Improved Resume Parsing based on Contextual Meaning Extraction using BERT
V. V. Satyanarayana Tallapragada, Vasundhara Raj, U Deepak, PagadalaManoj Sai, Trishul Mallikarjuna
Abstract
With the advent of technological advancement recruitment process is almost online. Most companies will first identify resumes from the received resumes online based on the required qualifications. This, in turn, requires resume culling based on the word-matching technique. It is hard for recruiters to identify the perfect resume, as the existing techniques utilized for classification are purely based on the words written in the resume. Hence, it is necessary to understand the given document contextually. This study exploits this concept of understanding the document contextually. Hence, it is proposed to use Bidirectional Encoder Representations from Transformer (BERT) vectorization to identify the text contextually. Results show that the proposed algorithm parses the resume better than the existing techniques.