Published Paper Details:

CV/Resumes taken from different professions use different formats, fonts, styling and structuring and vocabulary. We are focusing the concept of Automated Recruitment process and the mail logic behind it is to provide ease to Recruitment Process with the help of Technology. In this paper we are using concept to Text-Mining, Pdf-Mining, tokenization, parts of speech and many other steps which would require to build algorithm for our thesis. Tokenization is a key step in the process of information extraction from the given input CVs, the tokenizer module of natural language processing libraries are compared with regard to per- formance, throughput and applicability. At the later stages the dissertation will look into the inner workings of open-source parsers, measures their precision and recall, comparing the result to the previous CVs. The performance of three commercial resume parsers were also evaluated. Lastly the steps of an iterative development are discussed, by which a new resume parser was developed, where the emphasis was on using the resume�s structural and visual information, rather than relying only on its raw text.We will evaluate the performance of newly developed resume parser is measured, and compare its output to another set of pre-annotated resumes. Each step of the parsing process will be examined closely and recommendations will made for future researches.