Hybrid systems: Artificial Intelligence

NATIONAL CONFERENCE ON SOFTWARE AGENTS (TEHRAN, IRAN). 2015

1

Hybrid systems: Artificial Intelligence M. Ress Abstract The hybrid forecasting system that has been developed consists of a case-based reasoning system integrated with a radial basis function artificial neural network. The results obtained from experiments in which the system operated in real time in the oceanographic environment, are presented. An approach to hybrid artificial intelligence problem solving is presented in which the aim is to forecast, in real time, the physical parameter values of a complex and dynamic environment: the ocean. In situations in which the rules that determine a system are unknown or fuzzy the prediction of the parameter values that determine the characteristic behaviour of the system can be a problematic task. In such a situation it has been found that a hybrid artificial intelligence model can provide a more effective means of performing such predictions than either connectionist or symbolic techniques used separately.

I. I NTRODUCTION Forecasting the behaviour of a dynamic system is, in general, a difficult task, especially if the prediction needs to be achieved in real time. In such a situation one strategy is to create an adaptive system which possesses the flexibility to behave in different ways depending on the state of the environment. This paper presents the application of a novel hybrid artificial intelligence (AI) model to a real time forecasting problem. The approach which is discussed is capable of producing satisfactory results in situations in which neither artificial neural network nor statistical models have been sufficiently successful [1]. The oceans of the world form a highly dynamic system for which it is difficult to create mathematical models. Although some statistical models have been formulated to describe partial oceanographic water masses, there are, as yet, no accurate general models. It is well known that the behaviour and characteristics of the oceans change seasonally and spatially. However, current knowledge of the ocean structure is still too weak to create a comprehensive model. Ocean water masses are extremely heterogeneous; each water mass has certain properties that differentiate it from other water masses. Moreover, the convergence area between different water masses can be very noisy; for example, both the Arctic and Antarctic convergence zones are extremely heterogeneous and very variable [2]. An artificial intelligence approach to the problem of forecasting in the ocean environment offers potential advantages over alternative approaches, because it is able to deal with uncertain [3], incomplete and even inconsistent data. In these experiments it has been discovered that it is very difficult to train a neural network to forecast successfully over the whole time series, especially if the data relate to a dynamic system [4]. Statistical models such as Auto-Regressive Integrated Moving Averages (ARIMA) have been applied, but the results [5-9] so far obtained have indicated that neural networks have a greater facility for forecasting using this type of time series data. An attempt was also made to create an ANN able to train itself, in real time, with a number of consecutive values from a time series in order to forecast future parameter values [10-16]. The aim was to build a system having the ability to create a local model, thus capturing the characteristics of each particular section of the time series. Although this network produced better results that any of the other models, it was considered not to be sufficiently accurate. Following earlier experiments using case-based reasoning (CBR) as a problem solving strategy in other domains, a casebased approach to the forecasting problem was considered worthy of investigation [17-21]. Consequently a prototype system based on this approach was developed in the belief that a CBR mechanism might, as a data mining strategy, make better use of the vast database of oceanographic data held at PML [22]. The results of subsequent experiments employing a CBR approach indicated that case-based reasoning methods could facilitate the organisation of data, the recovery of relevant data necessary to make an accurate forecast and incremental system learning. The adaptation of the recovered data is a crucial factor in obtaining an accurate result [23]. With the aim of providing a more effective case adaptation procedure it was decided to investigate a hybrid systems approach in which a neural network would be employed as a means of case adaptation [24-48]. As a result, a Radial Basis Function (RBF) network has been found to be effective in the case adaptation stage of the CBR system. An important aim in the current work is to develop a universal forecasting mechanism, in the sense that it might operate effectively anywhere, at any point on the surface of the ocean, and at any time of the year without human intervention. The results obtained to date suggest that the approach to be described in this paper appears to fulfil this aim. The structure of the paper is as follows. First a brief overview of the basic concepts of case-based reasoning is given. Then the oceanographic problem domain is briefly outlined. Following this, a review of the application of CBR elsewhere as a strategy for forecasting is presented. The hybrid neural network enhanced CBR system is then explained, and, finally, an outline of some of the results obtained to date is presented. II. CBR S YSTEMS OVERVIEW Although knowledge-based systems (KBS) represent one of the commercial successes of the outcome of artificial intelligence research, developers of these systems have encountered several problems. Knowledge elicitation, a necessary process in the


2

development of rule-based systems, can be problematic [49-73]. The implementation of a KBS can also be complex, and, once implemented, may also be difficult to maintain. With the aim of overcoming these problems proposed a revolutionary approach, case-based reasoning, which is in fact a model of human reasoning [74-87]. The idea underlying CBR is that people frequently rely on previous problem-solving experiences when solving new problems [88-120]. This assertion may be verified in many day to day problem solving situations by simple observation or by psychological experimentation [121-145]. Since the ideas underlying case-based reasoning were first proposed, CBR systems have been found to be successful in a wide range of application areas [146]. A case-based reasoning system solves new problems by adapting solutions that were used to solve previous problems [147189]. The case base holds a number of cases, each of which represents a problem together with its corresponding solution [190-245]. Once a new problem arises, a possible solution to it is obtained by retrieving similar cases from the case base and studying their recorded solutions. A CBR system is dynamic in the sense that, in operation, cases representing new problems together with their solutions are added to the case base, redundant cases are eliminated and others are created by combining existing cases. A CBR system analyses a new problem situation, and by means of indexing algorithms, retrieves previously stored cases, together with their solution, by matching them against the new problem situation, then adapts them to provide a solution to the new problem by reusing knowledge stored in the form of cases in the case base [246-274]. All of these actions are self contained and may be represented by a cyclic sequence of processes, in which human interaction may possibly be needed. Case-base reasoning can be used by itself or as part of another intelligent or conventional computing system. Furthermore, case-based reasoning can be a particularly appropriate problem solving strategy when the knowledge required to formulate a rule-based model of the domain is difficult to obtain, or when the number or complexity of rules relating to the problem domain is too great for conventional knowledge acquisition methods [275-307]. The methodology of CBR has been successfully used in disciplines such as law, medicine and in various diagnostic situations where a rich history of cases may be accumulated. A typical CBR system is composed of four sequential steps. The purpose of the retrieval step is to search the case base and to select from it one or more previous cases that most closely match the new problem situation, together with their solutions. The selected cases are reused to generate a solution appropriate to the current problem situation. This solution is revised if necessary and finally the new case (i.e. the problem description together with the obtained solution) is stored in the case base. Cases may be deleted if they are found to produce inaccurate solutions, they may be merged together to create more generalised solutions, and they may be modified, over time, through the experience gained in producing improved solutions. If an attempt to solve a problem fails and it is possible to identify the reason for the failure, then this information should also be remembered in order to avoid the same mistake in the future. This corresponds to a common learning strategy employed in human problem solving [53-58]. Rather than creating general relationships between problem descriptors and conclusions as is the case with rule-based reasoning, or relying on general knowledge of the problem domain, CBR systems are able to utilise the specific knowledge of previously experienced concrete problem situations. A CBR system provides an incremental learning process because each time that a problem is solved a new experience is retained, thus making it available for future reuse. In the CBR cycle there is normally some human interaction. Whilst case retrieval and reuse may be automated, case revision and retention are often undertaken by human experts. This is a current weakness of CBR systems and one of their major challenges. In this paper a method of automating the process of case adaptation (revision) is presented for the solution of problems in which the cases are characterised predominantly by numerical information. A. CBR Systems for Forecasting Several researchers have used k-nearest-neighbour algorithms for time series predictions. Although a k-nearest-neighbour algorithm does not, in itself, constitute a CBR system, it may be regarded as a very basic and limited form of CBR operation in numerical domains. uses a relatively complex hybrid CBR-ANN system [59]. In contrast otherforecast a data set just by searching in a given sequence of data values for segments that closely match the pattern of the last n measurements and then by supposing that similar antecedent segments are likely to be followed by similar consequent segments. All the models presented use flat memories with simple data representation structures. Inspection reveals that the majority of the authors use k-nearest neighbour metrics in the reuse phase of the CBR cycle. The implementation of k-nearest neighbour methods requires a detailed analysis of the problem in order to define effective comparison rules. In most of the models, the cases are represented in the form of a vector of data values, which facilitates comparisons amongst them, especially if they hold quantitative or numerical values. K-nearest neighbour metrics are acceptable if the system is relatively stable and well understood. But if the system is dynamic and forecasts are required in real time, it may not be easily possible to update the limits, thresholds and variables governing the k-nearest neighbour metrics. The systems use similarity metrics or statistical models; in some systems the case adaptation is accomplished manually. If the problem is very complex, there may be no planned adaptation strategy, in which case the most similar case is used directly. In the current research it is believed that an adequate adaptation mechanism is the key to successful application of the CBR paradigm, as will be explained in subsequent sections of this paper.


