Evaluation of unstructured resumes using the Word2Vec model

The object of the study is the Word2Vec natural language processing model. The article presents an overview of the main principles on which this model is based and conducts a comparative experiment to evaluate its effectiveness in the context of unstructured resumes. The aim of the work is to improve the efficiency and accuracy of automated job candidate selection systems. It is proposed to use the Word2Vec model, which, unlike traditional methods such as TF-IDF, is capable of considering semantic relationships between words. This allows the system to more accurately assess candidates by taking into account not only direct skill matches but also synonyms and related competencies, thereby increasing the overall effectiveness of the selection process. Ref. 8, pic. 2, tabl. 2