Showing papers by "Daniel Rodriguez published in 2007"

Detecting Fault Modules Applying Feature Selection to Classifiers

Abstract: At present, automated data collection tools allow us to collect large amounts of information, not without associated problems. This paper, we apply feature selection to several software engineering databases selecting attributes with the final aim that project managers can have a better global vision of the data they manage. In this paper, we make use of attribute selection techniques in different datasets publicly available (PROMISE repository), and different data mining algorithms for classification to defect faulty modules. The results show that in general, smaller datasets with less attributes maintain or improve the prediction capability with less attributes than the original datasets.

Attribute Selection in Software Engineering Datasets for Detecting Fault Modules

Abstract: Decision making has been traditionally based on managers experience. At present, there is a number of software engineering (SE) repositories, and furthermore, automated data collection tools allow managers to collect large amounts of information, not without associated problems. On the one hand, such a large amount of information can overload project managers. On the other hand, problems found in generic project databases, where the data is collected from different organizations, is the large disparity of its instances. In this paper, we characterize several software engineering databases selecting attributes with the final aim that project managers can have a better global vision of the data they manage. In this paper, we make use of different data mining algorithms to select attributes from the different datasets publicly available (PROMISE repository), and then, use different classifiers to defect faulty modules. The results show that in general, the smaller datasets maintain the prediction capability with a lower number of attributes than the original datasets.

Software project effort estimation based on multiple parametric models generated through data clustering

Abstract: Parametric software effort estimation models usually consists of only a single mathematical relationship. With the advent of software repositories containing data from heterogeneous projects, these types of models suffer from poor adjustment and predictive accuracy. One possible way to alleviate this problem is the use of a set of mathematical equations obtained through dividing of the historical project datasets according to different parameters into subdatasets called partitions. In turn, partitions are divided into clusters that serve as a tool for more accurate models. In this paper, we describe the process, tool and results of such approach through a case study using a publicly available repository, ISBSG. Results suggest the adequacy of the technique as an extension of existing single-expression models without making the estimation process much more complex that uses a single estimation model. A tool to support the process is also presented.

Convertibility between IFPUG and COSMIC functional size measurements

Abstract: Since 1984 the International Function Point Users Group (IFPUG) has produced and maintained a set of standards and technical documents about a functional size measurement methods, known as IFPUG, based on Albrecht Fuction Points. On the other hand, in 1998, the Common Software Measurement International Consortium (COSMIC) proposed an improved measurement method known as Full Function Points (FFP). Both the IFPUG and the COSMIC methods both measure functional size of software, but produce different results. In this paper, we propose a model to convert functional size measures obtained with the IFPUG method to the corresponding COSMIC measures. We also present the validation of the model using 33 software projects measured with both methods. This approach may be beneficial to companies using both methods or migrating to COSMIC such that past data in IFPUG can be considered for future estimates using COSMIC and as a validation procedure.

Comparación de diferentes algoritmos de clustering en laestimación de coste en el desarrollo de software

Abstract: The software cost estimation models that obtain an only mathematical relation between the effort and some other attribute characteristic of the software projects, provide good results when the projects data base, from which by means of regression methods the mentioned relation is obtained previously, is formed by homogenous projects. Nevertheless, for projects data bases coming from very diverse sources, such as the ISBSG projects data base formed by thousands of heterogeneous projects, using an only mathematical relation to represent all these projects, does not offer so good results as if the projects were homogenous. In this work one considers, like improvement of the estimation process, to segment the ISBSG data base in different groups from projects by means of the use of three different clustering algorithms: COBWEB, EM, and k-means, so that for each one of these clusters (formed by homogenous projects to each other) a different mathematical relation is obtained. The carried out segmentation by these algorithms improves the estimation with respect to the model that uses the data base without segmenting. On the other hand if the

Defining a Legal Risk Management Strategy: Process, Legal Risk and Lifecycle

Abstract: All systems during their lifecycle, no matter how simple, will generate legal implications that need to be managed. The potential cost of an inadequate management of legal aspects can even imply the failure of the project. As a consequence, legal risk management should not only be a major activity of the development lifecycle, but it needs to be performed by qualified personnel following well-defined procedures and standards. However, current software process improvement models do not properly include processes for legal audits and more concretely legal risks management for each phase of the software development lifecycle. Neither in industry related to manage legal risks of software projects is possible to find well-defined and standardised projects. This lack of standardised process means that legal risks are handled reactively instead of proactively. This work presents a process for managing legal risks. It is organised by a series of activities to be performed at each stage of the software development lifecycle to eliminate or minimize the risk of project failures for legal reasons.

