Gen-AI Powered Student Academic Analysis System Large Language Models

Abstract

This paper addresses the challenge of improving student academic outcomes through intelligent performance analysis and personalized advisory support aimed at reducing failure rates. The rapid digital transformation of higher education demands academic platforms capable of providing predictive insights and adaptive guidance beyond traditional result-reporting systems. This work presents an AI-driven Student Academic Analysis and Advisory System designed to enable proactive academic monitoring within a multi-role institutional framework. The proposed platform integrates academic performance trend analysis with generative artificial intelligence to deliver forward-looking academic insights, individualized feedback, and interactive guidance. Implemented using a modular Django-based architecture, the system provides role-based access for students, faculty, and management. Historical academic records—including semester performance, subject outcomes, and attendance patterns—are analysed to identify performance trajectories and detect potential academic risks at an early stage. A large language model (Gemini-2.5-flash) is utilized to generate context-aware study recommendations, structured performance evaluations, and real-time chatbot-based academic assistance. The framework further incorporates percentile-based peer comparison, ranking analytics, attendance–performance correlation analysis, and visualization dashboards to enhance interpretability and institutional decision support. Experimental evaluation using institutional academic datasets demonstrates improved performance awareness and actionable AI-generated recommendations. The proposed architecture establishes a scalable foundation for AI-enabled academic advisory systems in modern higher education environments.

Introduction

This study presents an LSTM-based Traffic Flow Prediction System designed to address urban traffic congestion caused by rapid population growth, increasing vehicle ownership, and limited road infrastructure. Traffic congestion leads to economic losses, higher fuel consumption, environmental pollution, and commuter inconvenience, making accurate traffic forecasting essential for modern Intelligent Transportation Systems (ITS).

Traditional traffic prediction methods, such as statistical models (Linear Regression, ARIMA) and classical machine learning techniques (SVM, ANN), often fail to capture the complex nonlinear and temporal patterns present in traffic data. To overcome these limitations, the proposed system utilizes Long Short-Term Memory (LSTM) neural networks, which are highly effective in modeling sequential data and long-term temporal dependencies.

The system is implemented as a web-based application that enables real-time traffic prediction. Its architecture consists of three main layers:

1. User Interface Layer – Developed using HTML, CSS, and JavaScript for user interaction and input collection.
2. Backend Processing Layer – Built with Flask, responsible for data preprocessing, normalization, model selection, and communication between the frontend and prediction engine.
3. LSTM Prediction Layer – Uses PyTorch-based LSTM models trained on historical traffic datasets to forecast future traffic flow.

The workflow includes historical traffic data collection, preprocessing, sequence generation, model training for different cities, user input through the web interface, and real-time traffic prediction display. Separate city-specific LSTM models improve localization and prediction accuracy.

Experimental results show that the LSTM models successfully capture temporal traffic patterns and outperform traditional machine learning approaches by reducing prediction errors and improving robustness. The modular architecture ensures scalability, maintainability, and efficient deployment, while the web interface enhances accessibility and practical usability.

Conclusion

This paper presented an AI-driven Student Academic Analysis and Advisory System designed to support intelligent academic monitoring and personalized guidance in higher education environments. By integrating performance trend analysis with Gemini AI–based generative advisory capabilities, the proposed framework transforms conventional academic portals into intelligent decision-support platforms. The system enables early identification of academic risk, provides contextualized recommendations, and enhances institutional visibility through visualization dashboards and multi-role access mechanisms. Experimental evaluation demonstrates that AI-generated feedback improves interpretability and supports proactive academic intervention. Future work will focus on incorporating extended longitudinal datasets and additional contextual indicators to further enhance advisory precision and scalability.

References

[1] D. Kabakchieva, “Student performance prediction by using data mining classification algorithms,” International Journal of Computer Science and Management Research, vol. 1, no. 4, pp. 686–690, 2012. [2] Z. Iqbal, J. Qadir, A. N. Mian, and F. Kamiran, “Machine learning based student grade prediction: A case study,” arXiv preprint arXiv:1708.08744, 2017. [3] R. Ghorbani and R. Ghousi, “Comparing different resampling methods in predicting students’ performance using machine learning techniques,” IEEE Access, vol. 8, pp. 67899–67911, 2020. [4] V. Kumar and M. L. Garg, “Comparison of machine learning models in student result prediction,” in ICANI, Springer, 2019. [5] M. F. Masood et al., “Towards the selection of best machine learning model for student performance analysis and prediction,” in ISCMI, IEEE, 2019. [6] M. Nirmala, T. S. Selvi, and V. Saravanan, “Student academic performance prediction under various machine learning classification algorithms,” IJRASET, vol. 9, no. 11, 2021. [7] [7] M. Pojon, “Using machine learning to predict student performance,” Master’s thesis, 2017. [8] [8] S. Rai et al., “Machine learning approach for student academic performance prediction,” in FICTA 2020, Springer, 2021. [9] [9] M. S. Ram et al., “Machine learning based student academic performance prediction,” in ICIRCA, IEEE, 2021. [10] K. Kishor, R. Sharma, and M. Chhabra, “Student performance prediction using technology of machine learning,” Springer, 2021. [11] B. Albreiki, N. Zaki, and H. Alashwal, “A systematic literature review of student performance prediction using machine learning techniques,” Education Sciences, 2021. [12] P. Balaji et al., “Contributions of machine learning models towards student academic performance prediction: A systematic review,” Applied Sciences, 2021. [13] S. Gupta and J. Agarwal, “Machine learning approaches for student performance prediction,” in ICRITO, IEEE, 2022. [14] D. T. Ha et al., “An empirical study for student academic performance prediction,” IJCSIS, 2020.

Copyright

Copyright © 2026 Venkateswari Gavini, Kotcherla Venkata Sunitha, Gummadi Keerthana , Duddu Likhitha Priya . 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.