3

In the majority of the systems surveyed case revision (if carried out at all) is performed by human expert. The forecasting error is used to improve the metrics employed in the previous stages of the CBR cycle. Such modification constitutes what might be considered to be the learning aspect of the CBR process. CBP: Comparison Based Prediction Model The CBP model can not be considered a true CBR system since it does not include any of the four stages that characterise the CBR cycle. It has been included in this classification because it is based on the same theoretical foundation as the CBR systems. CBP used a methodology based on analogical reasoning and the author recommends its application to planning problems. Designed to assist experts, it is claimed to increase the success in solving planning problems by up to 30%. III. T HE O CEANOGRAPHIC P ROBLEM A. The Ocean Environment Oceanography is a science that concentrates on the understanding of the physical principles that drive the oceans, and which uses the tools of mathematics and theoretical fluid dynamics to forecast their behaviour. Oceanic waters are divided into provinces (also called water masses) which are moderately homogenous. The boundaries between provinces or water masses are known as fronts. These areas are very dynamic and their properties depend on the nature of the two water masses converging to create each front. Some frontal areas, e.g. the Arctic and Antarctic convergence zones, are extremely heterogeneous and are very variable; therefore forecasting the temperature of the water in these areas can be difficult. The movement of water masses makes the oceans water temperature change in a complex manner in both spatial and temporal domains. The analysis and interpretation of large volumes of oceanographic data have traditionally been achieved through the use of statistical software tools. However the processing speed when using such tools is limited by the need for frequent user intervention. B. The Application of Artificial Intelligence Methods Knowledge-based systems have been developed to assist in weather prediction. However, apart from the work of, there appears to be little evidence of the application of knowledge based approach as a means of predicting the location and movements of large water masses. The motivation for that work was derived from the oceanographers’ need for a better understanding of the ocean environment, for which, because of the complexities of the ocean, it is very difficult to formulate adequate and complete mathematical models. The results, using rule-based methods, provided some insight into the nature and complexities of the ocean. Building on this experience, it was decided to carry out further investigations into the application of alternative artificial intelligence problem solving approaches. The focus in recent research has been to investigate ways to forecast the thermal structure of the water ahead of an ongoing vessel, in real time. Several researchers have been working in this field and have obtained partial success. Forecasting in such an environment is a difficult task due to the nature and behaviour of the ocean waters, which are in a continuous state of movement. The scales of physical motion of the oceans and the atmosphere range from being ocean-wide down to tiny eddies, which are present in the neighbourhood of fronts. In order to obtain acceptable predictions an autonomous universal methodology, capable of forecasting changes in the water temperature as an expert oceanographer might do, is desirable. In addition, the system should be able to analyse, in real time, the variation in the temperature of the water on a local basis, to analyse and select the most relevant local knowledge from the huge database of information available (in the form of satellite images and thermal data profiles) to produce a forecast, taking into account factors such as the season of the year and the geographical location of the vessel from which the forecast is made. IV. F ORECASTING S YSTEMS In the current work the aim is to investigate a hybrid case-based approach to forecasting as a methodology for predicting the values of physical parameters (in particular, sea temperature) at a given depth around a sea going vessel from data acquired in real time, and also from past records of sea temperature (and possibly other oceanographic parameters) surrounding the vessel at some point ahead which will be reached in the immediate future. This information may also then be used to provide a forewarning of an impending oceanographic front. The approach builds on the methods and expertise previously developed in the earlier research referred to above. The problem of forecasting, which is currently being addressed, may be simply stated as follows: Given: a sequence of data values (which may be obtained either in real-time, or from stored records) relating to some physical parameters, Predict: the value of that parameter at some future point(s) or time(s). The raw data (on sea temperature, salinity, density and other physical characteristics of the ocean) which are measured in real time by sensors located on the vessel consist of a number of discrete sampled values of a parameter in the form of a time series. These data values are supplemented by additional data derived from satellite images, which are received weekly. In the present work the parameter used is the temperature of the water mass at a fixed depth. Values are sampled along a single horizontal dimension, thus forming a set of data points.


4

This data must be pre-processed in order to eliminate noise, to enhance interesting features, to smooth stable areas and to transform the data set into a form which may be represented on an absolute scale. There are several techniques that can be applied to transform the original data set to reduce noise, sharpen data and aid in the detection of fronts. The approach adopted employs a Sobel Filter, the operation of which is based on the idea that local variations, corresponding to edge transitions, occur at a slower rate than those corresponding to noise. A. Hybrid CBR - Neural Network System In order to produce a forecast, in real-time, of ocean temperature a certain distance ahead of a vessel as it traverses the ocean, a problem case is generated every 2 km. A problem case consists of a sequence of the N sampled data values (after suitable filtering and pre-processing) immediately preceding the data value corresponding to the current position of the vessel. A value of 40 for N has been found empirically to produce satisfactory results. The problem case also includes various other numerical values, including the current geographical location of the vessel and the time and date when the case was recorded. The set of N data values forms an input vector, which is then used to produce a forecast of the ocean temperature, several km ahead. (Note that, in practice, it is the set of differences (Ti Ti-1, Ti Ti-2 etc.) between the temperature Ti at the current point and the temperature at successive earlier points which is used as the input vector. The forecasted values are created using a neural network enhanced case-base reasoning system. The CBR mechanism allows the experience recorded in previous forecasting situations to be reused. The role of the neural network lies in the case adaptation process. The four basic phases in the CBR cycle are shown as ellipses. Superimposed on the fundamental CBR cycle is a cycle of neural network operations during which the network parameters are retrieved from a neural network knowledge base, employed in case adaptation, and then are revised, with their updated values being stored back in the knowledge base. The full cycle of operations of the hybrid system is explained in the following section. The particular type of neural network of interest in the current research is the Radial Basis Function, in which the input layer is a receptor for the input data, whilst the hidden layer performs a non-linear transformation from the input space to the hidden layer space. The hidden neurons form a basis for the input vectors; the output neurons merely calculate a linear combination of the hidden neurons’ outputs. Activation is fed forward from the input layer to the hidden layer where a Basis Function is calculated. The weighted sum of the hidden neurons activations is calculated at the single output neuron. Radial Basis Functions (RBF) are better at interpolating that at extrapolating. Furthermore, RBFs are less sensitive to the order in which data is presented to them than is the case with other neural network models, such as Multi-Layer Perceptrons. Radial Basis Functions are of potential use in hybrid systems because of their fast learning capability. B. Hybrid System Operation The forecasting system uses data from two sources: (i) the real-time data are used to create a succession of problem cases, characterising the current forecasting situation; (ii) data derived from satellite images are stored in a database. The satellite image data values are used to generate cases, which are then stored in the case base and subsequently updated during the CBR operation. The cycle of forecasting operations (which is repeated every 2 km) proceeds as follows. First a new problem case is created from the pre-processed real-time data. A set of k cases, which most closely matches this current problem case, is then obtained from the case base during the CBR retrieve phase, using nearest neighbour matching. In the reuse phase, the values of the weights and centres of the neural network used in the previous forecast are retrieved from the neural network knowledge base. These network parameters together with the k closest matching cases are then used to create a forecast of the temperature a distance 5 km, say, ahead. At this point the parameters of the network are modified by taking into account the information contained in the retrieved cases. The effect of this is to allow the system to learn from all these k cases (rather than simply using the single adjudged closest matching case) in making a new forecast. During each forecasting cycle the RBF network is retrained, using the retrieved weights and centres, with the input vectors contained in the k matching cases applied as inputs to the network. This process adapts the network, by accommodating the retrieved cases, thus updating the values of the network parameters (empirically, a value for k of between 500 to 1000 has been found to be appropriate). The input vector from the problem case is then fed into the trained network to produce a proposed forecast. In the revise phase, the proposed forecast is modified by taking into account the accuracy of the previous forecasts, which were reused in obtaining the new forecast. Each case has associated with it a measure of the average error over the previous forecasts for which that particular case was used to train the neural network. Error limits are calculated by averaging the average error of the k cases used to train the ANN in producing the current forecast. The revised forecast is then expressed, using the error limits, as an interval between upper and lower limits rather than as a single value. The revised forecast is then retained in a temporary store the forecasts database. When the vessel has travelled a further 5 km the actual value of the water temperature at that point is measured. The forecasted value for the temperature at this point can then be evaluated, by comparison of the actual and forecasted values, and the error obtained. A new case, corresponding to this forecasting operation, is then entered in the case base. Knowledge of the forecasting error is also, at this point, used to


5

update the average error of all the k cases that were reused to obtain that forecast. 4.3 Radial Basis Function Operation The RBF network uses nine input neurons, between twenty and thirty five neurons in the hidden layer and a single neuron in the output layer. Input vectors (explained earlier) form the input to the network; the output of the network is the difference between the temperature at the present point and the temperature a fixed distance ahead. Initially, twenty vectors are randomly chosen from the first training data set and used as centres in the middle layer of the RBF network. All the centres are associated with a Gaussian function, the width of which, for all the functions, is set to the mean value of the Euclidean distance between the two centres that are separated the most from each other. Training of the network is done by presenting pairs of corresponding input and desired output vectors. After an input vector activates every Gaussian unit the activations are propagated forward through the weighted connections to the output units which sum all incoming signals. The comparison of actual and desired output values enables the mean square error (the quantity to be minimised) to be calculated. The closest centre to each particular input vector is moved toward the input vector by a percentage of the present distance between them. By using this technique the centres are positioned close to the highest densities of the input vector data set. The aim of this adaptation is to force the centres to be as close as possible to as many vectors from the input space as possible. The value of is initialised to a value of 20 each time that the network is retrained, and its value is linearly decreased with the number of iterations until its value becomes zero; then the network is trained for a number of iterations (between 10 and 30 iterations for the whole training data set, depending on the time left for the training) in order to obtain the best possible weights for the final value of the centres. A new centre is inserted into the network when the average error in the training data set does not fall by more than 10% after 10 iterations (using the whole training set). In order to determine the most distant centre C, the Euclidean distance between each centre and each input vector is calculated and the centre whose distance from the input data vectors is largest is chosen. A new centre is inserted between C and the centre closest to it. Centres are also eliminated when they do not contribute significantly to the output of the neural network. Thus, a neuron is eliminated if the absolute value of the weight associated with that neuron is smaller than twenty per cent of the average value of the absolute value of the five smallest weights. The number of neurons in the middle layer is maintained above 20. V. R ESULTS The hybrid forecasting system has been tested in the Atlantic Ocean in September on a research cruise going from the UK to the Falkland Islands. The cruise crossed several water masses and oceanographic fronts (areas where two water masses with different characteristics converge). Although, the system was monitored and improved from a computing point of view, over a month, the forecasting method explained in previous sections remained unchanged. The prototype system used in this experiment was set up to forecast the temperature 5 km ahead. The strategy adopted was to create an accurate successful method which was able to forecast a short distance ahead, and then to extend it so as to produce forecasts a further distance ahead. The average error in the forecast was found to be 0.02 C . Only 4.5% of the forecasts have an error higher than 0.5 C, 8.3% higher than 0.04 C, 32% higher than 0.02 C. These data indicate that the hybrid system is able to produce a forecast with an average error of 0.02 C and with a probability of 0.96 that the error in the forecast is smaller than 0.05 C Although the experiment was carried out using a limited data set (11000 km between the latitudes 50 North and 50 South), eleven water masses with different characteristics were crossed, six fronts were traversed. The Falkland Front (km 10000) in particular is one of the most chaotic oceanographic areas in the world. It is believed that these error value results are significant enough to be extrapolated over the whole Atlantic Ocean. The use of error limits can substantially improve the accuracy of the forecast. The error limit values are determined and modified dynamically from information relating to the past forecasting performance of the system and which is contained in the stored cases. These error limits indicate the range of forecast values than may be expected to be produced through the adaptation of particular stored cases. The value of the error limits used during the experiment and the forecast error outside the error limits. (If, using error limits, the forecasted temperature value at a particular location is, say, 6 +- 0.5 C - i.e. from 5.5 C to 6.5 C - then, if the actual temperature value is found to be 6.8 C, the value of the forecast error outside the error limits will be 6.8 - 6.5, i.e. 0.3 C.) Although 45.5% of the predictions were outside the limits of the error band, only 3.4% of the predictions were more than 0.005 C outside the error limits. The average error of the predictions using error limits is 0.00099 C. This means that the error limits adapt themselves to the pattern of temperatures in the different water masses. Both the error and the error limits are, on average, higher in frontal water masses than in homogeneous water masses. This was to be expected due to the higher dynamic nature and heterogeneity of such areas. A similar experiment was carried out using the data recorded by the vessel during the cruise, but this time using cases obtained from satellite images recorded more than one week previously. It also shows that when using satellite pictures collected exactly one year back the error in the forecast may be similar or smaller than the error obtained using pictures that are three or more weeks old. This is the reason why data up to one year old is kept in the database and in the case base; if for technical reasons (e.g. clouds covering a certain area or problems with the data telecommunications) recent satellite images can not be obtained,


