Artificial intelligence is one of the most rapidly evolving technologies today, and it has been used in a variety of application domains. Recently, AI technology has shone in the field of recruiting. Many researchers are working on expanding its capabilities with various applications that will aid in the recruitment process. However, importantly, having an open labour market without a cohesive data centre makes it difficult to monitor, integrate, analyse, and build an evaluation matrix that helps reach the best match of a job candidate to a job vacancy. Moreover, previous research in this domain, where AI serves in the recruiting field, has focused solely on linear and non-linear models that, once trained, lack user preference personalisation. Furthermore, it was more focused on commercial applications than on government services.
This paper implements the AIRM architecture and found impressive results by exploring AI and Natural Language Processing matching job candidates. We used a suitable data repository technique with three processing layers, each with a different appropriate model. First, the Initial Screening layer usesthe BIRCH clustering algorithm. The Mapping layer then performs an approximate nearest neighbour search using a Sentence transformer and ranking with a Facebook AI Similarity Search
Finally, the Preferences layer takes the user's preferences as a list and sorts the results using a pretrained Cross-Encoders model that considers the weight of the more important words. We completed the implementation of AIRM and obtained some promising results. We discovered that when at least one expert agrees with the system's choice, the AIRM achieves an overall matching accuracy of 82%. In the time performance test, AIRM outperforms human performance in terms of task execution time. It completed the task in 2.4 minutes, but humans took more than three days on average, which is essential when pre-selecting a set of candidates and positions. AIRM will empower the Saudi government's immediate and strategic decisions by gaining comprehensive insight into the labour market and speeding up the recruiting process.