An Intelligent Book Recommendation System Using Machine Learning: Bridging Content-Based and Popularity- Driven Approaches

Abstract

The document presents an intelligent tour planner that simplifies travel planning by generating personalized itineraries using a rule-based AI approach. Users input preferences such as destination, duration, budget, and activity type, and the system generates a structured day-wise travel plan including activities, hotel suggestions, travel details, and budget estimates. It is developed using Python and Streamlit, with JSON-based storage for user data, login credentials, and travel history, and also includes features like authentication, PDF download, and trip tracking. The motivation is to reduce the complexity of travel planning, which is often scattered across multiple platforms, and to provide customized recommendations tailored to individual user needs. The problem addressed is the inefficiency and confusion in manual travel planning due to fragmented information sources. The literature survey shows that existing AI and ML-based travel systems are often complex and dependent on external APIs, while simpler rule-based systems lack adaptability. The proposed system balances this by using a rule-based approach without external dependencies. The methodology includes requirement analysis, system design, frontend and backend development, rule definition, storage integration, testing, and deployment. The system architecture processes user inputs through predefined rules to generate itineraries and cost estimates. Future improvements include integrating real-time APIs (weather, maps, bookings), adding machine learning for personalization, developing mobile apps, using cloud databases, and introducing advanced features like chatbot assistants, route optimization, and payment integration to make the system more intelligent and scalable.

Introduction

The document describes an intelligent tour planner that uses a rule-based AI system to generate personalized travel itineraries based on user inputs like destination, budget, duration, and interests. The system provides a complete travel plan including day-wise schedules, hotel suggestions, travel details, and budget estimates, and is built using Python and Streamlit with JSON-based storage and features like authentication, PDF export, and trip history.

It addresses the difficulty and fragmentation in traditional travel planning by offering a single, structured, and customized solution. While existing systems often rely on complex AI/ML models or external APIs, this approach uses a simpler rule-based method for ease of use and independence from external services.

The system is developed through structured stages including design, rule definition, implementation, testing, and deployment. Future enhancements aim to improve it with real-time data integration, machine learning-based personalization, mobile app support, cloud storage, and advanced features like chatbots, route optimization, and booking services.

Conclusion

This research paper presented a comprehensive hybrid book recommendation system addressing the fundamental challenge of balancing content relevance with recommendation diversity. The dual-architecture approach—combining academic-grade 70/30 TF-IDF weighting with production-optimized 60/40 configuration—demonstrates that thoughtful algorithm design can effectively mitigate popularity bias while maintaining user satisfaction. Key contributions of this work include: 1) Problem Identification: Documented popularity bias phenomenon in academic recommendation systems. 2) Solution Design: Engineered 60/40 optimization achieving 15% diversity improvement 3) Implementation: Developed production-ready Streamlit application for real-time deployment 4) Evaluation: Conducted comprehensive performance analysis across relevance, diversity, and bias metrics 5) Impact: Demonstrated practical pathway from research algorithm to user-facing application The results validate the hypothesis that reducing popularity weighting from 30% to 40% while correspondingly increasing content weighting significantly enhances recommendation diversity without sacrificing content relevance. Statistical analysis (?² = 23.4, p < 0.001) provides strong evidence for the optimization effectiveness. As recommendation systems become increasingly critical to digital content discovery, this research contributes methodological insights for designing ethical AI systems that balance algorithmic precision with equitable content access. The implementation serves as a functional prototype for educational purposes and demonstrates viability for production deployment with appropriate scaling infrastructure.

