• Share
  • Email
  • Embed
  • Like
  • Save
  • Private Content
Elane - Promise08
 

Elane - Promise08

on

  • 870 views

Comparing Negative Binomial and Recursive Partitioning Models for Fault Prediction

Comparing Negative Binomial and Recursive Partitioning Models for Fault Prediction

Statistics

Views

Total Views
870
Views on SlideShare
739
Embed Views
131

Actions

Likes
0
Downloads
2
Comments
0

1 Embed 131

http://promisedata.org 131

Accessibility

Upload Details

Uploaded via as Microsoft PowerPoint

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    Elane - Promise08 Elane - Promise08 Presentation Transcript

    • Comparing Negative Binomial and Recursive Partitioning Models for Fault Prediction Elaine Weyuker Thomas Ostrand Robert Bell AT&T Labs - Research
    • High Level Goal To determine which files of a large industrial software system with multiple releases are particularly likely to be fault-prone.
    • Why is this Important?
      • Help testers prioritize testing efforts.
      • Help developers decide what to rearchitect.
      • Help verifiers decide what to verify.
    • APPROACH
      • Identify properties that are likely to affect fault-proneness, and then build a statistical model to make predictions. In the past we’ve used a Negative Binomial Regression Model.
    • Past Systems Studied 75% 2.25 years Voice Resp 83% 2 years Provisioning 83% 4 years Inventory 20% Files Period Covered System Type
    • Can We Do Better?
      • Compare results for three different systems making predictions using the negative binomial regression model and recursive partitioning.
    • Recursive Partitioning System A, Releases 1-26, cp = 0.01
    •  
    •  
    •  
    • Percent Faults in 20% Files 67.9% 76.1% System C 84.8% 93.4% System B 76.1% 80.5% System A RP NBR