A study of mutual information based feature selection for case based reasoning in software cost estimation

Li, Yan‐Fu; Xie, Min; Goh, T.N.

doi:10.1016/j.eswa.2008.07.062

Cited by 94 publications

(71 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the case of effort estimation, it has been reported that reduced datasets improve the estimation accuracy [29,9,34,8]. It is known, however, that feature selection algorithms do not perform well with imbalanced datasets, resulting in a selection of metrics that cannot be adequate for the learning algorithms, decreasing the quality and usefulness of the rules.…”

Section: Reduction Of Imbalanced Datasetsmentioning

confidence: 99%

Searching for rules to detect defective modules: A subgroup discovery approach

Rodríguez

Ruíz

Riquelme

et al. 2012

Information Sciences

View full text Add to dashboard Cite

Data mining methods in software engineering are becoming increasingly important as they can support several aspects of the software development life-cycle such as quality. In this work, we present a data mining approach to induce rules extracted from static software metrics characterising fault-prone modules. Due to the special characteristics of the defect prediction data (imbalanced, inconsistency, redundancy) not all classification algorithms are capable of dealing with this task conveniently. To deal with these problems, Subgroup Discovery (SD) algorithms can be used to find groups of statistically different data given a property of interest. We propose EDER-SD (Evolutionary Decision Rules for Subgroup Discovery), a SD algorithm based on evolutionary computation that induces rules describing only faultprone modules. The rules are a well-known model representation that can be easily understood and applied by project managers and quality engineers. Thus, rules can help them to develop software systems that can be justifiably trusted. Contrary to other approaches in SD, our algorithm has the advantage of working with continuous variables as the conditions of the rules are defined using intervals. We describe the rules obtained by applying our algorithm to seven publicly available datasets from the PROMISE repository * Corresponding author Email addresses: daniel.rodriguezg@uah.es (D. Rodríguez), robertoruiz@upo.es (R. Ruiz), riquelme@lsi.us.es (J.C. Riquelme), aguilar@upo.es (J.S. Aguilar-Ruiz) February 4, 2011 showing that they are capable of characterising subgroups of fault-prone modules. We also compare our results with three other well known SD algorithms and the EDER-SD algorithm performs well in most cases. Preprint submitted to Information Sciences

show abstract

Section: Reduction Of Imbalanced Datasetsmentioning

confidence: 99%

Searching for rules to detect defective modules: A subgroup discovery approach

Rodríguez

Ruíz

Riquelme

et al. 2012

Information Sciences

View full text Add to dashboard Cite

show abstract

“…AI and data mining techniques help in this filed [93,94]. The selection, mining and extraction of relevant features for case representation and weights for these features are open problems [95,96]. The problem becoming complicated in the recent medical CBR systems due to a complex data format where the data are coming from sensors, images, time series or free-text format.…”

Section: ) Case Solution Adaptation Has Many Techniques Range From Mmentioning

confidence: 99%

Case Based Reasoning: Case Representation Methodologies

El–Sappagh¹,

Elmogy²

2015

ijacsa

View full text Add to dashboard Cite

Abstract-Case Based Reasoning (CBR) is an important technique in artificial intelligence, which has been applied to various kinds of problems in a wide range of domains. Selecting case representation formalism is critical for the proper operation of the overall CBR system. In this paper, we survey and evaluate all of the existing case representation methodologies. Moreover, the case retrieval and future challenges for effective CBR are explained. Case representation methods are grouped in to knowledge-intensive approaches and traditional approaches. The first group overweight the second one. The first methods depend on ontology and enhance all CBR processes including case representation, retrieval, storage, and adaptation. By using a proposed set of qualitative metrics, the existing methods based on ontology for case representation are studied and evaluated in details. All these systems have limitations. No approach exceeds 53% of the specified metrics. The results of the survey explain the current limitations of CBR systems. It shows that ontology usage in case representation needs improvements to achieve semantic representation and semantic retrieval in CBR system.

show abstract

“…missing values problem). Feature set selection is another major research subject in cost estimation where some papers tried to evaluate the minimum set of attributes that can best predict the effort estimation [6,7].…”

Section: Related Workmentioning

confidence: 99%

Evaluation of Cost Estimation Metrics: Towards a Unified Terminology

Alsmadi

Nuser

2013

CIT

View full text Add to dashboard Cite

Cost overrun of software projects is major cause of their failures. In order to facilitate accurate software cost estimation, there are several metrics, tools and datasets. In this paper, we evaluate and compare different metrics and datasets in terms of similarities and differences of involved software attributes. These metrics forecast project cost estimations based on different software attributes. Some of these metrics are public and standard while others are only employed in a particular metric tool/dataset.Sixteen public cost estimation datasets are collected and analyzed. Different perspectives are used to compare and classify those datasets. Tools for feature selection and classification are used to find the most important attributes in cost estimation datasets toward the goal of effort prediction. In order to have better estimation, it is needed to correlate cost estimation from different resources, which requires a unified standard for software cost estimation metric tools and datasets. It is pertinent that a common cost estimation model may not work for each project due to diverse project size, application areas etc. We suggest having a standardized terminology of project attributes used for cost estimation. This would improve cost estimation as multiple metrics could be applied on a project without much additional effort.

show abstract

A study of mutual information based feature selection for case based reasoning in software cost estimation

Cited by 94 publications

References 35 publications

Searching for rules to detect defective modules: A subgroup discovery approach

Searching for rules to detect defective modules: A subgroup discovery approach

Case Based Reasoning: Case Representation Methodologies

Evaluation of Cost Estimation Metrics: Towards a Unified Terminology

Contact Info

Product

Resources

About