Semantics based text matching for unstructured text documents

点击翻译按钮获取中文摘要

In this project our aim is to develop techniques that can find matches between job postings that are in natural language text and in descriptive format and job seeker profiles that are also in free-text format without much accessible structured content. The unstructured and natural language form of data poses challenges for the accurate matching and ranking in this domain. Research challenges include (i) vocabulary mismatch within the job postings and job seeker profiles, (ii) the processing of the interaction logs that are curated as a result of the visits that both job seekers and recruiters make on the website, and (iii) the possibility of processing interaction logs for the purpose of understanding the demand and trend in the job market for certain skill sets or qualifications. The research work and required development will be carried out in close concert with the industrial partner BC Jobs Online Inc.