Within the human resources Industry there are still many challenges ahead, by example, keyword based searches or manual resume screening could contribute to the omission of valuable opportunities for all the parties involved or to lengthy hiring procedures. Therefore, we would like to present a Recommender System in the field of HR, which comprises a series of Natural Language Processing techniques and Deep Learning models, that allow achieving a fully automated process which will propose semantically related and explainable suggestions to job seekers and companies in real time. The described system is running in production at Janzz.technology where it's used to match candidates and job seekers on a daily-basis. The architecture of the system combines several NLP techniques like Named Entity Recognition, Text Classification, and Entity Relationships based on HR data collected by Domain Specialized Curators and it also includes Language Detection, specific Text Preprocessing and Language Models pretrained over domain specific data, which allows to extracts and processes more than 50 different characteristics from job postings and resumes such as occupations, skills, education, etc., with various levels of granularity across multiple languages.

Download the slides for this talk.Download ( PDF, 8796.33 MB)