Publication Date: 2023/11/23
Abstract: Online job portals have rapidly expanded, making it simpler for job searchers to find employment. But it can take much work for job searchers to find the ideal position that matches their skills and preferences due to the abundance of job postings. To solve this issue, the author present a system for recommending relevant job listings to students using machine learning and natural language processing techniques. There has never been any prior interaction between user data and job listing data in the dataset collected for our research. The system employs a hybrid strategy to generate precise suggestions, combining collaborative filtering and content-based filtering. To provide the most pertinent job suggestions, the system examines the student's resume, specifications, and posting. Additionally, the system suggests the top jobs to the user by analyzing and gauging the similarity between the user choice and explicit job listing features. The Recommender System is then evaluated using precision, recall, and F1 score.
Keywords: NLP, Cosine Similarity, Word2Vec, Content- Based Filtering.
DOI: https://doi.org/10.5281/zenodo.10200263
PDF: https://ijirst.demo4.arinfotech.co/assets/upload/files/IJISRT23NOV818.pdf
REFERENCES