Problem Statement:
Our client, a referral-based recruitment portal, was facing suboptimal results from their current ranking algorithm that matched candidates’ skills with job descriptions. They wanted to improve their algorithm to better rank candidates for job applications.
Solution:
To improve the ranking algorithm, we developed spiders to collect data from multiple sites on the Internet to quantify the quality of candidate skills and characterize companies and educational institutes to predict the cultural match between the candidates’ employment history and prospective employers. We also extracted information from the spidered text to develop an influence network to measure the “network worth” of a referrer using social network analysis. We used Genetic Algorithms to optimize a hybrid model using skills, cultural match, and referrer quality. Techniques, Technologies, and Tools used in this solution were: Text analysis, topic models, genetic algorithm, conditional random fields, R, jsoup.
Benefits:
The new algorithm resulted in a significant reduction in the number of candidates that needed to be interviewed, saving a lot of time and money for the client. The inclusion of cultural match and referrer quality factors in the algorithm improved the accuracy of the candidate rankings, leading to better hiring decisions. The client was able to provide better value to both job seekers and employers, ultimately leading to a better reputation and increased business.
