Fault Prediction in Object Oriented System projects main aim is to capture different aspects of cohesion in object oriented software. In software system development reusability and maintainability are important facts, in order to maintain these factors there should be high cohesion. In present system attributes and references are two methods used for calculating cohesion. . Comments and Identifiers are two unstructured information calculated from source code.

This unstructured information is integrated in to source code. Latent semantic indexing is the technique used for extracting unstructured information from source code. This application is tested on three open source software systems and compares different features with the existing system and provides solution to develop new models which will predict software faults.

Hardware Requirements:

    • PROCESSOR                                  :     PENTIUM III 866 MHz
    • RAM                                   :       128 MB DD RAM
    • MONITOR                          :       15” COLOR
    • HARD DISK                       :       20 GB
    • FLOPPY DRIVE                 :       1.44 MB
    • CDDRIVE                          :       LG 52X
    • KEYBOARD                                  :       STANDARD 102 KEYS
    • MOUSE                              :       3 BUTTONS 

Software Requirements:           

  • LANGUAGE                   :    JAVA
  • FRONT-END TOOL      :    SWING