International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 10 | Oct 2025
p-ISSN: 2395-0072
www.irjet.net
AI-Driven Placement Prediction and Recommendation System for College Students Manish Khodaskar1, Adesh Gajare2, Chetan Bochare3, Shailesh Patil4 1Professor, Dept. of IT, Pune Institute of Computer Technology, Pune, Maharashtra, India. 2Student, Dept. of IT, Pune Institute of Computer Technology, Pune, Maharashtra, India. 3Student, Dept. of IT, Pune Institute of Computer Technology, Pune, Maharashtra, India.
4Student, Dept. of IT, Pune Institute of Computer Technology, Pune, Maharashtra, India.
---------------------------------------------------------------------***---------------------------------------------------------------------
admissions. Therefore, it is the responsibility of educational institutions and placement cells to guide students effectively and provide them with better opportunities. Traditional placement preparation methods may not always be sufficient, as they often fail to provide personalized insights into a student’s strengths and weaknesses.
Abstract -Campus placements are one of the most
important milestones in a student’s academic journey. However, the process is often uncertain and many students lack the right guidance to meet industry expectations. Companies look for strong technical skills, problem-solving ability, and consistent academic performance, but students are often unaware of these requirements. To bridge this gap, we propose an AI-driven placement prediction and recommendation system that helps students understand their placement chances and improve their preparation. The system predicts both the likelihood of a student being placed and the expected salary package using features such as CGPA, aptitude test scores, coding profiles, and backlog status. In addition to prediction, the system integrates a Generative AI module that provides personalized recommendations, including suitable job roles, skill-gap analysis, and focused preparation strategies. To ensure transparency, explainable AI methods high light the key factors influencing the predictions. A user-friendly dashboard built with Streamlit allows students to view their placement chances in real time and receive actionable suggestions for improvement. This project aims to support students in achieving better placement outcomes while also helping institutions strengthen their overall placement performance.
To address this issue, we propose an AI-driven placement prediction and recommendation system for PICT students. The system analyzes historical placement records along with student data such as CGPA, AMCAT scores, subject grades, coding profiles, and backlog status to predict both the expected placement package and the likelihood of a student being placed. Students are classified into three categories: Low, Average, and Strong placement chances. In addition to prediction, the system incorporates Generative AI to provide tailored recommendations, helping students enhance their technical and academic skills for improved career outcomes. This project not only assists students in understanding their placement readiness but also helps institutions strengthen their placement outcomes. By combining data-driven prediction with actionable guidance, the proposed model bridges the gap between student preparation and industry demands, thereby improving both the efficiency and transparency of the placement process.
Key Words: —Campus Placement, Placement Prediction, Ma chine Learning, Regression, Classification, Generative AI, Explainable AI, Recommendation System,
2. LITERATURE SURVEY In this section, we review previous research studies that applied supervised machine learning and deep learning techniques for campus placement prediction. These works demonstrate different datasets, features, and algorithms, highlighting their strengths and limitations.
1. INTRODUCTION Campus placements have become one of the most significant turning points in a student’s academic journey. While many engineering graduates enter the workforce each year, only a small percentage are able to meet the expectations of employers. Since the IT sector is continuously evolving, organizations look for students with strong foundational knowledge, problem-solving skills, and relevant technical expertise. However, a gap often exists between academic performance and industry readiness, as many students are unaware of these industry requirements.
In [1], the authors developed a campus placement prediction system using logistic regression to estimate the probability of a student being placed. The dataset included academic features such as CGPA, attendance, and test results from the placement management system. Historical data from past students were used to train the model. The work emphasizes how logistic regression, with properly selected training tuples, can provide reliable predictions, helping both students and faculty to identify skill gaps and improve academic planning.
Placement statistics also play a crucial role in defining a college’s reputation and attracting new
© 2025, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 77