Classification and visualization on eligibility rate of applicant’s LinkedIn account using Naïve Bayes / Nurul Atirah Ahmad

The recruitment process is crucial in organizations, as it involves selecting qualified applicants from a diverse pool. In the digital era, social media platforms like LinkedIn have become famous for recruitment, and recruiters widely use them to find potential employees. Selecting suitable applicants is essential for an organization’s success. However, recruiting unqualified applicants may affect the organizations. These can happen because the recruitment process requires much time, high costs, and possibly biased results when done manually. Thus, this project aims to classify and generate a list of potential job applicants by analyzing various attributes of their LinkedIn accounts, such as title, location, skills, education, language, certification, and experience. This project implements the Naive Bayes algorithm as the classification algorithm. The collected data from LinkedIn profiles then undergoes data preprocessing. It is set to label since it has no label class. The classification is set to two categories: Eligible or Ineligible. The dataset is split into training and testing sets to develop the Naive Bayes (NB) model. The NB model achieved a commendable accuracy of 89.8%, indicating good performance in classifying potential job applicants. The system’s functionality based on the use case and usability tested by the Human Resources expert has been tested to evaluate its system requirements. The usability testing yields a score of 80% which indicates that the system is acceptable. The classification results are visualized, allowing users to identify eligible applicants efficiently. Thus, this project can help users find suitable applicants for the job. Future research could also focus on expanding the criteria considered for classification, such as cultural fit and behavioural attributes.