José del Sagrado

Qualitative combination of Bayesian networks

Directed graphic models based on conditional independence provide a compact and concise representation of an experts subjective belief about existing relationships between variables. Faced with the task of building a greater model, each expert must be a specialist in some subset of the whole knowledge domain. It would be desirable to aggregate the knowledge provided by those specialists under the form of graphical models into a single and more general representation. This article studies the consensus model that would be obtained by combining two graphs associated with Bayesian networks and applying the union and intersection of their independencies.

Aggregation of Imprecise Probabilities

Methods to aggregate convex sets of probabilities are proposed. Source reliability is taken into account by transforming the given information and making it less precise. An important property of the aggregation will be that the precision of the result will depend on the initial compatibility of sources. Special attention will be paid to the particular case of probability intervals giving adaptations of aggregation procedures.

Ant Colony Optimization for the Next Release Problem: A Comparative Study

The selection of the enhancements to be included in the next software release is a complex task in every software development. Customers demand their own software enhancements, but all of them cannot be included in the software product, mainly due to the existence limited resources. In most of the cases, it is not feasible to develop all the new functionalities suggested by customers. Hence each new feature competes against each other to be included in the next release. This problem of minimizing development effort and maximizing customers’ satisfaction is known as the next release problem (NRP). In this work we study the NRP problem as an optimisation problem. We use and describe three different meta-heuristic search techniques for solving NRP: simulated annealing, genetic algorithms and ant colony system (specifically, we show how to adapt the ant colony system to NRP). All of them obtain good but possibly sub optimal solution. Also we make a comparative study of these techniques on a case study. Furthermore, we have observed that the sub optimal solutions found applying these techniques include a high percentage of the requirements considered as most important by each individual customer.

Multi-objective ant colony optimization for requirements selection

The selection of a set of requirements between all the requirements previously defined by customers is an important process, repeated at the beginning of each development step when an incremental or agile software development approach is adopted. The set of selected requirements will be developed during the actual iteration. This selection problem can be reformulated as a search problem, allowing its treatment with metaheuristic optimization techniques. This paper studies how to apply Ant Colony Optimization algorithms to select requirements. First, we describe this problem formally extending an earlier version of the problem, and introduce a method based on Ant Colony System to find a variety of efficient solutions. The performance achieved by the Ant Colony System is compared with that of Greedy Randomized Adaptive Search Procedure and Non-dominated Sorting Genetic Algorithm, by means of computational experiments carried out on two instances of the problem constructed from data provided by the experts.

Requirements interaction in the next release problem

The selection of a set of requirements between all those proposed by the customers is an important process in software development, that can be addressed using heuristic optimization techniques. Dependencies or interactions between requirements can be defined to denote common situations in software development: requirements that follow an order of precedence, requiments exclusive of each other, requirements that must be included at the same time, etc. This paper shows how requirements interactions affect the search space explored by optimization algorithms. Three search techniques, i.e. a greedy randomized adaptive search procedure (GRASP), a genetic algorithm (GA) and an ant colony system (ACS), have been adapted to the requirements selection problem considering interaction between requirements. We describe the adaptation of the three meta-heuristic algorithms to solve this problem and compare their performance.

Bayesian networks for enhancement of requirements engineering: a literature review

AbstractRequirements analysis is the software engineering stage that is closest to the users’ world. It also involves tasks that are knowledge intensive. Thus, the use of Bayesian networks (BNs) to model this knowledge would be a valuable aid. These probabilistic models could manage the imprecision and ambiguities usually present in requirements engineering (RE). In this work, we conduct a literature review focusing on where and how BNs are applied on subareas of RE in order to identify which gaps remain uncovered and which methods might engineers employ to incorporate this intelligent technique into their own requirements processes. The scarcity of identified studies (there are only 20) suggests that not all RE areas have been properly investigated in the literature. The evidence available for adopting BNs into RE is sufficiently mature yet the methods applied are not easily translatable to other topics. Nonetheless, there are enough studies supporting the applicability of synergistic cooperation between RE and BNs. This work provides a background for understanding the current state of research encompassing RE and BNs. Functional, non-functional and -ilities requirements artifacts are enhanced by the use of BNs. These models were obtained by interacting with experts or by learning from databases. The most common criticism from the point of view of BN experts is that the models lack validation, whereas requirements engineers point to the lack of a clear application method for BNs and the lack of tools for incorporating them as built-in help functions.

Olive Fly Infestation Prediction Using Machine Learning Techniques

This article reports on a study on olive-fly infestation prediction using machine learning techniques. . The purpose of the work was, on the one hand, to make accurate predictions and, on the other, to verify whether the Bayesian network techniques are competitive with respect to classification trees. We have applied the techniques to a dataset and, in addition, performed a previous phase of variables selection to simplify the complexity of the classifiers. The results of the experiments show that Bayesians networks produce valid predictors, although improved definition of dependencies and refinement of the variables selection methods are required.

REQUIREMENT RISK LEVEL FORECAST USING BAYESIAN NETWORKS CLASSIFIERS

Requirement engineering is a key issue in the development of a software project. Like any other development activity it is not without risks. This work is about the empirical study of risks of requirements by applying machine learning techniques, specifically Bayesian networks classifiers. We have defined several models to predict the risk level for a given requirement using three dataset that collect metrics taken from the requirement specifications of different projects. The classification accuracy of the Bayesian models obtained is evaluated and compared using several classification performance measures. The results of the experiments show that the Bayesians networks allow obtaining valid predictors. Specifically, a tree augmented network structure shows a competitive experimental performance in all datasets. Besides, the relations established between the variables collected to determine the level of risk in a requirement, match with those set by requirement engineers. We show that Bayesian networks are valid tools for the automation of risks assessment in requirement engineering.

Architecture for the use of synergies between knowledge engineering and requirements engineering

The application of Artificial Intelligence techniques in the processes of Software Engineering is achieving good results in those activities that require the use of expert knowledge. Within Software Engineering, the activities related to requirements become a suitable target for these techniques, since a good or bad execution of these tasks has a strong impact in the quality of the final software product. Hence, a tool to support the decision makers during these activities is highly desired. This work presents a three-layer architecture, which provides a seamless integration between Knowledge Engineering and Requirement Engineering. The architecture is instantiated into a CARE (Computer-Aided Engineering Requirement) tool that integrates some Artificial Intelligence techniques: Requisites, a Bayesian network used to validate the specification of the requirements of a project, and metaheuristic techniques (simulated annealing, genetic algorithm and an ant colony system) to the selection of the requirements that have to be included into the final software product.

Three steps multiobjective decision process for software release planning

This paper deals with how to determine which features should be included in the software to be developed. Metaheuristic techniques have been applied to this problem and can help software developers when they face contradictory goals. We show how the knowledge and experience of human experts can be enriched by these techniques, with the idea of obtaining a better requirements selection than that produced by expert judgment alone. This objective is achieved by embedding metaheuristics techniques into a requirements management tool that takes advantage of them during the execution of the development stages of any software development project.