SLA: A legal assurance process model for software engineering management

Abstract: The legal assurance activities and measures are a key element for the viability of information systems projects because nowadays there can arise legal risks in some cases, which can be a serious threat for project commercial and financial success. In spite of this, there does not exist in the main evaluation and improvement processes models a process of legal assurance that systematizes and orders the activities and measures precisely by to manage such legal risks. On the other hand, the professional practice does not generally incorporate standardized processes in order to discipline the legal assurance activities and measures. This circumstance can generate the appearance of deficits in the project's legal security. This work proposes to consider the legal assurance activities and measures as a process to implement more in the evaluation and improvement processes models, with the objective to provide a suitable instrument for the management of inherent legal risks to any information systems project. This concept of the legal assurance activities and measures as a process allows the exceeding of the present reactivity characteristic of the effective professional practice and elevates it to proactive management, suitable for avoiding the legal risks that can threaten the project. Copyright © 2007 John Wiley & Sons, Ltd.

Epistemological and Ontological Representation in Software Engineering

Abstract: This paper provides an overview of how empirical research can be a valid approach to improve epistemological foundations and ontological representations in Software Engineering (SE). Despite of all the research done in SE, most of the results have not been yet been stated as laws, theories, hypothesis or conjectures, i.e., from an epistemological point of view. This paper explores such facts and advocates that the use of empirical methods can help to improve this situation. Furthermore, it is also imperative for SE experiments to be planned and executed properly in order to be valid epistemologically. Finally, this paper presents some epistemological and ontological results obtained from empirical research in SE.

Suggestions for Improving Measurement Plans: A BMP application in Spain

Abstract: Time and Cost are most often in industry the two main (often solely) dimensions of analysis against which a project is monitored and controlled, excluding other possible dimensions such as Quality, Risks, impact on society and Stakeholders’ viewpoints in a broader sense. Another issue of interest is the proper amount of measures and indicators to implement in an organiseation to optimising the two sides of the cost of quality (COQ - cost of quality - and CONQ - cost of non quality). How can multiple concurrent control mechanisms across several dimensions of analysis be balanced? The approach of Balancing Multiple Perspectives (BMP) has been designed to help project managers choose a set of project indicators from several concurrent viewpoints. After gathering experiences from Canada, Germany and Turkey, this paper presents the results from a new BMP application in Spain, using a list of 14 candidate measures interviewing a double set of respondents from industry. Lessons learned are presented for improving measurement plans.

Personal metadata for cataloguing microcontent – linking to large ontologies?

Abstract: Microlearning objects can be related by means of freely created annotations or tagging. Although such solution can help with filtering and searching, it is not enough for formalizing microlearning towards the semantic Web approach. In this paper, we propose the use of upper ontologies and more concretely, OpenCyc, to provide the required formal semantics needed by the semantic Web.

Continuous renal replacement therapy with a pulsatile tubular pump: parameters must be adjusted in each case.

Abstract: A three-valve, pulsatile tubular pump was used in 24 pigs weighing 10.2 ± 3.2 kg; the pump was connected to a neonatal hemofiltration circuit. Ninety-two periods of 30 min were studied to analyze the efficacy of the system with variations in the percentage time in diastole, the diastolic speed, the systolic speed, and the percentage time in systole during which the postfilter valve was closed. System efficacy was determined by the blood flow through the filter, the ultrafiltrate volume, the vascular overload measured by the inlet aspiration pressure, and the filter overload measured by the cross-filter pressure drop and the transmembrane pressure. The variations in the pump parameters did not lead to significant differences in the efficacy of the system or in the vascular or filter overload. The parameters must be adjusted in each case to obtain the best yield with the lowest vascular and filter overload.