The growing popularity of automated processes, empowered by Artificial Intelligence (AI), is changing how companies offer and fill their jobs. The growth of AI has created the need for a more sophisticated way to find a job, through the use of CRM-based applications. Most job search sites offer simple keyword matching (keyword searches) with little or no customization, resulting in a low accuracy rate. This paper presents a new type of Job Recommendation Model, based on the use of Hybrid AI models. The purpose of this new hybrid model is to improve job matching through three separate processes: skill-based filtering, sentiment analysis, and predictive modeling. The proposed hybrid model will take user calculated skill sets into account for the job description and will use AI adoption levels, automation risk, expected salary, and anticipated growth opportunity for the job type. The hybrid model will produce two scores for each job type; the first score represents the potential to match with the specific job type based on the applicant’s skills and the second score represent the applicants’ overall profile. Once combined to create a matching score, the job recommendation system will recommend jobs based upon applicant's (1) Skills, (2) Ability to do the job, (3) Salary Range, and (4) Job Growth Potential (area of interest on the applicant).
Dutta et al. (Mon,) studied this question.