6

data recorded one year back can be used by the system and may in fact produce better results that data recorded three or four weeks previously. This is due to the annual cycle of most of the water masses. However, such results can not be guaranteed as there are also other factors that determine the pattern of ocean temperature variations. Further experiments have been carried out to compare the performance of the CBR-ANN hybrid forecasting system with several other forecasting approaches. These include standard statistical forecasting algorithms and the application of several neural network methods. Using the hybrid system the forecasting error is less than 20% of the corresponding value produced by any of the other forecasting methods. VI. C ONCLUSIONS This paper has presented a novel hybrid problem solving method that combines a case-based reasoning system integrated with an artificial neural network, which is able to identify trends present in a large data set in order to create forecasts in real time. As such, it could be regarded as an application of data mining in a real time situation. The forecasting task which is addressed in this work is difficult for two reasons: the unpredictable and complexity of the media in which the forecast must to be done, and the fact that the forecast must be done in real time. Essentially, the cases stored in the case base are created through the selection of data values which correspond to the current location and track of the ship. The time series data values obtained in real time are then matched by the CBR mechanism against the data patterns of the stored cases in order to produce the required forecast. The data are derived from the extensive database of earlier historical values, or from satellite images. The evaluation of the forecasts produced and the consequent modifications of the stored cases enable the system to learn and to improve its performance over time. The hybrid forecasting system is able to produce a forecast with an acceptable degree of accuracy and within the time constraints imposed by the real time nature of the problem. Although the accuracy of the forecast depends, to a great extent, on the quality of the cases and on the actual date when the data from which the cases were created was collected, it has been demonstrated that good quality forecasts may be obtained even with data collected one year before the forecast was made. The method combines the ability of case-based reasoning to index, organise and retrieve relevant data with the generalisation, learning and adaptation capabilities of the radial basis function neural network. The resulting hybrid system thus combines complementary properties of both connectionist and symbolic AI methods. The neural network plays an important role in the system; it adapts the cases selected during the case-based operations, combines aspects of the knowledge contained in several cases and supports the generation of the prediction. The Radial Basis Function network adapts its structure, in an unsupervised way, to the characteristics of the environment in which the system is operating and acts as a function that facilitates system learning by extracting the relevant characteristics of a number of closely matching cases and combining them in the form of a representative case. The limitations of this method of forecasting in its present form are as follows. Forecasts can only be produced while the vessel is proceeding in a straight line. The present system is not able to forecast while the vessel is changing its direction (or has changed it in the last 40 km). However, it would be possible to adapt the way in which the data is extracted from the database during the case creation process, to enable it to overcome this limitation. The present system operates satisfactorily only if there are no discontinuities in the data greater than 2 km in length. However, after a discontinuity the forecast can be resumed by interpolating the missing data with data from both ends of the case vector. This strategy has been found to be successful if the discontinuity is no longer than 5 km. The system can not function in a particular area if there are no stored cases from that area. In this situation, the only solution is to use a back-up mechanism to prime the system; for this, the experimental results obtained in comparing neural network methods suggest that a Finite Impulse Response network may be the most appropriate method to use. Once the system is in operation and is producing forecasts, a succession of cases will be generated, thus enabling the hybrid forecasting mechanism to function autonomously. In conclusion, the hybrid problem solving approach provides an effective strategy for forecasting in an environment in which the raw data is derived from three distinct sources: a large database of historical data, satellite image data, and time series data obtained in real-time. R EFERENCES [1] A. Abraham, E. Corchado, and J. M. Corchado, “Hybrid learning machines,” 2009. [2] F. Fdez-Riverola, E. L. Iglesias, F. D´ıaz, J. R. Méndez, and J. M. Corchado, “Applying lazy learning algorithms to tackle concept drift in spam filtering,” Expert Systems with Applications, vol. 33, no. 1, pp. 36–48, 2007. [3] M. G. Bedia and J. Corchado, “A planning strategy based on variational calculus for deliberative agents.” 2002. [4] F. Fdez-Riverola and J. M. Corchado, “Fsfrt: forecasting system for red tides,” Applied Intelligence, vol. 21, no. 3, pp. 251–264, 2004. [5] F. Fdez-Riverola, E. L. Iglesias, F. Diaz, J. R. Méndez, and J. M. Corchado, “Spamhunting: An instance-based reasoning system for spam labelling and filtering,” Decision Support Systems, vol. 43, no. 3, pp. 722–736, 2007. [6] F. D´ıaz, F. Fdez-Riverola, and J. M. Corchado, “gene-cbr: A case-based reasonig tool for cancer diagnosis using microarray data sets,” Computational Intelligence, vol. 22, no. 3-4, pp. 254–268, 2006. [7] C. Fyfe, E. Corchado Rodr´ıguez, M. González Bedia, and J. M. Corchado Rodr´ıguez, “Analytical model for constructing deliberative agents.” 2002. [8] C. Fyfe and J. M. Corchado, “Automating the construction of cbr systems using kernel methods,” International Journal of Intelligent Systems, vol. 16, no. 4, pp. 571–586, 2001.


7

[9] J. M. Corchado, E. S. Corchado, J. Aiken, C. Fyfe, F. Fernandez, and M. Gonzalez, “Maximum likelihood hebbian learning based retrieval method for cbr systems,” in International Conference on Case-Based Reasoning. Springer, Berlin, Heidelberg, 2003, pp. 107–121. [10] J. M. Corchado and C. Fyfe, “Unsupervised neural method for temperature forecasting,” Artificial Intelligence in Engineering, vol. 13, no. 4, pp. 351–357, 1999. [11] F. Fdez-Riverola and J. M. Corchado, “Cbr based system for forecasting red tides,” Knowledge-Based Systems, vol. 16, no. 5-6, pp. 321–328, 2003. [12] J. M. Corchado and J. Aiken, “Hybrid artificial intelligence methods in oceanographic forecast models,” IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), vol. 32, no. 4, pp. 307–313, 2002. [13] J. M. Corchado, J. Aiken, and N. Rees, Artificial intelligence models for oceanographic forecasting. Plymouth Marine Laboratory, 2001. [14] R. Laza, R. Pavón, and J. M. Corchado, “A reasoning model for cbr bdi agents using an adaptable fuzzy inference system,” in Current topics in artificial intelligence. Springer, Berlin, Heidelberg, 2004, pp. 96–106. [15] F. Fernandez-Riverola, F. Diaz, and J. M. Corchado, “Reducing the memory size of a fuzzy case-based reasoning system applying rough set techniques,” IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), vol. 37, no. 1, pp. 138–146, 2007. [16] J. R. Méndez, F. Fdez-Riverola, E. L. Iglesias, F. D´ıaz, and J. M. Corchado, “Tracking concept drift at feature selection stage in spamhunting: an anti-spam instance-based reasoning system,” in European conference on case-based reasoning. Springer, Berlin, Heidelberg, 2006, pp. 504–518. [17] B. Baruque, E. Corchado, A. Mata, and J. M. Corchado, “A forecasting solution to the oil spill problem based on a hybrid intelligent system,” Information Sciences, vol. 180, no. 10, pp. 2029–2043, 2010. [18] B. Lees and J. Corchado, “Case based reasoning in a hybrid agent-oriented system,” in 5th German Workshop on Case-Based Reasoning, 1997, pp. 139–144. [19] J. M. Corchado, M. L. Borrajo, M. A. Pellicer, and J. C. Yán˜ ez, “Neuro-symbolic system for business internal control,” in Industrial Conference on Data Mining. Springer, Berlin, Heidelberg, 2004, pp. 1–10. [20] J. R. Mendez, F. Fdez-Riverola, F. Diaz, E. L. Iglesias, and J. M. Corchado, “A comparative performance study of feature selection methods for the anti-spam filtering domain,” in Industrial Conference on Data Mining. Springer, Berlin, Heidelberg, 2006, pp. 106–120. [21] E. Corchado, J. M. Corchado, L. Sáiz, and A. Lara, “Constructing a global and integral model of business management using a cbr system,” in International Conference on Cooperative Design, Visualization and Engineering. Springer, Berlin, Heidelberg, 2004, pp. 141–147. [22] J. M. Corchado, B. Lees, and J. Aiken, “Hybrid instance-based system for predicting ocean temperatures,” International Journal of Computational Intelligence and Applications, vol. 1, no. 01, pp. 35–52, 2001. [23] J. M. Corchado, J. Aiken, E. S. Corchado, N. Lefevre, and T. Smyth, “Quantifying the oceans co2 budget with a cohel-ibr system,” in European Conference on Case-Based Reasoning. Springer, Berlin, Heidelberg, 2004, pp. 533–546. [24] D. Glez-Peña, F. D´ıaz, J. M. Hernández, J. M. Corchado, and F. Fdez-Riverola, “genecbr: a translational tool for multiple-microarray analysis and integrative information retrieval for aiding diagnosis in cancer research,” BMC bioinformatics, vol. 10, no. 1, p. 187, 2009. [25] J. M. Corchado, B. Lees, and N. Rees, “A multi-agent system test bed for evaluating autonomous agents,” in Proceedings of the first international conference on Autonomous agents. ACM, 1997, pp. 386–393. [26] J. M. Corchado, R. Laza, L. Borrajo, J. Yañes, and M. Valiño, “Increasing the autonomy of deliberative agents with a case-based reasoning system,” International Journal of Computational Intelligence and Applications, vol. 3, no. 01, pp. 101–118, 2003. [27] J. M. Corchado, J. M. Corchado, and J. M. M. López, “Introducción a la teor´ıa de agentes y sistemas multiagente,” 2002. [28] A. Mata and J. M. Corchado, “Forecasting the probability of finding oil slicks using a cbr system,” Expert Systems with Applications, vol. 36, no. 4, pp. 8239–8246, 2009. [29] J. R. Méndez, D. Glez-Peña, F. Fdez-Riverola, F. D´ıaz, and J. M. Corchado, “Managing irrelevant knowledge in cbr models for unsolicited e-mail classification,” Expert Systems with Applications, vol. 36, no. 2, pp. 1601–1614, 2009. [30] B. Lees, B. Kumar, A. Mathew, J. Corchado, B. Sinha, and R. Pedreschi, “A hybrid case-based neural network approach to scientific and engineering data analysis,” in Applications and Innovations in Expert Systems VI. Springer, London, 1997, pp. 245–260. [31] J. M. Corchado and R. Laza, “Construction of bdi agents from cbr systems,” in Proceedings of the 1st German Workshop on on Experience Management: Sharing Experiences about the Sharing of Experience. GI, 2002, pp. 47–60. [32] C. Fyfe and J. Corchado, “A comparison of kernel methods for instantiating case based reasoning systems,” Advanced Engineering Informatics, vol. 16, no. 3, pp. 165–178, 2002. [33] F. Fdez-Riverola, F. D´ıaz, M. L. Borrajo, J. C. Yán˜ ez, and J. M. Corchado, “Improving gene selection in microarray data analysis using fuzzy patterns inside a cbr system,” in International Conference on Case-Based Reasoning. Springer, Berlin, Heidelberg, 2005, pp. 191–205. [34] J. Corchado, C. Fyfe, and B. Lees, “Unsupervised learning for financial forecasting,” in Computational Intelligence for Financial Engineering (CIFEr), 1998. Proceedings of the IEEE/IAFE/INFORMS 1998 Conference on. IEEE, 1998, pp. 259–263. [35] J. Pavón, J. M. Corchado, J. J. Gómez-Sanz, and L. F. C. Ossa, “Mobile tourist guide services with software agents,” in International Workshop on Mobile Agents for Telecommunication Applications. Springer, Berlin, Heidelberg, 2004, pp. 322–330. [36] F. Fdez-Riverola, J. M. Corchado, and J. M. Torres, “An automated hybrid cbr system for forecasting,” in European Conference on Case-Based Reasoning. Springer, Berlin, Heidelberg, 2002, pp. 519–533. [37] F. Fdez Riverola and J. M. Corchado, “Sistemas h´ıbridos neuro-simbólicos: una revisión,” Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial, vol. 4, no. 11, 2000. [38] J. R. Méndez, C. González, D. Glez-Peña, F. Fdez-Riverola, F. D´ıaz, and J. M. Corchado, “Assessing classification accuracy in the revision stage of a cbr spam filtering system,” in International Conference on Case-Based Reasoning. Springer, Berlin, Heidelberg, 2007, pp. 374–388. [39] J. Corchado, B. Lees, and C. Fyfe, “Project monitoring intelligent agent system,” 1997. [40] F. Fdez-Riverola and J. M. Corchado, “Fsfrt: Forecasting system for red tides. a hybrid autonomous ai model,” Applied Artificial Intelligence, vol. 17, no. 10, pp. 955–982, 2003. [41] J. R. Méndez, E. L. Iglesias, F. Fdez-Riverola, F. D´ıaz, and J. M. Corchado, “Tokenising, stemming and stopword removal on anti-spam filtering domain,” in Conference of the Spanish Association for Artificial Intelligence. Springer, Berlin, Heidelberg, 2005, pp. 449–458. [42] F. D´ıaz, F. Fdez-Riverola, D. Glez-Peña, and J. M. Corchado, “Using fuzzy patterns for gene selection and data reduction on microarray data,” in International Conference on Intelligent Data Engineering and Automated Learning. Springer, Berlin, Heidelberg, 2006, pp. 1087–1094. [43] ——, “Applying gcs networks to fuzzy discretized microarray data for tumour diagnosis,” in International Conference on Intelligent Data Engineering and Automated Learning. Springer, Berlin, Heidelberg, 2006, pp. 1095–1102. [44] B. Baruque, E. Corchado, A. Mata, and J. M. Corchado, “Ensemble methods for boosting visualization models,” in International Work-Conference on Artificial Neural Networks. Springer, Berlin, Heidelberg, 2009, pp. 165–173. [45] F. Garc´ıa, J. Corchado, and M. Laguna, “Cbr applied to development with reuse based on mecanos,” in Proceedings of the 13th International Conference on Software Engineering and Knowledge Engineering, Buenos Aires, Argentina, 2001. [46] F. D´ıaz and J. Corchado, “Rough sets for detecting dependence relationships and bayesian networks construction,” in Proc. of IPMU, 2000. [47] F. Fernández-Riverola and J. M. Corchado, “Employing tsk fuzzy models to automate the revision stage of a cbr system,” in Current Topics in Artificial Intelligence. Springer, Berlin, Heidelberg, 2004, pp. 302–311. [48] M. L. Borrajo, J. M. Corchado, J. C. Yán˜ ez, F. Fdez-Riverola, and F. D´ıaz, “Autonomous internal control system for small to medium firms,” in International Conference on Case-Based Reasoning. Springer, Berlin, Heidelberg, 2005, pp. 106–121. [49] J. M. Corchado, R. Laza, L. Borrajo, J. Yañez, A. De Luis, and M. Gonzalez-Bedia, “Agent-based web engineering,” in International Conference on Web Engineering. Springer, Berlin, Heidelberg, 2003, pp. 17–25.


