TrueMatch: An AI-Driven Smart Matrimonial System

Abstract

With the digitization of matrimonial services, manual data entry remains time-consuming and error-prone, particularly for marriage bureaus managing multiple biodatas. This paper presents TrueMatch, a smart matrimonial platform that automates profile creation and enhances matchmaking accuracy. The system uses OCR to extract text from scanned biodatas and NLP to structure and autofill candidate profiles, reducing manual effort and improving data accuracy. For intelligent matchmaking, a K-Nearest Neighbors (KNN) algorithm computes compatibility scores based on user attributes and preferences, providing personalized match recommendations. Experimental evaluation demonstrates that the OCR-based biodata extraction module achieves an accuracy of approximately 85–90%, while the KNN based matchmaking engine provides compatibility recommendations with an accuracy of around 80–85%. These results validate the system’s effectiveness in improving matchmaking quality, reducing manual effort, and enhancing operational efficiency in digital matrimonial platforms.

Introduction

The text presents TrueMatch, an AI-powered smart matrimonial platform designed to automate profile management and improve matchmaking efficiency for both individual users and marriage bureaus.

• Traditional matrimonial platforms often rely on manual data entry, which is time-consuming, error-prone, and inefficient, especially when handling large numbers of biodatas. TrueMatch addresses these challenges through automation and intelligent matching.
• The platform supports two user roles: individual candidates, who manage their own profiles, and marriage bureau administrators, who manage multiple client profiles.
• A major feature is the use of Optical Character Recognition (OCR) and Natural Language Processing (NLP) to automatically extract and structure information such as personal details, education, profession, and partner preferences from scanned biodatas. This reduces manual effort and improves data accuracy.
• The system is developed using the MERN stack (MongoDB, Express.js, React.js, and Node.js), providing a scalable, secure, and user-friendly web platform.
• For intelligent matchmaking, the platform employs the K-Nearest Neighbors (KNN) algorithm. Candidate attributes and preferences are converted into numerical vectors, and compatibility is calculated using weighted similarity scores to generate personalized match recommendations.
• Additional technologies include Named Entity Recognition (NER), rule-based pattern matching, JWT authentication, and RESTful APIs to enhance data processing, security, and system performance.
• The OCR module automates biodata digitization and profile autofill, making it easier for marriage bureaus to manage large numbers of profiles efficiently.
• The KNN-based recommendation engine analyzes factors such as age, education, profession, location, and preferences to provide accurate and meaningful match suggestions.
• Testing results showed successful implementation of key modules, including profile creation, biodata upload, automatic profile generation, AI-based matchmaking, and administrative management dashboards.
• Overall, TrueMatch offers a secure, automated, and intelligent matrimonial solution that improves operational efficiency, enhances user experience, reduces errors, and delivers more personalized matchmaking services.

Conclusion

This paper presented an AI-driven matrimonial management system that automates profile creation, enhances matchmaking accuracy, and ensures secure data handling. The integration of OCR and NLP reduced manual data entry by approximately 70%, while the KNN-based matching engine improved recommendation accuracy by nearly 25% over manual methods. The system also supports secure authentication, role-based access, and fake profile detection, ensuring reliability. Overall, the proposed solution effectively achieved its automation and accuracy objectives. Future enhancements may include deep learning-based matching models, multilingual biodata processing, and real-time compatibility analytics to further improve system performance.

References

[1] Kaundilya, Chandni, Diksha Chawla, and Yatin Chopra. ”Automated text extraction from images using OCR system.” 2019 6th International Conference on Computing for Sustainable Global Development (INDIACom). IEEE, 2019. [2] Vasani, Vaibhav, et al. ”Bogus user profile detection on matrimonial sites using machine learning approach.” 2021 fourth international conference on electrical, computer and communication technologies (ICECCT). IEEE, 2021. [3] Shinde, Shruti, and Sunil B. Mane. ”Malicious profile detection on social media: a survey paper.” 2021 9th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions)(ICRITO). IEEE, 2021. [4] Chawla, Akhil, Aarushi Gupta, and K. S. Shushrutha. ”Intelligent information retrieval: techniques for character recognition and structured data extraction.” J. Emerg. Technol. Innov. Res.(JETIR) 9.7 (2022). [5] Chia, Yong-Fang, Chian-Wen Too, and Kok-Chin Khor. ”Data-Driven Similarity Measures for Matrimonial Application.” Information Science and Applications: Proceedings of ICISA 2020. Singapore: Springer Singapore, 2021. 277-286. [6] Tomita, Yoji, Riku Togashi, and Daisuke Moriwaki. ”Matching theory based recommender systems in online dating.” Proceedings of the 16th ACM Conference on Recommender Systems. 2022. [7] Lalvani, Simiran, and Joyojeet Pal. ”The moral orders of matchmaking work: Digitization of matrimonial services and the future of work.”Proceedings of the ACM on Human-Computer Interaction 6.CSCW1 (2022): 1-23. [8] Joshi, Kedar, and Sushil Kumar. ”Matchmaking using fuzzy analytical hierarchy process, compatibility measure and stable matching for online matrimony in India.” Journal of Multi-Criteria Decision Analysis 19.1-2 (2012): 57-66.

Copyright

Copyright © 2026 Pooja Patil, Sanika Patil, Saloni Phale, Samruddhi Shinde, Sanika Sankpal, Nandini Patil. 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.