Abstract
Understanding the skill sets and knowledge required for any career is of utmost importance, but it is increasingly becoming harder in today’s world. This is caused by rapid innovations in the tools used and the nature of the industry, making it harder to catch up. Due to the conflicting situation, there is a need to produce relevant career insights on the skill sets and knowledge based on the latest data. This paper proposed the use of a data-driven recommendation engine approach to extract and generate insights. The dataset used will be the job descriptions published by companies for hiring, attributing to the ground truth. Keywords related to skill sets will be extracted from the job description, passed through a data cleaning pipeline before feeding it to a model to learn the relationship. Modelling techniques such as a frequency-based bag of words and word embedding using FastText and BERT will be explored further for this task. These techniques are driven by how skill sets are analyzed manually but at a higher frequency and larger dataset. In the business world, this problem presents a unique market opportunity in the career industry. With the proposed data-driven recommendation engine, businesses can seek to venture and provide the service either to companies (B2B) or the public (B2C). Different business models such as Software as a Service (Saas) and API as a Service (Aaas) are possible routes the business can take. With the huge potential present in the market as of now, this industry provides a prime opportunity for businesses to jump in.