José A. Castellanos-Garzón

Intelligent business processes composition based on multi-agent systems

This paper proposes a novel model for automatic construction of business processes called IPCASCI (Intelligent business Processes Composition based on multi-Agent systems, Semantics and Cloud Integration). The software development industry requires agile construction of new products able to adapt to the emerging needs of a changing market. In this context, we present a method of software component reuse as a model (or methodology), which facilitates the semi-automatic reuse of web services on a cloud computing environment, leading to business process composition. The proposal is based on web service technology, including: (i) Automatic discovery of web services; (ii) Semantics description of web services; (iii) Automatic composition of existing web services to generate new ones; (iv) Automatic invocation of web services. As a result of this proposal, we have presented its implementation (as a tool) on a real case study. The evaluation of the case study and its results are proof of the reliability of IPCASCI.

An Agent-Based Clustering Approach for Gene Selection in Gene Expression Microarray

Gene selection is a major research area in microarray analysis, which seeks to discover differentially expressed genes for a particular target annotation. Such genes also often called informative genes are able to differentiate tissue samples belonging to different classes of the studied disease. Despite the fact that there is a wide number of proposals, the complexity imposed by this problem remains a challenge today. This research proposes a gene selection approach by means of a clustering-based multi-agent system. This proposal manages different filter methods and gene clustering through coordinated agents to discover informative gene subsets. To assess the reliability of our approach, we have used four important and public gene expression datasets, two Lung cancer datasets, Colon and Leukemia cancer dataset. The achieved results have been validated through cluster validity measures, visual analytics, a classifier and compared with other gene selection methods, proving the reliability of our proposal.

A Framework for Knowledge Discovery from Wireless Sensor Networks in Rural Environments: A Crop Irrigation Systems Case Study

This paper presents the design and development of an innovative multiagent system based on virtual organizations. The multiagent system manages information from wireless sensor networks for knowledge discovery and decision making in rural environments. The multiagent system has been built over the cloud computing paradigm to provide better flexibility and higher scalability for handling both small- and large-scale projects. The development of wireless sensor network technology has allowed for its extension and application to the rural environment, where the lives of the people interacting with the environment can be improved. The use of “smart” technologies can also improve the efficiency and effectiveness of rural systems. The proposed multiagent system allows us to analyse data collected by sensors for decision making in activities carried out in a rural setting, thus, guaranteeing the best performance in the ecosystem. Since water is a scarce natural resource that should not be wasted, a case study was conducted in an agricultural environment to test the proposed system’s performance in optimizing the irrigation system in corn crops. The architecture collects information about the terrain and the climatic conditions through a wireless sensor network deployed in the crops. This way, the architecture can learn about the needs of the crop and make efficient irrigation decisions. The obtained results are very promising when compared to a traditional automatic irrigation system.

A CBR framework with gradient boosting based feature selection for lung cancer subtype classification

Molecular subtype classification represents a challenging field in lung cancer diagnosis. Although different methods have been proposed for biomarker selection, efficient discrimination between adenocarcinoma and squamous cell carcinoma in clinical practice presents several difficulties, especially when the latter is poorly differentiated. This is an area of growing importance, since certain treatments and other medical decisions are based on molecular and histological features. An urgent need exists for a system and a set of biomarkers that provide an accurate diagnosis. In this paper, a novel Case Based Reasoning framework with gradient boosting based feature selection is proposed and applied to the task of squamous cell carcinoma and adenocarcinoma discrimination, aiming to provide accurate diagnosis with a reduced set of genes. The proposed method was trained and evaluated on two independent datasets to validate its generalization capability. Furthermore, it achieved accuracy rates greater than those of traditional microarray analysis techniques, incorporating the advantages inherent to the Case Based Reasoning methodology (e.g. learning over time, adaptability, interpretability of solutions, etc.).

