Student Placement and Job Role Predictions

Abstract

Career selection remains a critical challenge for students, influenced by external pressures and insufficient guidance. This study proposes a machine learning-based job role prediction system to address career uncertainty. Leveraging algorithms like Random Forest, Decision Tree, and SVM, the model analyzes academic performance, skills, interests, and personality traits from a dataset of 20,000 records. Data preprocessing included handling missing values, categorical encoding, and feature scaling, while Recursive Feature Elimination identified key predictors. Hyperparameter tuning via Grid Search optimized the Random Forest model, achieving 80% accuracy. A dynamic quiz integrated into the system assesses logical reasoning, reducing self-reporting bias. Implemented as a web tool using Django and Tailwind CSS, the platform offers personalized career recommendations, particularly for computer science students. Results highlight the efficacy of multi-dimensional analysis in reducing career indecision. Future work may explore ensemble methods and real-time skill gap analysis to enhance predictions.

Introduction

The career choices students make are critical to their future success, yet the growing number of options and external pressures often lead to confusion and poor decisions. Many students struggle due to unclear understanding of their own skills and interests, which can affect their job satisfaction and productivity. Early career planning that aligns with a student’s strengths and preferences is essential.

To address this, the authors propose a web-based job role prediction system that uses quizzes and assessments to gather data on students' academic performance, personality traits, and interests. Unlike existing tools, this system incorporates multiple factors to offer personalized and accurate career recommendations. The model, built using Python, Scikit-learn, and Django, utilizes a dataset of 20,000 records from Kaggle and GitHub, focusing on key attributes like coding skills, logical reasoning, and teamwork.

The literature review highlights various machine learning-based career guidance systems, emphasizing the need for personalized and multi-factor approaches. The methodology section details the data preprocessing, feature engineering, and model training steps, with Random Forest achieving 80% accuracy after hyperparameter tuning. Integration of a quiz to dynamically assess logical reasoning improved accuracy by 5%.

The system’s architecture combines Django backend with a Tailwind CSS frontend and uses machine learning models for prediction. Evaluation shows promising results, with potential for further improvements through ensemble methods.

Conclusion

The challenges students face in selecting an appropriate career path, compounded by external pressures and an overwhelming array of choices, necessitate intelligent solutions to guide informed decision-making. This study addresses these challenges by proposing a robust job role prediction system that leverages machine learning algorithms to analyze academic performance, skills, interests, and personality traits. By integrating a dynamic quiz to assess logical reasoning and preprocessing diverse datasets, the system ensures accurate and personalized recommendations. Among the tested models, Random Forest outperformed others with an 80% accuracy after hyperparameter tuning, while feature engineering and automated assessments further enhanced reliability. The system’s web-based implementation using Django and Tailwind CSS offers an accessible platform for students, particularly in computer science, to align their strengths with viable career paths. Results underscore the efficacy of combining multi-faceted data with advanced ML techniques to reduce uncertainty and improve career outcomes. Future work could explore ensemble methods, larger datasets, and real-time skill gap analysis to refine predictions further. Ultimately, this tool empowers students to navigate career complexities confidently, fostering productivity and long-term professional satisfaction

References

[1] P. Shahane, P. Rinke, T. Datar, and S. Badjate, “Student Career Interest Prediction using Machine Learning,” International Research Journal of Engineering and Technology (IRJET), Nov. 2022. [2] D. Chaudhary, H. Prajapati, R. Rathod, P. Patel, and R. K. Gurjwar, “Student Future Prediction Using Machine Learning,” ResearchGate, Jul. 2020. [3] S. Udayasri and C. Varsha, “Career Prediction System Based on Student Preferences,” International Research Journal of Modernization in Engineering, Technology and Science (IRJMETS), Jul. 2017. [4] G. Mahalakshmi and A. A. Kumar, “Job Recommendation System Based on Skill Sets,” 2022. [5] S. A. Alsaif, “Learning-Based Matched Representation System for Job Recommendation,” 2022. [6] H. L. H. S. Warnars, “Trends and Characteristic of Career Recommendation Systems for Fresh Graduates,” 2022. [7] A. H. A. Rashid, M. Mohamad, S. Masrom, and A. Selamat, “Student Career Recommendation System Using Content-Based Filtering Method,” 2022. [8] A. Jose-Garcia, A. Sneyd, A. Merlo, et al., “C3-IoC: A Career Guidance System for Assessing Student Skills Using Machine Learning and Network Visualization,” 2022. [9] A. R. Panhalkar and D. D. Doye, “A novel approach to build accurate and diverse decision tree forest,” Evolutionary Intelligence, vol. 15, no. 1, pp. 439–453, 2022. [10] D. Varghese, “Comparative study on classic machine learning algorithms,” Retrieved July 28, 2018. [11] B. Sekeroglu, K. Dimililer, and K. Tuncal, “Student performance prediction and classification using machine learning algorithms,” Proceedings of the 2019 8th International Conference on Educational and Information Technology, 2019. [12] O. Baval, “Career Prediction Dataset,” GitHub Repository, Accessed 2025.

Copyright

Copyright © 2025 Kanishk Mandrelia, Rahul Chauhan, Ananta Pandey, Anam Khan. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.