SmartAssess - An Intelligent Assessment Recommenda on System

Abstract

SmartAssess is a web-based intelligent assessment recommendation system designed to assist recruiters and students in identifying the most suitable assessments for specific job roles. Traditional keyword-based systems fail to capture the actual intent of job descriptions, leading to inaccurate recommendations. SmartAssess uses semantic search and embedding-based similarity techniques to understand user requirements more effectively. Assessment data is converted into vector embeddings and stored using FAISS for efficient retrieval. This system reduces manual effort, improves accuracy, and supports informed decision-making in recruitment and career preparation.

Introduction

SmartAssess is an AI-powered assessment recommendation system designed to provide personalized and efficient evaluation in educational and recruitment contexts. Traditional assessment methods, which use the same tests for all learners or rely on keyword-based search for recruiters, often fail to reflect individual skills or job requirements. SmartAssess addresses these limitations by using large language models (Google Gemini) to extract skills from job descriptions, semantic embeddings via Sentence Transformers, and FAISS vector similarity search to identify the most relevant assessments.

The system features a web-based client-server architecture with secure user authentication, a recommendation engine, and a MySQL database for storing user data and assessment catalogs. Its workflow allows users to input job requirements, extract key skills, and quickly receive ranked assessment suggestions.

Results and Benefits:

• Generates accurate, context-aware assessment recommendations in seconds.
• Reduces manual effort and errors in assessment selection.
• Supports adaptive, personalized learning and recruitment evaluation.
• Ensures secure access and consistent system performance.

SmartAssess demonstrates how AI, semantic search, and LLMs can improve evaluation processes by aligning assessments with individual or job-specific requirements, enhancing efficiency, and reducing human effort.

Conclusion

The SmartAssess system provides an intelligent and efficient solution for recommending relevant assessments based on job descriptions and required skills. The system uses artificial intelligence techniques such as semantic search, skill extraction, and vector similarity search to automate the assessment selection process. By integrating technologies like FastAPI, Flask, Google Gemini API, Sentence Transformers, and FAISS, the platform can analyze job requirements and provide suitable assessment recommendations within a short time. The system reduces the need for manual browsing of assessment catalogs and improves the accuracy of recommendations by understanding the contextual meaning of job descriptions. The user-friendly web interface allows recruiters and users to easily interact with the system and obtain results quickly. Overall, SmartAssess improves efficiency, reduces effort, and provides a practical approach for intelligent assessment recommendation.

References

[1] A. Das and M. A. Akour, \"Intelligent Recommendation System for E-Learning using Membership Optimized Fuzzy Logic Classifier,\" 2020 IEEE Pune Section International Conference (PuneCon), Pune, India, 2020, pp. 1-10, doi: 10.1109/PuneCon50868.2020.9362416. [2] Q. Zhang and M. Abisado, \"Towards Designing an Intelligent Recommender System using Adaptive Collaborative Filtering Technique,\" 2023 3rd International Symposium on Computer Technology and Information Science (ISCTIS), Chengdu, China, 2023, pp. 506-511, doi: 10.1109/ISCTIS58954.2023.10213045. [3] R. Leniowski, L. Leniowska, M. Wro?ski, K. Tomecki, M. Grochowina and ?. Ryk, \"Intelligent candidate recommendation system based on experimental calculation of the similarity model,\" 2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), Padua, Italy, 2022, pp. 1-7, doi: 10.1109/FUZZ-IEEE55066.2022.9882820 [4] Lu, J., Wu, D., Mao, M., Wang, W. and Zhang, G., 2015. Recommender system application developments: a survey. Decision support systems, 74, pp.12-32. [5] H. Xu and F. Yan, \"Optimization of Intelligent Recommendation Algorithms and System Design for Online Music Education,\" 2025 International Conference on Digital Analysis and Processing, Intelligent Computation (DAPIC), Incheon, Korea, Republic of, 2025, pp. 652-657, doi: 10.1109/DAPIC66097.2025.00126. [6] H. Zhu, \"Intelligent Recommendation Algorithm for Website Content Update and Maintenance,\" 2024 International Conference on Internet of Things, Robotics and Distributed Computing (ICIRDC), Rio De Janeiro, Brazil, 2024, pp. 315-320, doi: 10.1109/ICIRDC65564.2024.00061. [7] V. C.P, P. Poornachandra and R. Shetty, \"EVAL IQ: Intelligent Educational Assessment and Recommendation Platform,\" 2024 IEEE 4th International Conference on ICT in Business Industry & Government (ICTBIG), Indore, India, 2024, pp. 1-6, doi: 10.1109/ICTBIG64922.2024.10911676. [8] W. Jinghua and F. Rong, \"An Intelligent Agent System for Borrower\'s Recommendation in P2P Lending,\" 2010 International Conference on Multimedia Communications, Hong Kong, China, 2010, pp. 179-182, doi: 10.1109/MEDIACOM.2010.69. [9] N. Kerimbayev, K. Adamova, V. Jotsov, R. Shadiev, Z. Umirzakova and A. Nurymova, \"Organization of Feedback in the Intelligent Learning Systems,\" 2024 IEEE 12th International Conference on Intelligent Systems (IS), Varna, Bulgaria, 2024, pp. 1-7, doi: 10.1109/IS61756.2024.10705178. [10] Y. Guo, \"Optimization and Effect Evaluation of Deep Learning in Intelligent Advertising Recommendation System,\" 2025 5th International Conference on Neural Networks, Information and Communication Engineering (NNICE), Guangzhou, China, 2025, pp. 1545-1548, doi: 10.1109/NNICE64954.2025.11064692. [11] Xin Sui, Suozhu Wang and Zhaowei Li, \"Research on the model of Integration with Semantic Web and Agent Personalized Recommendation System,\" 2009 13th International Conference on Computer Supported Cooperative Work in Design, Santiago, 2009, pp. 233-237, doi: 10.1109/CSCWD.2009.4968064. [12] S. S. Potter and D. D. Woods, \"Event driven timeline displays: beyond message lists in human-intelligent system interaction,\" Conference Proceedings E International Conference on Systems, Man, and Cybernetics, Charlottesville, VA, USA, 1991, pp. 1283-1288 vol.2, doi: 10.1109/ICSMC.1991.16 [13] S. S. Potter and D. D. Woods, \"Event driven timeline displays: beyond message lists in human-intelligent system interaction,\" Conference Proceedings 1991 IEEE International Conference on Systems, Man, and Cybernetics, Charlottesville, VA, USA, 1991, pp. 1283-1288 vol.2, doi: 10.1109/ICSMC.1991.169864. [14] K. Ehimwenma, M. Beer and P. Crowther, \"Pre-assessment and Learning Recommendation Mechanism for a Multi-agent System,\" 2014 IEEE 14th International Conference on Advanced Learning Technologies, Athens, Greece, 2014, pp. 122-123, doi: 10.1109/ICALT.2014.43. [15] \"IEEE Draft Standard for Transparency of Autonomous Systems,\" in IEEE P7001/D3, September 2021 , vol., no., pp.1-75, 27 Sept. 2021.

Copyright

Copyright © 2026 Bandaru Vijaya Lakshmi Satya Tulasi, Yerubandi Jaya Sai Vinay Kumari, Bheemala Naga Durga Rama Kumari , Bandaru Divakara Rama Lakshmi Ganapathi , Ethakota Shyam Prasad Varma , Ganji Lakshmi. 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.