References

[1] Anderson, C. (2006). The Long Tail: Why the Future of Business is Selling Less of More. Hyperion. [2] Ricci, F., Rokach, L., & Shapira, B. (2015). Recommender systems handbook. Springer.https://doi.org/10.1007/978-1-4899-7637-6 [3] McSherry, D. (2005). Diversity-conscious retrieval. In European Conference on Case- Based Reasoning (pp. 849-849). Springer Berlin Heidelberg. [4] Ekstrand, M. D., Harper, F. M., &Konstan, J. A. (2011). iLike: Personalized recommendations for music. In Proceedings of the 5th ACM conference on Recommender systems (pp. 199-206). [5] Jannach, D., Zanker, M., Felfernig, A., & Friedrich, G. (2010). Recommender Systems: An Introduction. Cambridge University Press. [6] Burke, R. (2002). Hybrid recommender systems: Survey and evaluation. User modeling and user-adapted interaction, 12(4), 331-370. https://doi.org/10.1023/A:1021240730564 [7] Pazzani, M. J., &Billsus, D. (2007). Content-based recommendation systems. The adaptive web, 4321, 325-341.https://doi.org/10.1007/978-3-540-72079-9_10 [8] Lops, P., de Gemmis, M., & Sera no, G. (2011). Content-based recommender systems: State of the art and trends. Recommender systems handbook, 73-105.https://doi.org/10.1007/978-0-387-85820-3_3 [9] Koren, Y., Bell, R., & Volinsky, C. (2009). Matrix factorization techniques for recommender systems. Computer, 42(8), 30-37.https://doi.org/10.1109/MC.2009.263 [10] Goldberg, D., Nichols, D., Oki, B. M., & Terry, D. (1992). Using collaborative filtering to weave an information tapestry. Communications of the ACM, 35(12), 61-70. https://doi.org/10.1145/138859.138867 [11] Desrosiers, C., &Karypis, G. (2011). A comprehensive survey of neighborhood-based recommendation methods. Recommender systems handbook, 107-149.https://doi.org/10.1007/978-0-387-85820-3_4 [12] Lika, B., Kolomvatsos, K., &Hadjiefthymiades, S. (2014). Facing the cold start problem in recommender systems. Expert Systems with Applications, 41(4), 2065-2073.https://doi.org/1 [13] 0.1016/j.eswa.2013.09.005 [14] Melville, P., Mooney, R. J., & Nagarajan, R. (2002). Content-boosted collaborative filtering for improved recommendations. In AAAI (Vol. 2, pp. 187-192). [15] Burke, R. (2002). Hybrid recommender systems: Survey and evaluation. User Modeling and User-Adapted Interaction, 12(4), 331-370. [16] Montaner, M., López, B., & de la Rosa, J. L. (2003). A taxonomy of recommended agents on the internet. Arti cial Intelligence Review, 19(4), 285-330. https://doi.org/10.1023/A:1022903627593 [17] Abdollahpouri, H., Burke, R., & Mobasher, B. (2017). Controlling popularity bias in learning-to-rank recommendation. In Proceedings of the 11th ACM conference on recommender systems (pp. 42-46).https://doi.org/10.1145/3109859.3109912 [18] Park, Y. J., &Tuzhilin, A. (2008). The long tail of recommender systems and how to exploit it. In Proceedings of the 2008 ACM conference on Recommender systems (pp. 11-18). https://doi.org/10.1145/1454008.1454012 [19] Kunaver, M., &Požrl, T. (2017). Diversity in recommender systems. Journal of computer and system sciences, 94, 73-94.https://doi.org/10.1016/j.jcss.2017.06.012 [20] Pariser, E. (2011). The Filter Bubble: What the Internet is Hiding from You. Penguin Press. [21] Said, A., &Bellogín, A. (2014). Comparative recommender system evaluation. In Proceedings of the 8th ACM conference on recommender systems (pp. 129-136). https://doi.org/10.1145/2645710.2645746 [22] Salton, G., & McGill, M. J. (1983). Introduction to Modern Information Retrieval. McGraw-Hill. [23] Robertson, S. (2004). Understanding inverse document frequency: on theoretical arguments for IDF. Journal of documentation, 60(5), 503-520.https://doi.org/10.1108/00220410410560573 [24] Steinbach, M., Karypis, G., & Kumar, V. (2000). A comparison of document clustering techniques. In KDD workshop on text mining (Vol. 400, pp. 525-526). [25] Ozdal, B., & Uyar, A. (2013). A book recommendation system using collaborative and content-based filtering. In 2013 12th International Conference on Machine Learning and Applications (Vol. 1, pp. 104-109). IEEE.https://doi.org/10.1109/ICMLA.2013.23 [26] Cremonesi, P., Garzotto, F., & Turrin, R. (2012). Investigating the persuasion potential of recommender systems from a quality perspective. In Proceedings of the fth ACM conference on Recommender systems (pp. 255-262).https://doi.org/10.1145/2365952.2366007 [27] Wang, X., Li, Y., & Huang, Z. (2014). A collaborative ltering recommendation algorithm incorporating user preference intensity. In 2014 IEEE International Conference on Systems, Man, and Cybernetics (SMC) (pp. 1240-1245). IEEE. [28] Ziegler, C. N., McNee, S. M., Konstan, J. A., & Lausen, G. (2005). Improving recommendation lists through topic diversification. In Proceedings of the 14th international conference on World Wide Web (pp. 22-32). https://doi.org/10.1145/1060745.1060754 [29] Kaggle Books Dataset. Retrieved fromhttps://www.kaggle.com/datasets/... [30] Herlocker, J. L., Konstant, J. A., Terveen, L. G., & Riedl, J. T. (2004). Evaluating collaborative filtering recommender systems. ACM Transactions on Information Systems (TOIS). [30]Shani, G., & Gunawardana, A. (2011). Evaluating recommendation systems. [31] Recommender systems handbook, 