To investigate the problem of classifying source code reviews, we have created a dataset suitable for evaluating and testing various methods for solving this problem. We combined four open datasets, and manually marked up 3200 code review comments. We have created our own classification based on the available datasets. The final dataset contains 10045 comments and is divided into 16 classes and 5 groups for hierarchical classification. The dataset contains comment classes such as:Style - readability, code layout, indentation issues, and other common programming conventionsNaming - uniform style and convenience of naming variables, methods, classesQuestioning - questions to the author of the code, requests clarification of the code or examples of useResponse - appointment of other reviewers, writes acknowledgements, agreements with others, additions to the developer's opinionConvention - discussion of the software development processTesting - requests tests to verify the functionality of the codeDesign - architecture and code design, program structure controlRefactoring - logical structure, object creation, logical errorsFunctionality - identification of code defectsRoadmap - further development of the programOptimization - code optimization, parallelism, synchronizationError - identifies problems with exception and error handlingDocumentation - problems with documentation or comments in the source codeSupport - compatibility with other systems, support systemsInput/Output - input/output in the graphical user interface, problems with pop-up windows.Other - comments that do not carry a semantic load without context Union of classes:Code style (Style, Naming)Discussion (Questioning, Response, Convention, Testing)Development (Design, Refactoring, Functionality, Roadmap, Optimization, Error)User (Documentation, Support, Input/Output)Other (Other)

To investigate the problem of classifying source code reviews, we have created a dataset suitable for evaluating and testing various methods for solving this problem. We combined four open datasets, and manually marked up 3200 code review comments. We have created our own classification based on the available datasets. The final dataset contains 10045 comments and is divided into 16 classes and 5 groups for hierarchical classification. The dataset contains comment classes such as:Style - readability, code layout, indentation issues, and other common programming conventionsNaming - uniform style and convenience of naming variables, methods, classesQuestioning - questions to the author of the code, requests clarification of the code or examples of useResponse - appointment of other reviewers, writes acknowledgements, agreements with others, additions to the developer's opinionConvention - discussion of the software development processTesting - requests tests to verify the functionality of the codeDesign - architecture and code design, program structure controlRefactoring - logical structure, object creation, logical errorsFunctionality - identification of code defectsRoadmap - further development of the programOptimization - code optimization, parallelism, synchronizationError - identifies problems with exception and error handlingDocumentation - problems with documentation or comments in the source codeSupport - compatibility with other systems, support systemsInput/Output - input/output in the graphical user interface, problems with pop-up windows.Other - comments that do not carry a semantic load without context Union of classes:Code style (Style, Naming)Discussion (Questioning, Response, Convention, Testing)Development (Design, Refactoring, Functionality, Roadmap, Optimization, Error)User (Documentation, Support, Input/Output)Other (Other)