superhen/Automated-Job-Resume-Matching-Solution

According to a 2015 study on job seeking behavior by Pew Research Center, 79% of the job seekers utilized the online resources for their most recent employment (Aaron ,2015). This study result suggests that the online job boards become the major channel for job seekers in the digital era. However, another finding in the study indicates that most of the job seekers fail to match their experiences with the job requirements and spend hours on job board to apply job which is not seen to be suitable (Aaron, 2015). Additionally, Dr. John Sullivan conducted a similar research in 2013 which highlighted some interesting aspects: on average, 250 resumes are received for each job opening by the major organizations, more than 50% of the resumes does not meet the minimum requirement (John, 2013). This means the time our recruiter spends on these 50% of the resumes for each job is wasted. From both candidate and recruiter’s points of view, the phenomenon may suggest that the traditional online job board does not seem to simplify the job application process or reduce the effort required from both parties. With this challenge getting bigger and bigger, the demand to automate the resume - job matching process is getting increased as well. For instance, the content - based recommendation system (CBR) is introduced to analyze the job description to identify the potential area of interest to the job seekers (Shiqiang et al., 2016). To apply the concept in Singapore local context, our team has conducted a text mining project based on the data acquired from the major online job board in Singapore. The primary objective of this project is to create a machine learning model to accelerate the job - resume matching process. The detail of the text mining methodology and results are presented in the following sections.

Jupyter NotebookMIT