257-297.https://doi.org/10.1007/978-0-387-85820-3_8 [32] Vargas, S., & Castells, P. (2011). Rank and relevance in novelty and diversity metrics for recommender systems. In Proceedings of the fth ACM conference on Recommender systems (pp. 109-116).https://doi.org/10.1145/2365952.2365976 [33] Zhou, T., Kuscsik, Z., Liu, J. G., Medo, M., Wakeling, J. R., & Zhang, Y. C. (2010). Solving the apparent diversity-accuracy dilemma of recommender systems. Proceedings of the national academy of sciences, 107(10), 4511-4515.https://doi.org/10.1073/pnas.1000488107 [34] Tatar, A., Antonopoulos, M., de Amorim, M. D., &Fdida, S. (2014). Ranking page importance in web search. arXiv preprint arXiv:1402.6273. [35] Kirnap, Ö., Kumova, B. I., & Kilinc, D. (2017). The e ect of popularity bias on recommendation systems. In 2017 International Conference on Information Systems and Software Technology (ICISST) (pp. 1-5). IEEE. [36] Ekstrand, M. D., Riedl, J. T., &Konstan, J. A. (2010). Collaborative filtering recommender systems. Foundations and Trends® in Human–Computer Interaction, 4(2), 81-173. https://doi.org/10.1561/1100000009 [37] Wu, Y., & Liu, S. (2009). Collaborative filtering with heterogeneous datatypes. Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining (pp. 1373-1382).https://doi.org/10.1145/1557019.1557185 [38] Järvelin, K., & Kekäläinen, J. (2002). Cumulated gain-based evaluation of IR techniques. [39] ACM Transactions on Information Systems (TOIS), 20(4), 422-446. https://doi.org/10.1145/582415.582418 [40] McNemar, Q. (1947). Note on the sampling error of the di erence between correlated proportions or percentages. Psychometrika, 12(2), 153-157.https://doi.org/10.1007/BF02295996 [41] Schütze, H., Manning, C. D., & Raghavan, P. (2008). Introduction to Information Retrieval (Vol. 39). Cambridge University Press.https://nlp.stanford.edu/IR-book/ [42] Covington, P., Adams, J., & Sargin, E. (2016). Deep neural networks for youtube recommendations. In Proceedings of the 10th ACM conference on recommender systems (pp. 191-198).https://doi.org/10.1145/2959100.2959190 [43] Williams, M. R., Ren, F., & Li, S. (2016). Food for thought: Assessing the in uence of book reviews on reader choice. In ICWSM. [44] Abdollahpouri, H., & Burke, R. (2019). Unfairness in recommendation ranking through intra-list bias in pairwise comparisons. In Proceedings of the 27th ACM conference on user modeling, adaptation and personalization (pp. 374-383).https://doi.org/10.1145/3298589.3298646 [45] Dean, J., & Ghemawat, S. (2008). MapReduce: simpli ed data processing on large clusters. Communications of the ACM, 51(1), 107-113.https://doi.org/10.1145/1327452.1327492 [46] Mikolov, T., Sutskever, I., Chen, K., Corrado, G. S., & Dean, J. (2013) . [47] Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. (2018). BERT: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805. [48] Ning, X., Karypis, G., & Schreiber, R. (2015). Evaluation of item-based top-n recommendation algorithms. In Proceedings of the tenth ACM conference on recommender systems (pp. 285-292).https://doi.org/10.1145/2800128.2800177 [49] Matignon, R., & Anand, S. (2015). Handling recommendation freshness and diversity in the context of multi-stakeholder recommendation systems (Doctoral dissertation, University of Louisville). [50] He, R., & McAuley, J. (2016). Ups and downs: Modeling the visual evolution of fashion trends with one-class collaborative filtering. In proceedings of the 25th international conference on world wide web (pp. 507-517).https://doi.org/10.1145/2872427.2883037 [51] Cheng, H. T., Koc, L., Harmsen, J., Shaked, T., Chandra, T., Aradhye, H., ... & Levin, D. (2016). Wide & deep learning for recommender systems. In Proceedings of the 1st workshop on deep learning for recommender systems (pp. 7-10).https://doi.org/10.1145/2988450.2988454 [52] Xue, H. J., Dai, X., Zhang, J., Huang, S., & Chen, B. (2017). Deep matrix factorization models for recommender systems. In IJCAI (pp. 3882-3888). [53] Li, L., Chu, W., Langford, J., &Schapire, R. E. (2010). A contextual-bandit approach to personalized news recommendation. In Proceedings of the 19th international conference on the World wide web (pp. 661-670).https://doi.org/10.1145/1772690.1772758 [54] Schein, A. I., Popescul, A., Ungar, L. H., & Pennock, D. M. (2002). Methods and metrics for cold-start recommendations. In SIGIR (Vol. 2, pp. 253-260). https://doi.org/10.1145/564376.564421 [55] Kohavi, R., Longbotham, R., Sommer eld, D., & Henne, R. M. (2009). Controlled experiments on the web: survey and practical guide. Data Mining and Knowledge Discovery, 18(1), 140-181.https://doi.org/10.1007/s10618-008-0114-1 [56] Zhou, T., Kuscsik, Z., Liu, J. G., Medo, M., Wakeling, J. R., & Zhang, Y. C. (2010). [57] Solving the apparent diversity-accuracy dilemma of recommender systems. Proceedings of the [58] National Academy of Sciences, 107(10), 4511-4515.https://doi.org/10.1073/pnas.1000488107 [59] Cochran, W. G. (1954). Some methods for strengthening the common ?2 tests. Biometrics, 10(4), 417-451.https://doi.org/10.2307/3001616

Copyright

Copyright © 2026 Dr. Manish Madhava Tripathi, Umrah Meraj, Zainab Parveen, Samna Iqbal. 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.