8

[50] J. M. Corchado, J. Aiken, E. S. Corchado, and F. Fdez-Riverola, “Evaluating the air–sea interactions and fluxes using an instance-based reasoning system,” AI Communications, vol. 18, no. 4, pp. 247–256, 2005. [51] J. M. Corchado, “Agencia: Una puerta hacia la convergencia de la inteligencia artificial,” Universidad de Salamanca, 1999. [52] ——, “Modelos y arquitecturas de agente,” Univerisdad de Salamanca, España, 1999. [53] F. Fdez-Riverola, F. D´ıaz, and J. M. Corchado, “Applying rough sets reduction techniques to the construction of a fuzzy rule base for case based reasoning,” in Ibero-American Conference on Artificial Intelligence. Springer, Berlin, Heidelberg, 2004, pp. 83–92. [54] F. Fdez-Riverola and J. M. Corchado, “Forecasting red tides using an hybrid neuro-symbolic system,” AI Communications, vol. 16, no. 4, pp. 221–233, 2003. [55] R. Laza, A. Gómez, R. Pavón, and J. M. Corchado, “A case-based reasoning approach to the implementation of bdi agents.” in ECCBR Workshops, 2002, pp. 27–30. [56] M. Glez-Bedia and J. Corchado, “Variational calculus bases planning for deliberative agents,” in I Workshop on Planning, Scheduling and Temporal Reasoning. Sevilla, Spain, vol. 12, 2002. [57] F. Sandoval, A. Prieto, and J. M. Corchado, “Bio-inspired systems: Computational and ambient intelligence,” 2009. [58] F. D´ıaz and J. Corchado, “The selection of relevant features and rough sets,” in Conferencia de la Asociación Española de Inteligencia Artificial. Gijón, 2001, pp. 14–16. [59] E. S. Corchado, J. M. Corchado, L. Sáiz, and A. Lara, “A beta-cooperative cbr system for constructing a business management model,” in Industrial Conference on Data Mining. Springer, Berlin, Heidelberg, 2004, pp. 42–49. [60] D. Glez-Pena, F. D´ıaz, F. Fdez-Riverola, J. R. Méndez, and J. M. Corchado, “Fuzzy patterns and gcs networks to clustering gene expression data,” in Fuzzy Systems in Bioinformatics and Computational Biology. Springer, Berlin, Heidelberg, 2009, pp. 103–125. [61] J. Corchado, J. Torres, and M. Sacai Caudrado, “Forecasting red tides caused by pseudo-nitzschia spp. using an artificial intelligence model,” Harmful algae, pp. 531–533, 2002. [62] E. Corchado, A. Mata, B. Baruque, J. M. Corchado, and B. Perez-Lancho, “A topology-preserving system for environmental models forecasting,” International Journal of Computer Mathematics, vol. 88, no. 9, pp. 1979–1989, 2011. [63] B. Lees and J. Corchado, “Neural network support in a hybrid case-based forecasting system,” Case-Based Reasoning Integrations, pp. 85–90, 1998. [64] J. M. Corchado, E. S. Corchado, and M. A. Pellicer, “Design of cooperative agents for mobile devices,” in International Conference on Cooperative Design, Visualization and Engineering. Springer, Berlin, Heidelberg, 2004, pp. 205–212. [65] J. M. Corchado, A. Mata, and S. Rodriguez, “Osm: A multi-agent system for modeling and monitoring the evolution of oil slicks in open oceans,” in Advanced Agent-Based Environmental Management Systems. Birkhäuser Basel, 2009, pp. 91–117. [66] M. Bedia and J. Corchado, “Constructing autonomous distributed systems using cbr-bdi agents,” Innovations in Knowledge Engineering, 2013. [67] R. Laza, F. Fernández, and J. Corchado, “Sistemas de razonamiento basado en casos para el soporte a la toma de decisiones,” 1999. [68] J. M. Corchado, E. S. Corchado, and J. Aiken, “An ibr system to quantify the oceans carbon dioxide budget,” in Industrial Conference on Data Mining. Springer, Berlin, Heidelberg, 2004, pp. 33–41. [69] M. L. Borrajo, R. Laza, and J. M. Corchado, “A complex case-based advisor,” Applied Artificial Intelligence, vol. 22, no. 5, pp. 377–406, 2008. [70] D. MacDonald, J. Koetsier, E. Corchado, C. Fyfe, and J. Corchado, “A kernel method for classification,” in Mexican International Conference on Artificial Intelligence. Springer, Berlin, Heidelberg, 2004, pp. 823–832. [71] E. Corchado, B. Baruque, A. Mata, and J. M. Corchado, “A wevos-cbr approach to oil spill problem,” in International Workshop on Hybrid Artificial Intelligence Systems. Springer, Berlin, Heidelberg, 2008, pp. 378–384. [72] O. H. Franco, L. F. Castillo, J. M. Corchado, and C. A. Lopez, “Multiagent system for software monitoring and users activities in a network equipment,” Scientia et technica, vol. 1, no. 34, 2007. [73] J. R. Méndez, F. Fdez-Riverola, D. Glez-Peña, F. D´ıaz, and J. M. Corchado, “Relaxing feature selection in spam filtering by using case-based reasoning systems,” in Portuguese Conference on Artificial Intelligence. Springer, Berlin, Heidelberg, 2007, pp. 53–62. [74] M. Valiño and J. Corchado, “Voip: the convergence of networks,” 1999. [75] V. F. López, R. Aguilar, L. Alonso, M. N. Moreno, and J. M. Corchado, “Grammatical inference with bioinformatics criteria,” Neurocomputing, vol. 75, no. 1, pp. 88–97, 2012. [76] F. Fdez-Riverola, F. D´ıaz, and J. M. Corchado, “Using rough sets theory and minimum description length principle to improve a β-tsk fuzzy revision method for cbr systems,” in Brazilian Symposium on Artificial Intelligence. Springer, Berlin, Heidelberg, 2004, pp. 424–433. [77] J. Corchado and B. Lees, “Evaluation of symbolic and connectionist approaches in a multi-agent system,” 1996. [78] J. PAVON and J. M. CORCHADO, “Agents for the web,” International journal of Web engineering and technology, vol. 1, no. 4, pp. 393–396, 2004. [79] M. G. Bedia, J. M. Corchado, and L. F. Castillo, “Bio-inspired memory generation by recurrent neural networks,” in International Work-Conference on Artificial Neural Networks. Springer, Berlin, Heidelberg, 2007, pp. 55–62. [80] F. Fdez-Riverola and J. M. Corchado, “A hybrid cbr model for forecasting in complex domains,” in Ibero-American Conference on Artificial Intelligence. Springer, Berlin, Heidelberg, 2002, pp. 101–110. [81] J. Koetsier, E. Corchado, D. MacDonald, J. Corchado, and C. Fyfe, “Kernel maximum likelihood hebbian learning,” in International Conference on Computational Science. Springer, Berlin, Heidelberg, 2004, pp. 650–653. [82] B. Lees and J. Corchado, “The application of hybrid artificial intelligence systems for forecasting,” in AIP Conference Proceedings, vol. 465, no. 1. AIP, 1999, pp. 259–267. [83] F. D´ıaz and J. Corchado, “Rough sets para la generación de redes de bayes,” Universidad de Vigo, 1999. [84] F. Fdez-Riverola and J. Corchado, “Fsfrt: Un sistema para la predicción de mareas rojas en las costas gallegas,” 2004. [85] F. D´ıaz, L. Borrajo, F. F. Riverola, A. Usero, J. Corchado, and C. U. A. Lagoas, “Negative feedback network for financial prediction,” 2004. [86] S. Rodr´ıguez, V. Julián, A. L. Sánchez, C. Carrascosa, V. F. López, J. M. Corchado, and E. Corchado, “Trends on the development of adaptive virtual organizations,” in Distributed Computing and Artificial Intelligence. Springer, Berlin, Heidelberg, 2010, pp. 113–121. [87] J. M. Corchado and J. F. Paz, Practical Applications of Computational Biology and Bioinformatics: 2nd International Workshop (IWPACBB 2008). Springer, 2008. [88] E. Corchado, J. Corchado, and C. Fyfe, “Kernel-sim preprocessing in a variational calculus based planning strategy,” 2001. [89] F. Fdez-Riverola, J. M. Corchado, and J. M. Torres, “Neuro-symbolic system for forecasting red tides,” in Irish Conference on Artificial Intelligence and Cognitive Science. Springer, Berlin, Heidelberg, 2002, pp. 45–52. [90] F. Fernandez-Riverola and J. M. Corchado, “Selected papers from the 10th conference of the spanish association for artificial intelligence (caepia03)employing tsk fuzzy models to automate the revision stage of a cbr system,” Lecture Notes in Computer Science, vol. 3040, pp. 302–311, 2004. [91] M. P. Rocha, F. F. Riverola, H. Shatkay, and J. M. C. Rodr´ıguez, “Advances in bioinformatics: 4th international workshop on practical applications of computational biology and bioinformatics 2010 (iwpacbb 2010),” 2010. [92] A. Abraham, D. Dumitrescu, Y. Chen, J. M. Corchado, Y. Dote, B. De Baets, O. François, P. Frisco, R. Ghosh, C. Groúan et al., “Nca 2006workshop on natural computing and applications,” 2006. [93] R. Cano and J. M. Corchado, “Painalli: Plataforma multiagente para la gestión inteligente de documentos y mensajes organizacionales,” 2007. [94] C. I. Pinzón and J. M. Corchado, “Arquitectura de un sistema multiagente para la clasificación de consultas con inyección sql,” 2010. [95] M. Borrajo, E. Corchado, M. A. Pellicer, and J. M. Corchado, “Cbr based engine for business internal control,” in Soft Computing Applications in Business. Springer, Berlin, Heidelberg, 2008, pp. 243–260. [96] J. M. Corchado and P. Cuesta, “Improving the interoperability in multi-agent systems,” in ECOOP Workshops, 1999, pp. 14–15.


