Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...
Research proposal
1. Research Topics: Automatic Bug Localization from bug reports
Problem statement: In software engineering, quality of the software must be ensured. To
provide good software quality, the first and foremost condition is to build up bug free software.
A software bug is an error, flaw, failure, or fault in a computer program or system that causes it
to produce an incorrect or unexpected result. Most bugs arise from mistakes and errors made by
people in either a program's source code or its design. To make bug free software is not as easy
as it hears. In tradition and mostly current system follows a way in where a quality assurance
team will continuously check the errors by running the program manually. After finding the
errors of any unexpected behavior, the QA team will provide bug report which contains normally
bug title, bug description, etc. When developer team observes the bug report, the team leader will
assign bugs to developers. For fixing the bugs, the developer at first needs to find out the
problematic location of the bug. For finding the location of bug is much more time consuming.
The developer attaches the debug pointers, and then sequentially debugs the program. If the
developer can easily find out the location of the problematic area, it would reduce the huge time
and cost of the project.
Motivation: In my internship company “Orion informatics Ltd”, I solved lots of bugs of the real
life project of banking system. At first quality assurance team provides bug reports. After the bug
reports, my team leader assigned bugs on me. During the bug fixing time, I had to do lot of work
and from that i understood that for fixing a bug only 5% time is required for proving logic and
other 95% time is needed for locating the bug. So, if the bug location can be automatically find
out, the development time will be very faster. As a result, the cost can be reduced significantly.
Benefit: Faster development is possible and the hassle of the developers can be reduced. The cost
of the project can be dramatically minimized.
Solution Direction: Form bug report at first through tokenization “bag of words” can be made.
From that “bag of word”, stop words will be reduced and by calculating conditional probability
among the each remained words with each program file may provide some probability. Then
based on the probability values descending order affected file will be suggested to the developer
where bug may exist.
the highest probable file will keep at the top
Konpositionaerror hoiche + konvalur zonnohoyechhe ai dui tadekhate parle developerkhubsohoze
solv Krte parto