|
|
2019 » Papers » Volume 3 » A New Method to Help the Human Resources Staff to Find the Right Candidates, Based on Deep Learning 1. A NEW METHOD TO HELP THE HUMAN RESOURCES STAFF TO FIND THE RIGHT CANDIDATES, BASED ON DEEP LEARNING Authors: Vasilescu Cristian, Suciu George, Pasat Adrian Volume 3 | DOI: 10.12753/2066-026X-19-170 | Pages: 240-246 | Download PDF | Abstract
The paper presents a new way of finding the internship candidates'opinion about the company they intend to work in. Candidates go to the company, and Human Resources staff outlines the company's main areas of activity. Afterwards, candidates are asked to express their views/opinions on each field of activity in a few sentences. These sentences are automatically analyzed using Artificial Intelligence (AI), namely Deep Learning (DL). In this case, the necessary analysis is the "Sentiment Analysis" which indicates how much the candidate likes or not to work in each field of activity, as a score in the range from very negative to very positive. This score is used by human resources staff to choose the right candidate for the right job matching their knowledge and attitude towards the job. Furthermore, personalized training programs are suggested to the candidate after the recruitment in order to accelerate their induction to the new work place and the impact of the training on the performance of the new employees is measured.
Therefore, we analyze several different versions of sentiment analysis algorithms based on DL, from different sources: Stanford University, Google, Microsoft and IBM, NLTK, OpenAI, Apache Open NLP, LexisNexis, Amazon. Some of these algorithms are open source and others are proprietary, but accessible using web services from the cloud and we have chosen the most appropriate.
The novelty of this paper is the method of objectively automatically finding the candidates' opinion about the company where they intend to work using the most appropriate algorithm for sentiment analysis which is based on DL. | Keywords
Artificial Intelligence; Deep Learning; Sentiment Analysis; Training; Recruiting. |
|
|
|