How to identify class comment types? A multi-language approach for class comment classification

Most software maintenance and evolution tasks require developers to understand the source code of their systems. Software usually inspect class comments gain knowledge about program behavior, regardless programming language they are using. Unfortunately, (i) different languages present language-specific commenting notations/guidelines; (ii) projects often lacks that adequately describe which complicates comprehension activities. To handle these challenges, this paper investigates practices three languages: Python, Java, Smalltalk. In particular, we systematically analyze similarities differences information types found in developed languages. We propose an approach leverages two techniques, namely Natural Language Processing Text Analysis, automatically identify various from i.e., specific semantic comments. best our knowledge, no previous work has provided a comprehensive taxonomy comment for with help common automated approach. Our results confirm can classify frequent high accuracy Smalltalk believe monitor assess quality languages, thus support tasks.