9

[97] V. F. López, R. Aguilar, L. Alonso, M. N. Moreno, and J. M. Corchado, “Data mining for grammatical inference with bioinformatics criteria,” in International Conference on Hybrid Artificial Intelligence Systems. Springer, Berlin, Heidelberg, 2010, pp. 53–60. [98] J. C. Rodr´ıguez, R. L. Fidalgo, L. B. Diz, and M. V. Garcia, “An autonomous agent based engineering sales support system.” 2001. [99] L. Castillo, M. Bedia, and J. Corchado, “Librer´ıa para el intercambio de mensajes fip-acl entre dispositivos móviles a través de bluetooth,” 2005. [100] J. Vázquez Abad, F. Fdez-Riverola, and J. Corchado, “Forecasting economic cycles with conectionist models,” in I international meeting on economic cycles, 2000. [101] A. Losada, F. Fdez-Riverola, and J. Corchado, “Monitorización remota mediante vhf de una red de procesos industriales que utilizan plcs,” 2002. [102] J. M. Corchado and J. Pavón, “Development and application of cbr-bdi agents,” 2004. [103] R. Laza, R. Pavon, and J. M. Corchado, “Selected papers from the 10th conference of the spanish association for artificial intelligence (caepia03)-a reasoning model for cbr-bdi agents using an adaptable fuzzy inference system,” Lecture Notes in Computer Science, vol. 3040, pp. 96–106, 2004. [104] F. Dıaz and J. Corchado, “Selecting relevant features using rough sets and the mdl principle,” 2002. [105] M. G. Bedia and J. M. Corchado, “Cbr-bdi,” Acta Nova, vol. 2, no. 3, p. 362, 2003. [106] F. Fdez-Riverola, F. Diaz, and J. M. Corchado, “Evolutionary computation, artificial life, and hybrid systems-using rough sets theory and minimum description length principle to improve a, b-tsk fuzzy revision method for cbr systems,” Lecture Notes in Computer Science, vol. 3171, pp. 424–433, 2004. [107] M. L. Borrajo, J. M. Corchado, E. S. Corchado, and M. A. Pellicer, “Neural business control system,” in Industrial Conference on Data Mining. Springer, Berlin, Heidelberg, 2007, pp. 242–254. [108] J. M. Corchado Rodr´ıguez, J. Aiken, and N. Rees, “Neuro-symbolic reasoning system for modeling complex behaviours.” 2003. [109] D. d. I. y Automdtica, “An automated hybrid reasoning system for forecasting,” Soft Computing Systems: Design, Management and Applications, vol. 87, p. 32, 2002. [110] E. S. Corchado, M. A. Pellicer, and J. M. Corchado, “A guiding agent: Smart dynamic technology for solving distributed problems,” 2006. [111] P. Carrión, R. Laza, E. González-Rufino, and J. M. Corchado, “Knowledge management with an agent network.” in LANOMS, 1999. [112] J. M. Corchado, M. L. Borrajo, and J. C. Yán˜ ez, “Autonomous internal control system for small to medium firms,” 2005. [113] J. Corchado, E. Corchado, M. Glez-Bedia, A. de Luis, and L. Castillo, “Constructing deliberative agents using k-sim case-based reasoning systems,” Agentes Inteligentes en el Tercer Milenio, 2003. [114] J. M. Corchado, “Ingenier´ıa de software orientada a agentes,” 2008. [115] A. Abraham, E. Corchado, R. Alhaj, and V. Snasel, “Computational aspects of social networks (cason),” Salamanca, vol. 19, p. 21, 2011. [116] A. Mata, B. Pérez, and J. M. Corchado, “Forest fires prediction by an organization based system,” in Advances in Practical Applications of Agents and Multiagent Systems. Springer, Berlin, Heidelberg, 2010, pp. 135–144. ´ [117] L. GONZALEZ, J. TORRES, J. CORCHADO, A. TURIEL, and E. GARCIA-LADONA, “Detection, monitoring and forecasting of hydrocarbons spills in the ocean using remote sensing and artificial intelligence techniques,” 2003. [118] M. G. Bedia, J. M. Corchado, and L. F. Castillo, “Bio-inspired dynamical tools for analyzing cognition,” in Encyclopedia of Artificial Intelligence. IGI Global, 2009, pp. 256–261. [119] D. Borrajo, L. Castillo, and J. M. Corchado, “Current topics in artificial intelligence: 12th conference of the spanish association for artificial intelligence, caepia 2007, salamanca, spain, november 12-16, 2007, selected papers,” 2007. [120] J. Corchado, N. Rees, B. Lees, and J. Aiken, “Data mining using example-based methods in oceanographic forecast models,” 1998. [121] J. M. Corchado Rodr´ıguez, S. Rodr´ıguez González, J. Llinas, and J. M. Molina, “International symposium on distributed computing and artificial intelligence 2008 (dcai 2008),” 2009. [122] J. M. Corchado, “Adaptive hybrid system architecture for forecasting,” in Proceedings of the fourteenth national conference on artificial intelligence and ninth conference on Innovative applications of artificial intelligence. AAAI Press, 1997, pp. 808–808. [123] A. Mata, M. D. Muñoz, E. Corchado, and J. M. Corchado, “Isotropic image analysis for improving cbr forecasting,” Journal of Mathematical Imaging and Vision, vol. 42, no. 2-3, pp. 212–224, 2012. [124] J. R. Mendez Reboredo, F. Fernandez-Riverola, E. L. Iglesias, F. Diaz Gomez, and J. M. Corchado, “Application papers-tracking concept drift at feature selection stage in spamhunting: An anti-spam instance-based reasoning system,” Lecture Notes in Computer Science, vol. 4106, pp. 504–518, 2006. [125] A. Tawil, W. Behrendt, R. Cunningham, A. Diacakis, J. Pitt, J. Corchado, B. Lees, C. Fyfe, T. Rose, and P. Wyard, “Iee colloquium intelligent world wide web agents,” 1997. [126] J. R. Mendez, E. L. Iglesias, F. Fdez-Riverola, F. Diaz, and J. M. Corchado, “Selected papers from the 11th conference of the spanish association for artificial intelligence (caepia 2005)-tokenising, stemming and stopword removal on anti-spam filtering domain,” 2006. [127] A. Abraham, E. Corchado, S.-Y. Han, W. Guo, J. Corchado, and A. Vasilakos, “Next generation web services practices,” Salamanca, vol. 19, p. 21, 2011. [128] D. Glez-Peña, F. D´ıaz, J. R. Méndez, J. M. Corchado, and F. Fdez-Riverola, “An evolutionary approach for sample-based clustering on microarray data,” in International Work-Conference on Artificial Neural Networks. Springer, Berlin, Heidelberg, 2009, pp. 972–978. [129] J. M. Corchado, N. Rees, C. Fyfe, and B. Lees, “Study and comparison of multilayer perceptron nn and radial basis function nn in oceanographic forecasting,” in Applications and Science of Artificial Neural Networks III, vol. 3077. International Society for Optics and Photonics, 1997, pp. 550–561. [130] F. D´ıaz, L. B. Diz, F. F. Riverola, A. Usero, and J. Corchado, “Unsupervised learning for power load forecasting,” 2008. [131] A. Dom´ınguez and J. M. Corchado Rodriguez, “Handling requirements with xml like system specifications,” 2001. [132] M. G. Bedia and J. M. Corchado, “Dynamical systems for dealing with classification tasks in industrial environments,” 2006. [133] P. Ferreiro, A. Blanco, F. Fdez-Riverola, F. D´ıaz, and J. Corchado, “Monitorización distribuida de una red de impresoras,” 2003. [134] J. M. Corchado and A. Mata, “Predicting the presence of oil slicks,” 2008. [135] J. M. Corchado, “Redes neuronales artificiales: un enfoque práctico,” 2000. ´ Román, S. Rodr´ıguez, and J. M. Corchado, “Distributed and specialized agent communities,” in Trends in Practical Applications of Agents and [136] J. A. Multiagent Systems. Springer, Cham, 2013, pp. 33–40. ´ [137] I. Alvarez, J. M. Górriz, J. Ram´ırez, D. Salas-Gonzalez, M. López, C. Puntonet, and F. Segovia, “Alzheimer’s diagnosis using eigenbrains and support vector machines,” Electronics Letters, vol. 45, no. 7, pp. 342–343, 2009. [138] A. Abraham, V. Achim, H. Ahn, S. Aine, C. Aladag, M. Albakoor, L. Ali, M. AL-Rousan, N. Amiri, T. Amraee et al., “Location system,” Arya, vol. 550, p. 1197, 2006. [139] J. Alonso-Meijide, F. Carreras, M. Fiestras-Janeiro, G. Owen, D. Ben-Arieh, T. Easton, F. Fdez-Riverola, E. Iglesias, F. Dıaz, J. Méndez et al., “Telework vs. central work: A comparative view of knowledge accessibility,” 2007. [140] J. Cabestany, F. Sandoval, A. Prieto, and J. M. Corchado, “Bio-inspired systems: Computational and ambient intelligence,” Lecture Notes in Computer Science, vol. 5517, 2009. [141] M. Abderrahim, R. Abdulwahhab, P. J. Adeodato, E. Agafonov, A. Alexandridis, J. Andreu, P. Angelov, C. Arismendi, F. Armetta, A. L. Arnaud et al., “Dal mas,” 2012. [142] J. Mendez, F. Fdez-Riverola, F. Diaz, and J. Corchado, “Intelligent systems for detecting and filtering spam e-mail: a survey,” INTELIGENCIA ARTIFICIAL, vol. 11, no. 34, 2007. ´ [143] J. Aalmink, D. Adzinets, J. Agüero, F. Alayón, L. Alonso, N. Alonso, R. S. Alonso, F. Alvarez-Rodrıguez, M. Amor, R. Antonello et al., “Calvillomoreno, edgar a. 1 camacho, david 371, 379 campos, c. 77,” Advances in Intelligent and Soft Computing 71, vol. 100, p. 731, 2010.


