E-recruiting support system based on text mining methods

摘要

Since web documents have different formats and contents; it is necessary for various documents to use standards to normalise their modelling in order to facilitate retrieval task. The model must take into consideration, both the syntactic structure, and the semantic content of the documents. Curriculum vitae (CV) is the document that summaries our education, skills, accomplishments, and experience. Job seekers submit their CV via the web. Therefore, in their recruitment process, companies arc requiring systems for extraction and analysis of information from CVs: identifying specific patterns, which meet with certain profile. To extract the essential component of CVs and to relate them with user's require, needs first, a study of their most significant elements and a better understanding of the CV feature. This work focuses on CVs' analysis. It introduces an approach for analysing and structuring CVs which are in French. To this end, we make an extension of General Architecture of Text Engineering (GATE). The extension affects essentially a formulation of logic rules for the generation of annotations used for CV handling. The goal is to normalise the CV content according to the structure adopted by Europass CV. This action is guided by the MR-XML standard. We experiment the proposed process and we showed that there is an improvement in the extraction phase.