Job Recommendation System
Chetas Parekh, Yuvraj Gupta, Manjot Singh
Department of Computer Science
Faculty Supervisor: Tracy Chen
This project presents a personalized job recommendation system designed to optimize the matching process between candidates and open positions. Leveraging advanced natural language processing and machine learning algorithms, the platform performs in situ analysis of job descriptions and candidate skills.
The underlying data pipeline utilizes a robust web crawling module to systematically extract real-world job postings across multiple platforms, such as Google and Yahoo. Following data acquisition, a specialized skill extraction pipeline—supported by both Hugging Face and Gemini models—identifies key competencies and technical requirements from unstructured text. These extracted features are then processed to compute precise similarity scores, ensuring that users receive highly tailored, relevant job recommendations.
By addressing the de facto challenges of information overload in modern recruiting, this system significantly reduces the manual effort required during the job hunt. Preliminary evaluations demonstrate a marked improvement in matching accuracy over traditional keyword-based searches. Future work will focus on expanding the training dataset and refining the recommendation algorithms a priori to further enhance personalization and scalability.