10

[144] R. Fuentes-Fernández, S. Astorga, M. Aulinas, I. Ayala, M. Bergeret, O. Boissier, T. Bosse, L. Búrdalo, B. Bouzy, R. A. Braga et al., “Dugdale, julie 25 esparcia, sergio 35 fernández-caballero, antonio 91 fuentes, lidia 279,” in Advances in Practical Applications of Agents and Multiagent Systems: 8th International Conference on Practical Applications of Agents and Multiagent Systems (PAAMS’10), vol. 70. Springer Science & Business Media, 2010, p. 291. [145] S. Deusdado, J. Alves, M. Alves, Y. Audic, N. F. Azevedo, C. Baldwin, R. Benito, E. Berger, L. Bertel-Paternina, J. Bisbal et al., “International conference on practical applications of computational biology & bioinformatics,” in 6th International Conference on Practical Applications of Computational Biology & Bioinformatics, 2012, p. 285. [146] A. Abdullah, K. Abe, M. Aguilera, M. Akiyoshi, T. Alamos, J. Alonso, R. S. Alonso, R. Amami, I. Angulo, A. Aouichat et al., “Distributed computing and artificial intelligence,” Distributed Computing and Artificial Intelligence, p. 775, 2012. ´ Alvarez, ´ ´ [147] E. Alba, A. Alcayde, L. Aldamiz-Echevarria, M. J. Algar, S. Altamirano, M. A. J. Alvarez-Bravo, F. Andrade, I. Angulo, N. Antón et al., “Distributed computing and artificial iintelligence,” Advances in Intelligent and Soft Computing 79, p. 699, 2010. [148] P. Garg, S. Bandyopadhyay, B. Barath Kumar, N. Bellamine, N. Ben Yahia, V. Bhadran, A. Bhandari, S. Bhardwaj, A. K. Bhartee, S. Bhowmik et al., “Ekbal, asif 9, 57 faili, heshaam 443 fliss, imtiez 129 gangwal, neeraj 357,” in Intelligent Informatics: Proceedings of the International Symposium on Intelligent Informatics ISI12 Held at August 4-5 2012, Chennai, India, vol. 182. Springer Science & Business Media, 2012, p. 519. [149] R. Igual, L. Cardoso, D. Carneiro, F. Catalao, E. Cerezo, S. Chessa, M. Corbalan, J. M. Corchado, and R. Costa, “Hermann, thomas 71, 229 heynderickx, ingrid 195 horvath, louise beltzung 163 ibanez-sánchez, gema 105,” in Ambient Intelligence-Software and Applications: 4th International Symposium on Ambient Intelligence (ISAmI 2013, vol. 219. Springer Science & Business Media, 2013, p. 245. [150] D. B. L. Castillo and J. M. Corchado, “Current topics in artificial intelligence,” 2007. [151] M. Fathi, J. Baptista, M. R. Barbosa, D. F. Barrero, B. Baruque, I. Bertini, E. Bittencourt, M. Boubekeur, K. N. Brown, J. Bulas-Cruz et al., “Soft computing models in industrial and environmental applications,” Advances in Intelligent and Soft Computing 73, p. 261, 2010. [152] Y. Aloimonos, E. Ambikairajah, C. An, D. Avrahami, A. Aztiria, A. A. Baptista, J. Bardram, J. Barroso, J. Blaya, T. Bosse et al., “Reviewers index,” Journal of Ambient Intelligence and Smart Environments, vol. 1, no. 377, p. 377, 2009. [153] C. Bernabéu, J. M. Corchado Rodr´ıguez, S. Rodr´ıguez, and P. Chamoso, “Aracnocóptero: An unmanned aerial vtol multi-rotor for remote monitoring and surveillance,” 2011. [154] T. Li, S. Sun, T. P. Sattar, and J. M. Corchado, “Fight sample degeneracy and impoverishment in particle filters: A review of intelligent approaches,” Expert Systems with applications, vol. 41, no. 8, pp. 3944–3954, 2014. [155] C. B. A. Salmerón, A. Alonso-Betanzos, J. I. H. L. Mart´ınez, A. T. E. Corchado, and J. M. Corchado, “Advances in artificial intelligence,” 2013. [156] J. A. Valle and J. M. Corchado, “Sistema multiagente basado en organizaciones para el control de v´ıdeo multidispositivo,” Avances en Informática y Automática, p. 149, 2013. [157] D. H. Alfageme and J. M. Corchado, “Sistema multiagente desplegado en un entorno cloud computing para gestión de zonas de aparcamiento,” Avances en Informática y Automática, p. 93, 2013. [158] A. Almeida and D. López-de Ipiña, “An approach to more reliable context-aware systems by assessing ambiguity-taking into account indetermination and vagueness in smart environments.” in PECCS, 2012, pp. 233–236. [159] C. Bielza, A. Salmerón, A. Alonso-Betanzos, J. I. Hidalgo, L. Mart´ınez, A. Troncoso, E. Corchado, and J. M. Corchado, “Advances in artificial intelligence: 15th conference of the spanish association for artificial intelligence, caepia 2013, madrid, september 17-20, 2013, proceedings,” 2013. [160] E. Corchado, J. M. C. Rodr´ıguez, and A. Abraham, “Innovations in hybrid intelligent systems,” 2007. [161] W. Coulier, F. Garijo, J. Gomez, J. Pavon, P. Kearney, and P. Massonet, “Message: a methodology for the development of agent-based applications,” Methodologies and Software Engineering for Agent Systems-The Agent-Oriented Software Engineering Handbook, F. Bergenti, MP Gleizes, F. Zambonelli (Eds.), Kluwer, pp. 177–194, 2004. [162] J. R. Méndez, F. Fdez Riverola, F. D´ıaz, and J. M. Corchado, “Sistemas inteligentes para la detección y filtrado de correo spam: una revisión,” Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial, vol. 11, no. 34, 2007. [163] L. F. Castillo, J. Corchado, and M. G. Bedia, “Modelo y desarrollo de w-planner: sistema multiagente on-line aplicado al turismo electrónico,” Ingenier´ıa y Ciencia, vol. 1, no. 1, 2005. [164] J. Méndez, E. Iglesias, F. Fdez-Riverola, F. D´ıaz, and J. Corchado, “Analyzing the impact of corpus preprocessing on anti-spam filtering software,” Research on Computing Science, vol. 17, pp. 129–138, 2005. [165] H. Billhardt, V. Julián, J. M. Corchado, and A. Fernández, “An architecture proposal for human-agent societies,” in International Conference on Practical Applications of Agents and Multi-Agent Systems. Springer, Cham, 2014, pp. 344–357. [166] K. Matsui, K. Kimura, A. Pérez, S. Rodr´ıguez, and J. M. Corchado, “Development of electrolarynx by multi-agent technology and mobile devices for prosody control,” in International Conference on Practical Applications of Agents and Multi-Agent Systems. Springer, Cham, 2014, pp. 54–65. [167] A. Sanchis, V. Julián, J. M. Corchado, H. Billhardt, and C. Carrascosa, “Using natural interfaces for human-agent immersion,” in International Conference on Practical Applications of Agents and Multi-Agent Systems. Springer, Cham, 2014, pp. 358–367. [168] J. A. Román, S. Rodr´ıguez, and J. M. Corchado, “Improving intelligent systems: Specialization,” in International Conference on Practical Applications of Agents and Multi-Agent Systems. Springer, Cham, 2014, pp. 378–385. [169] S. Omatu, T. Wada, S. Rodr´ıguez, P. Chamoso, and J. M. Corchado, “Multi-agent technology to perform odor classification,” in Ambient IntelligenceSoftware and Applications. Springer, Cham, 2014, pp. 241–252. [170] A. Almeida, I. Larizgoitia, D. Lopez-de Ipiña, X. Laiseca, A. Barbier, U. Aguilera, and P. Orduña, “Dynamic ontology enrichment and reasoningin ami environments,” in Proceedings of the 3rd Workshop on Artificial Intelligence, 2008. [171] T. Li, S. Sun, J. M. Corchado, and M. F. Siyau, “A particle dyeing approach for track continuity for the smc-phd filter,” in Information Fusion (FUSION), 2014 17th International Conference on. IEEE, 2014, pp. 1–8. [172] ——, “Random finite set-based bayesian filters using magnitude-adaptive target birth intensity,” in Information Fusion (FUSION), 2014 17th International Conference on. IEEE, 2014, pp. 1–8. [173] R. Evans, P. Besnard, E. Blasch, H. Blom, Y. Boers, D. Carevic, M. Cetin, M. Chan, K. Chang, H. Chen et al., “Information fusion,” in 16 International Conference on Information Fusion, 2013. [174] P. Chamoso, A. Pérez, S. Rodr´ıguez, J. M. Corchado, M. Sempere, R. Rizo, F. Aznar, and M. Pujol, “Modeling oil-spill detection with multirotor systems based on multi-agent systems,” in Information Fusion (FUSION), 2014 17th International Conference on. IEEE, 2014, pp. 1–8. [175] Y. W. Choon, M. S. Mohamad, S. Deris, R. M. Illias, C. K. Chong, L. E. Chai, S. Omatu, and J. M. Corchado, “Differential bees flux balance analysis with optknock for in silico microbial strains optimization,” PloS one, vol. 9, no. 7, p. e102744, 2014. [176] Y. W. Choon, M. S. Mohamad, S. Deris, C. K. Chong, S. Omatu, and J. M. Corchado, “Gene knockout identification using an extension of bees hill flux balance analysis,” BioMed research international, vol. 2015, 2015. [177] E. Sancristobal, S. Martin, R. Gil, R. Pastor, J. Garc´ıa-Zubia, P. Orduña, A. Pesquera, G. Diaz, J. Harward, and M. Castro, “Service-labs: Reusing services and laboratories from open learning management systems,” ICL2009. Interactive Computer aided Learning. Villach, Austria: September, pp. 23–25, 2009. [178] I. Hupont, F. Barber, C. Benito, G. Bocewicz, M. L. Borrajo, R. A. Bravo, F. Cacheda, J. Catalina, J. M. Corchado, R. Corchuelo et al., “Trends in practical applications of agents and multiagent systems,” Trends in Practical Applications of Agents and Multiagent Systems, p. 165, 2012. [179] E. Redondo-Gonzalez, L. N. de Castro, J. Moreno-Sierra, M. de las Casas, M. Luisa, V. Vera-Gonzalez, D. G. Ferrari, and J. M. Corchado, “Bladder carcinoma data with clinical risk factors and molecular markers: a cluster analysis,” BioMed research international, vol. 2015, 2015.


