Decision tree based project classification using activity diagram

Ace p.

Order the writing of a tailor-made Computer science Term papers

Free quote online

Thesis Format .pdf

Decision tree based project classification using activity diagram

Download

Read an extract

Reader
Abstract
Contents
Extract

Abstract

Several projects of different categories start every year by various companies and each project goes through the Software Development Life Cycle (SDLC). Even though SDLC is followed by all companies, many projects do not complete. The reasons are lack of technical skills, imprecise project estimation and inaccurate requirements, to name the few. In order to better manage projects and to produce higher quality software systems that can delivered on time and on budget, software engineers explore the potential of software engineering data using data mining techniques like classification, association rule and clustering. Data mining principles can also be applied on software engineering data to monitor, evaluate and identify project status.

Abstract
Introduction
Proposed architecture
1. Pre-processing
2. Feature extraction
3. Relevance analysis
4. Data transformation
5. Decision tree induction
6. Project classification
Conclusion
References

Get this table of contents for free after login.

Extract

[...] Information gain measure is used to select the test attribute at each node in the tree Fig.6 Input Dataset for Decision Tree Induction Results obtained at the end of first iteration is based on the below given values of the gain and shown in Fig Once continuous attributes are categorized, actual record is transformed into a form suitable for next phase. For sample activity diagram, as per the categorization shown in Table II, Invocation node falls into ‘Medium' category, Branch and Fork nodes matches with category ‘low' and edges go with ‘High' category. [...]

[...] Hassan, Tao Xie Mining API Patterns as Partial Orders from Source Code: From Usage Scenarios to Specifications, Mithun Acharya, Tao Xie, Jian Pei, Jun Xu. ESEC/FSE 2007 Mining Email Social Networks. Christian Bird, Alex Gourley, Prem Devanbu, Michael Gertz, and Anand Swaminathan. MSR 2006 Mining Version Histories to Guide Software Changes. Thomas Zimmermann, Peter Weibgerber, Stephan Diehl, and Andreas Zeller. ICSE 2004 Predicting Source Code Changes by Mining Change History. Annie Ying, Gail Murphy, Raymond Ng, and Mark Chu-Carroll. TSE 2004 Data Mining Concepts and Techniques, Jiawei Han & Micheline Kamber, Elsevier Science India. pp.284-285. Fig.8 Final Decision Tree VI. Project Classification Once each [...]

[...] Details visible in Fig of extracted features for a sample activity diagram are summarized in Table I. Mainly, complexity of Tag ‘< edge >' contains two other attributes other the activity diagram is dependent on fork, branch then ‘id'. As every edge of the Activity Diagram is and invocation nodes. Complexity is also affected directed, attributes like ‘source' and ‘destination' by the number of edges as they indicate various are also required. Edge tag also contains < peid > possible paths. tag. It is also clear from the snapshot that file does not provide detail of the type of nodes i.e. [...]

[...] Once software engineering data (such as code bases, each activity diagram for a given project is execution traces, historical code changes, mailing classified into one of the two classes, project is also lists, and bug databases which contains huge classified. information about a project's status, progress, and evolution) by using data mining techniques like I. Pre-Processing classification, association rule and clustering in order to better manage projects and to produce An Activity diagram for a project is shown in Fig. [...]

pdf