KNN-Driven Intelligent Job Recommendation and Automated Applicant Tracking for Smart Campus Placement Systems

In this paper, I will describe the Smart Campus Placement System (SCPS) which is a web-based tool that is based on the Django framework and automates the campus recruitment pipeline of academic institutions. The system offers student, Training and Placement Officers (TPOs) and recruiting company focused modules, replacing the normal disjointed manual processes of the traditional campus placement. The key automated processes are student registration, resume management, eligibility against established guidelines (CGPA, department, backlogs), job application status, scheduling of interviews, and publication of results. The technical contribution leading to publication is a K-Nearest Neighbors (KNN) recommendation engine which matches students to job opportunities by calculating Euclidean distance between TF-IDF weighted skill vectors, which exposes the five most compatible jobs per student. It was trained and tested on 215 student records of the CMR Engineering College (2022-26), providing an accuracy of 87.4, precision of 0.86, recall of 0.88 and a F1-score of 0.87, which is higher than Naive Bayes (F1: 0.79), Decision Tree (F1: 0.82), and SVM-RBF (F1: 0.85) baselines. The application of the system saved an average of 85 percent of shortlisting time per drive, minimized errors by 91 percent in data entry and increased the rate of institutional placement by 74 percent of 61 percent. These findings show that the SCPS is technically viable, has been shown to be measurably effective, and can be used with large groups of students.