11

[180] J. A. Román, S. Rodr´ıguez, J. M. Corchado, C. Carrascosa, and S. Ossowski, “Specialization: A new way to improve intelligent systems,” International Journal of Artificial Intelligence, vol. 13, no. 1, pp. 58–73, 2015. [181] A. Sanchis, V. Julián, J. M. Corchado, H. Billhardt, and C. Carrascosa, “Improving human-agent immersion using natural interfaces and cbr,” International Journal of Artificial Intelligence, vol. 13, no. 1, pp. 81–93, 2015. [182] H. Billhardt, V. Julián, J. M. Corchado, and A. Fernández, “Human-agent societies: challenges and issues,” International Journal of Artificial Intelligence, vol. 13, no. 1, pp. 28–44, 2015. [183] P. C. Marchante, “Reforma, interiorismo y diseño de mobiliario para una vivienda antigua,” Ph.D. dissertation, 2011. [184] P. Vieira and J. Corchado, “A formal machines as a player of a game,” in Distributed Computing and Artificial Intelligence, 12th International Conference. Springer, Cham, 2015, pp. 137–147. [185] C. Grosan and A. Abraham, “Knowledge representation and reasoning,” in Intelligent Systems. Springer, Berlin, Heidelberg, 2011, pp. 131–147. [186] G. Goos, J. Hartmanis, J. v. Leeuwen, W. Brauer, D. Gries, and J. Stoer, “Lecture notes in computer science,” Lecture Notes in Computer Science, vol. 1164, p. 1164, 1996. [187] A. H. M. Salleh, M. S. Mohamad, S. Deris, S. Omatu, F. Fdez-Riverola, and J. M. Corchado, “Gene knockout identification for metabolite production improvement using a hybrid of genetic ant colony optimization and flux balance analysis,” Biotechnology and bioprocess engineering, vol. 20, no. 4, pp. 685–693, 2015. [188] A. C. E. Lima, L. N. de Castro, and J. M. Corchado, “A polarity analysis framework for twitter messages,” Applied Mathematics and Computation, vol. 270, pp. 756–767, 2015. [189] S. E. Costa, J. J. Rodrigues, B. M. Silva, J. N. Isento, and J. M. Corchado, “Integration of wearable solutions in aal environments with mobility support,” Journal of medical systems, vol. 39, no. 12, p. 184, 2015. [190] J. M. Corchado, M. L. Borrajo, M. A. Pellicer, J. C. Yán˜ ez, and P. Perner, “Advances in data mining: Applications in image mining, medicine and biotechnology, management and environmental control, and telecommunications; 4th industrial conference on data mining, icdm 2004, leipzig, germany, july 4-7, 2004, revised selected papers,” 2005. [191] L. L. Kleine, V. A. V. Ruiz, M. P. Rocha, F. F. Riverola, H. Shatkay, and J. M. Corchado, “Advances in bioinformatics: 4th international workshop on practical applications of computational biology and bioinformatics 2010 (iwpacbb 2010),” 2010. [192] H. Ghasemzadeh, R. Jafari, B. Prabhakaran, R. Fletcher, K. Dobson, M. Goodwin, H. Eydgahi, O. Wilder-Smith, D. Fernholz, Y. Kuboyama et al., “Special section on affective and pervasive computing for healthcare,” 2010. [193] P. Wah Tang, P. San Chua, S. Kee Chong, M. Saberi Mohamad, Y. Wen Choon, S. Deris, S. Omatu, J. Manuel Corchado, W. Howe Chan, and R. Abdul Rahim, “A review of gene knockout strategies for microbial cells,” Recent patents on biotechnology, vol. 9, no. 3, pp. 176–197, 2015. [194] L. Snidaro, J. Garc´ıa, and J. M. Corchado, “Context-based information fusion [guest editorial],” 2015. [195] W. Raveane and M. A. G. Arrieta, “Texture classification with neural networks,” in Distributed Computing and Artificial Intelligence. Springer, Cham, 2013, pp. 325–332. [196] B. Lees and J. Corchado, “Integrated case-based neural network approach,” in XPS-99: Knowledge-Based Systems-Survey and Future Directions: 5th Biannual German Conference on Knowledge-Based Systems, Würzburg, Germany, March 3-5, 1999, Proceedings, vol. 1570. Springer Science & Business Media, 1999, p. 157. [197] M. D. Muñoz, A. Mata, E. Corchado, and J. M. Corchado, “(obifs) isotropic image analysis for improving a predicting agent based systems,” Expert Systems with Applications, vol. 40, no. 12, pp. 5011–5020, 2013. [198] J. V. Abad, F. Fdez-Riverola, and J. Corchado, “Forecasting economic cycles with conexionist models,” 2000. [199] J. Corchardo, B. Lees, C. Fyfe, N. Rees, and J. Aiken, “Neuro-adaptation method for a case-based reasoning system,” in Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on, vol. 1. IEEE, 1998, pp. 713–718. [200] E. Alpaydın, “Guest editorial,” 1999. [201] M. Bressan, J. Vitria, A. Pons-Porrata, R. Berlanga-Llavori, J. Ruız-Shulcloper, G. Guo, H. Wang, D. Bell, J. R. Alcobé, M. Macıas-Macıas et al., “Knowledge representation and reasoning,” 2010. [202] K.-D. Althoff, R. Bergmann, M. Minor, and A. Hanft, “Advances in case-based reasoning,” in 9th European Conference, ECCBR, 2008, pp. 1–4. [203] E. Corchado Rodr´ıguez and H. Yin, “Intelligent data engineering and automated learning-ideal 2009,” 2009. [204] J. Neves and J. M. Machado, “Progress in artificial intelligence,” 2007. [205] J. A. F. Canelas, Q. M. Martin, and J. M. C. Rodriguez, “Argumentative sox compliant and quality decision support intelligent expert system over the suppliers selection process,” Applied Computational Intelligence and Soft Computing, vol. 2013, p. 7, 2013. [206] M. S. Pérez Santos, J. M. Corchado Rodr´ıguez, J. M. Mateos Roco, F. Atrio Barandela, A. Calvo Hernández, and M. Egido Manzano, “Nuevos recursos en la práctica docente en el grado en f´ısica: la pizarra digital interactiva,” 2011. [207] J. A. Fernandez, Q. M. Martin, and J. M. C. Rodriguez, “Business intelligence expert system on sox compliance over the purchase orders creation process,” Intelligent Information Management, vol. 5, no. 3, p. 49, 2013. [208] M. Cristancho, G. Isaza, A. Pinzón, J. M. C. Rodr´ıguez, and L. F. Castillo, Advances in Computational Biology. Springer, 2014. [209] M. G. Bedia, J. M. C. Rodr´ıguez, and J. O. Garc´ıa, “Principios dinámicos en el estudio de la percepción,” in Una perspectiva de la inteligencia artificial en su 50 aniversario: Campus Multidisciplinar en Percepción e Inteligencia, CMPI 2006, Albacete, España, 10-14 de Julio del 2006: actas. Universidad de Castilla-La Mancha, 2006, pp. 463–474. [210] M. L. B. Diz, J. C. Y. López, R. L. Fidalgo, and J. M. C. Rodr´ıguez, “Sistema inteligente para la realización del control interno en el sector empresarial,” in Emergent solutions for the information and knowledge economy: proceedings of the Tenth International Association for Fuzzy-Set Management and Economy Congress: León, october 9-11, 2003. Secretariado de Publicaciones y Medios Audiovisuales, 2003, pp. 213–227. [211] J. M. Mateos Roco, P. Garc´ıa Estévez, J. I. Íñiguez de la Torre Bayo, T. González Sánchez, L. Plaja Rustein, M. S. Pérez Santos, I. Íñiguez-de-la Torre, F. d. l. Prieta Pintado, A. J. Sánchez Mart´ın, and J. M. Corchado Rodr´ıguez, “Actividades de producción digital destinadas a la difusión y promoción del máster universitario en f´ısica,” 2013. [212] L. S. Aguilar and J. M. C. Rodr´ıguez, “Desarrollo de sistemas multiagentes,” in 3rd International workshop on practical applications of agents and multiagent systems: IWPAAMS 2004. Universidad de Burgos, 2004, pp. 343–346. [213] J. M. Mateos Roco, M. S. Pérez Santos, and J. M. Corchado Rodr´ıguez, “Procedimientos de apoyo para la realización del seguimiento del t´ıtulo de grado en f´ısica,” 2012. ´ M. Giménez, “Utilización de datos de envisat para la detección de vertidos [214] L. G. Vilas, G. M. Iglesias, J. M. C. Rodr´ıguez, J. M. T. Palenzuela, and A. de hidrocarburos,” Revista de teledetección: Revista de la Asociación Española de Teledetección, no. 25, pp. 55–59, 2006. [215] L. Uden, L. S. Wang, J. M. C. Rodr´ıguez, H.-C. Yang, and I.-H. Ting, The 8th International Conference on Knowledge Management in Organizations: Social and Big Data Computing for Knowledge Management. Springer Science & Business Media, 2013. [216] J. M. C. Rodr´ıguez, J. C. Y. López, and M. L. B. Diz, “Case-based reasoning systems: tols to assit the business decision taken process,” in Emergent solutions for the information and knowledge economy: proceedings of the Tenth International Association for Fuzzy-Set Management and Economy Congress: León, october 9-11, 2003. Secretariado de Publicaciones y Medios Audiovisuales, 2003, pp. 157–171. [217] M. G. Bedia, L. F. Castillo, and J. M. C. Rodr´ıguez, “W-planer: Sistema multiagente de tiempo real aplicado al turismo electrónico,” in 3rd International workshop on practical applications of agents and multiagent systems: IWPAAMS 2004. Universidad de Burgos, 2004, pp. 135–146. [218] J. M. C. Rodr´ıguez, “Desarrollo de sistemas autónomos mediante agentes,” in Simposio de Informática y Telecomunicaciones, 2002, pp. 9–10. [219] ——, “Sistemas multiagente basados en algoritmos h´ıbridos de inteligencia artificial para predicciones en tiempo real y series complejas,” Ph.D. dissertation, Universidad de Salamanca, 1998.


