Abstract:
"
Anti-patterns were conceived in small talk circles until recently. The impact of anti-pattern
occurrences has caused many problems in human history, making engineers look more into
these. Antipatterns which refer to specific design violations or implementation styles can tell the
developers whether a design choice is “poor” or not. Poor designs can be fixed by refactoring.
Detecting anti-patterns and refactoring is integral to development. If proper detection is not done
properly, it can set back days, even weeks, and refactoring becomes riskier. Anti-pattern
detection models are often used to help allocate software quality assurance efforts.
The Anti-Pattern Detection analyzer detects two types of Anti-Patterns names Long Method and
Speculative Generality. To detect the occurrences, the user given repository is scanned a dataset
is created for the respective repository. Then the created dataset is saved and analyzed via big
data analytics map-reduce methods and pattern matching algorithms. The final result is emailed
to the user.
The project has been tested on open-source Java projects where have been examined and scanned
for the detection of Anti-Patterns. The Anti-Pattern Tool and the results from this evaluation
open a good approach to the domain area for the detection of patterns."