A Clustering-Based Method for Gene Selection to Classify Tissue Samples in Lung Cancer

This paper proposes a gene selection approach based on clustering of DNA-microarray data. The proposal has been aimed at finding a boundary gene subset coming from gene groupings imposed by a clustering method applied to the case study: gene expression data in lung cancer. Thus, we assume that such a found gene subset represents informative genes, which can be used to train a classifier by learning tumor tissue samples. To do this, we compare the results of several methods of hierarchical clustering to select the best one and then choose the most suitable clustering based on visualization techniques. The latter is used to compute its boundary genes. The results achieved from the case study have shown the reliability of this approach.

An Evolutionary Hierarchical Clustering Method with a Visual Validation Tool

In this paper, we propose a novel hierarchical clustering method based on evolutionary strategies. This method leads to gene expression data analysis, and shows its effectiveness with regard to other clustering methods through cluster validity measures on the results. Additionally, a novel visual validation interactive tool is provided to carry out visual analytics among clusters of a dendrogram. This interactive tool is an alternative for the used validity measures. The method introduced here attempts to solve some of the problems faced by other hierarchical methods. Finally, the results of the experiments show that the method can be very effective in the cluster analysis on DNA microarray data.

Evolutionary Techniques for Hierarchical Clustering Applied to Microarray Data

In this paper we propose a novel hierarchical clustering method that uses a genetic algorithm based on mathematical proofs for the analysis of gene expression data, and show its effectiveness with regard to other clustering methods. The analysis of clusters with genetic algorithms has disclosed good results on biological data, and several studies have been carried out on the latter, although the majority of these researches have been focused on the partitional approach. On the other hand, the deterministic methods for hierarchical clustering generally converge to a local optimum. The method introduced here attempts to solve some of the problems faced by other hierarchical methods. The results of the experiments show that the method could be very effective in the cluster analysis on DNA microarray data.

A Genetic Programming Approach Applied to Feature Selection from Medical Data.

Genetic programming represents a flexible and powerful evolutionary technique in machine learning. The use of genetic programming for rule induction has generated interesting results in classification problems. This paper proposes an evolutionary approach for logical rule induction, which is applied to clinical data. Since logical rules disclose knowledge from the analyzed data, we use such a knowledge to filter features from the target dataset. The results reached by the used dataset have been very promising when used in classification tasks and compared with other methods.

An Ensemble Framework Coping with Instability in the Gene Selection Process

This paper proposes an ensemble framework for gene selection, which is aimed at addressing instability problems presented in the gene filtering task. The complex process of gene selection from gene expression data faces different instability problems from the informative gene subsets found by different filter methods. This makes the identification of significant genes by the experts difficult. The instability of results can come from filter methods, gene classifier methods, different datasets of the same disease and multiple valid groups of biomarkers. Even though there is a wide number of proposals, the complexity imposed by this problem remains a challenge today. This work proposes a framework involving five stages of gene filtering to discover biomarkers for diagnosis and classification tasks. This framework performs a process of stable feature selection, facing the problems above and, thus, providing a more suitable and reliable solution for clinical and research purposes. Our proposal involves a process of multistage gene filtering, in which several ensemble strategies for gene selection were added in such a way that different classifiers simultaneously assess gene subsets to face instability. Firstly, we apply an ensemble of recent gene selection methods to obtain diversity in the genes found (stability according to filter methods). Next, we apply an ensemble of known classifiers to filter genes relevant to all classifiers at a time (stability according to classification methods). The achieved results were evaluated in two different datasets of the same disease (pancreatic ductal adenocarcinoma), in search of stability according to the disease, for which promising results were achieved.

An Ensemble Approach for Gene Selection in Gene Expression Data

Feature/Gene selection is a major research area in the study of gene expression data, generally dealing with classification tasks of diseases or subtype of diseases and identification of biomarkers related to a type of disease. In such a context, this paper proposes an ensemble approach of gene selection for classification tasks from gene expression datasets. This proposal provides a four-staged approach of gene filtering. Each stage performs a different gene filtering task, such as: data processing, noise removing, gene selection ensemble and application of wrapper methods to reach the end result, a small subset of informative genes. Our proposal has been assessed on two different datasets of the same disease (Pancreatic ductal adenocarcinoma) for which, good results have been achieved in comparison with other gene selection methods. Hence, the proposed strategy has proven its reliability with respect to other approaches.