12

[220] R. L. Fidalgo and J. M. C. Rodr´ıguez, “Cbr-bdi agentes in planning,” in Simposio de Informática y Telecomunicaciones, 2002, pp. 181–192. [221] J. M. Molina, J. M. C. Rodr´ıguez, and J. P. Mestras, “Modelos y arquitecturas de agente,” in Agentes software y sistemas multiagente: conceptos, arquitecturas y aplicaciones. Prentice Hall, 2005, pp. 29–64. [222] R. L. Fidalgo, J. M. C. Rodr´ıguez, and L. F. C. Ossa, “Arquitectura basada en agentes para desarrollar aplicaciones de internet,” Novática: Revista de la Asociación de Técnicos de Informática, no. 170, pp. 30–34, 2004. [223] J. M. C. Rodr´ıguez, D. G. Peña, F. F. Riverola, and F. D. Gómez, “Análisis de microarrays mediante genecbr,” in Actas del V Congreso Español sobre Metaheur´ısticas, Algoritmos Evolutivos y Bioinspirados, 2007, pp. 435–442. [224] M. S. Pérez Santos, J. M. Corchado Rodr´ıguez, J. M. Mateos Roco, and J. Mateos López, “Material para el estudio de la desviación de electrones en campos eléctricos y magnéticos,” 2010. [225] J. M. Mateos Roco, A. González Sánchez, T. González Sánchez, L. López D´ıaz, F. Fernández González, P. Garc´ıa Estévez, C. Rodr´ıguez Puebla, and J. M. Corchado Rodr´ıguez, “Difusión y promoción del grado en f´ısica y del máster en f´ısica,” 2013. [226] J. M. C. Rodr´ıguez, M. A. P. P. Figueras, M. L. B. Diz, and J. C. Y. López, “The impact of the multiagent technology in the e-business and e-commerce,” in Emergent solutions for the information and knowledge economy: proceedings of the Tenth International Association for Fuzzy-Set Management and Economy Congress: León, october 9-11, 2003. Secretariado de Publicaciones y Medios Audiovisuales, 2003, pp. 345–354. [227] J. M. C. Rodriguez, L. F. Castillo, and M. G. Bedia, “Librer´ıa para el intercambio de mensajes acl entre dispositivos móviles a través de bluetooth,” 2007. [228] M. G. Bedia, J. M. C. Rodr´ıguez, and J. O. Garc´ıa, “Arquitecturas emocionales en inteligencia artificial,” in Una perspectiva de la inteligencia artificial en su 50 aniversario: Campus Multidisciplinar en Percepción e Inteligencia, CMPI 2006, Albacete, España, 10-14 de Julio del 2006: actas. Universidad de Castilla-La Mancha, 2006, pp. 186–193. [229] J. M. C. Rodr´ıguez, “Gu´ıa docente de inteligencia artificial distribuida: Agentes y sistemas multiagente (1, 5 ects),” 2006. [230] C. A. González and J. M. C. Rodr´ıguez, “Ingenier´ıa del conocimiento: Introducción a la metodolog´ıa commonkads profesores,” 2005. [231] J. M. C. Rodr´ıguez, “Intervención de d. juan manuel corchado rodr´ıguez,” in Actas del 12o Congreso de Econom´ıa de Castilla y León. Consejer´ıa de Econom´ıa y Hacienda, 2011, p. 43. [232] M. González Bedia and J. M. Corchado Rodr´ıguez, “Constructing autonomous distributed systems using cbr-bdi agents.” 2003. ´ Fernández Canelas, M. Quint´ın Mart´ın, and J. M. Corchado Rodr´ıguez, “Decision making intelligent agent on sox compliance over the goods [233] J. A. receipt processs,” 2013. [234] F. de la Prieta Pintado and J. M. Corchado Rodr´ıguez, “Cloud computing and multiagent systems, a promising relationship.” 2015. [235] J. M. Corchado Rodr´ıguez and B. Lees, “Adaptation of cases for case based forecasting with neural network support,” 2001. [236] J. M. Corchado-Rodr´ıguez, R. Laza-Fidalgo, and L. F. Castillo-Ossa, “An agent-based architecture for developing internet-based applications,” The European Journal for the Informatics Professional, vol. 5, no. 4, 2004. [237] L. F. Castillo, M. Cristancho, G. Isaza, A. Pinzón, and J. M. C. Rodr´ıguez, Advances in Computational Biology: Proceedings of the 2nd Colombian Congress on Computational Biology and Bioinformatics (CCBCOL). Springer Science & Business Media, 2013, vol. 232. [238] A. Abraham, S. K. Pal, J. M. C. Rodriguez, and S. M. Thampi, Recent Advances in Intelligent Informatics. Springer, 2014. [239] J. Casillas, F. J. Mart´ınez-López, and J. M. C. Rodr´ıguez, Management Intelligent Systems: First International Symposium. Springer Science & Business Media, 2012, vol. 171. [240] L. F. Castillo, M. Cristancho, G. Isaza, A. Pinzón, and J. M. C. Rodr´ıguez, “Advances in computational biology,” 2014. [241] M. Rocha, N. Luscombe, J. Rodr´ıguez, and F. Fdez-Riverola, “Advances in intelligent and soft computing: Preface,” Advances in Intelligent and Soft Computing, vol. 154, 2012. [242] J. L. González-Sánchez, A. G. Cervero, J. Luis, G. Rivera, and J. M. M. Rodr´ıguez, “Mvc: Technical description of a proposal in multimedia and distance education,” 2000. [243] R. L. Fidalgo, R. P. Rial, A. G. Rodr´ıguez, and J. M. C. Rodr´ıguez, “Automatic generation of shellfish exploitation plans,” 2004. [244] J. Canelas, Q. Martin, and J. M. Corchado, “Decision making intelligent agent on sox compliance over the imports process,” 2014. [245] J. A. F. Canelas, Q. M. Martin, and J. M. C. Rodriguez, “Argumentative sox compliant and intelligent decision support systems for the suppliers contracting process,” in Intelligent Techniques in Engineering Management. Springer, 2015, pp. 333–375. [246] J. M. C. Rodr´ıguez, R. L. Fidalgo, M. R. P. Rial, and A. M. G. Rodr´ıguez, “Integrating rule based systems which change along time in a cbr environment,” in Simposio de Informática y Telecomunicaciones, 2002, pp. 193–202. [247] M. P. Rocha, J. M. C. Rodr´ıguez, F. F. Riverola, and A. Valencia, 5th International Conference on Practical Applications of Computational Biology & Bioinformatics. Springer Science & Business Media, 2011, vol. 93. [248] J. A. F. Canelas, Q. M. Martin, and J. M. C. Rodriguez, “Decision making intelligent agent on sox compliance over the goods receipt processs,” Decision Making, vol. 4, no. 10, 2013. [249] S. Rodr´ıguez González, “Modelo adaptativo para organizaciones virtuales de agentes,” 2010. [250] M. Graña, E. Corchado, M. Wozniak, D. Simic, I. Kovacevic, S. Simic, B. Kurlej, C.-M. Pintea, G. C. Crisan, C. Chira et al., “Logic journal of the igpl,” 2012. [251] J. C. Galván, J. R. Pérez, and J. M. A. Rodr´ıguez, “Tuberculosis bovina. cl´ınica y lesiones.” Bovis, no. 47, pp. 43–54, 1992. [252] J. M. A. Rodr´ıguez and J. C. Galván, “Epizootiolog´ıa y control de la ascosferiosis en españa,” in Jornadas técnicas: IX feria regional ap´ıcola de Castilla-La Mancha. Patronato Rector de la Feria Regional Ap´ıcola de Castilla-La Mancha, 1991, pp. 39–49. [253] J. G. Sánchez León and J. M. Rodr´ıguez, “Diseño optimo de experimentos aplicado al establecimiento de programas de bioensayos,” 2007. [254] J. Crochado and C. Fyfe, “A comparison of kernel methods for instantiating case based reasoning systems,” Computing and Information Systems, vol. 7, no. 2, pp. 29–42, 2000.