FinMate: A Cloud-Based AI and Machine Learning Systemfor Smart Finance Management
Keywords:
fintech, XGBoost, LSTM, Isolation Forest, expense categorization, tax optimization, anomaly detection, cloud microservicesAbstract
Managing your own money has become a lot harder since the number of digital payments in India has gone up a lot. The most popular tools still rely on fixed keyword heuristics, which means they can't learn, make predictions about the future, or give tax advice. FinMate fixes these problems with a cloud-native microservices platform that combines three machine learning models with a rule-based engine for optimising taxes. An XGBoost classifier puts incoming transactions into one of 18 spending categories and gets a weighted F1 score of 0.943, which is 32.8% better than a keyword baseline. A hybrid LSTM/ARIMA model makes spending predictions for 30, 60, and 90 days. It has a 30-day RMSE of ₹312, which is 23.4% less than a seasonal naive method.An Isolation Forest module spots strange transactions in real time with 91.3% accuracy, which is 14.3 percentage points better than a z-score detector. The tax engine connects different types of annual spending to deductions under the Indian Income Tax Act, including Sections 80C, 80D, HRA, 80E, and the standard deduction. This covers 85% of the most common types of deductions. An evaluation of 2,400 held-out transactions from a 12,000-record corpus, along with an ablation study and user acceptance testing (n=15), demonstrates a 73% reduction in time-to-financial-insight and a System Usability Scale score of 81.4. Anyone can see all the code, data, and a Docker deployment so they can make their own version.
References
[1] KPMG India and Google, "Digital Payments 2025: Fintech Adoption in Urban India," KPMG Rep., New Delhi, India, 2025.
[2] A. Bhatt and R. Mehta, "Limitations of conventional personal finance applications: A systematic review," Int. J. Financial Technol., vol. 4, no. 2, pp. 112–124, 2021.
[3] Y. Zhang et al., "An enhanced XGBoost algorithm for mobile price classification," in Proc. IEEE Int. Conf. Big Data, 2023, pp. 123–130.
[4] M. I. D. Hasibuan et al., "Classification of single tuition fee with imbalanced data using XGBoost," in Proc. IEEE ICDMW, 2024, doi: 10.1109/ICDMW.2024.10797103.
[5] S. Hochreiter and J. Schmidhuber, "Long short-term memory," Neural Comput., vol. 9, no. 8, pp. 1735–1780, Nov. 1997.
[6] M. A. M. B. Shaikh et al., "FinMate — cloud-based AI & ML financial management system: System concept and interface design," in Proc. Natl. Conf. Emerg. Trends Comput. Eng., Nashik, India, 2023/2024.
[7] M. L. Pervan and T. Matošević, "Comparative analysis of machine learning methods for credit card fraud detection," Appl. Sci., vol. 11, no. 10, Art. no. 4580, May 2021.
[8] J. V. M. Gonçalves, "A comparative analysis of financial time series forecasting: ARIMA vs. LSTM," Univ. São Paulo, Tech. Rep., 2023.
[9] "AI-driven financial insights for personal budget planning," IEEE Access, vol. 13, pp. 4567–4578, Feb. 2025, doi: 10.1109/ACCESS.2025.10914721. [10] F. T. Liu, K. M. Ting, and Z.-H. Zhou, "Isolation forest," in Proc. 8th IEEE ICDM, Pisa, Italy, Dec. 2008, pp. 413–422.
[11] F. Liu et al., "Isolation forest for financial transaction anomaly detection," New Digitals J., Oct. 2023.
[12] "Personal finance manager with predictive analytics using LSTM," Int. J. Adv. Res. Sci. Commun. Technol., vol. 5, no. 2, pp. 112–120, May 2025.
[13] G. E. P. Box et al., Time Series Analysis: Forecasting and Control, 5th ed. Hoboken, NJ, USA: Wiley, 2015.
[14] N. Patil and A. Deshmukh, "AI-driven personal finance management for Indian millennials," Int. J. Eng. Technol., vol. 11, no. 4, pp. 234–242, 2022.
[15] "How India's tax-tech startups are rising," BillCut Blog, 2025. [Online]. https://billcut.in
[16] Ministry of Finance, India, "Income tax deductions under Chapter VI-A," Income Tax Dept., New Delhi, Tech. Rep., 2026.
[17] R. Smith, "An overview of the Tesseract OCR engine," in Proc. 9th ICDAR, Sep. 2007, pp. 629–633.
[18] MeitY, Govt. of India, "The Digital Personal Data Protection Act 2023," Gazette of India, Aug. 2023.
[19] B. McMahan et al., "Communication-efficient learning of deep networks from decentralized data," in Proc. AISTATS, Apr. 2017, pp. 1273–1282.
[20] A. Bangor, P. Kortum, and J. Miller, "Determining what individual SUS scores mean," J. Usability Studies, vol. 4, no. 3, pp. 114–123, May 2009. [21] "Cloud computing and microservices architecture for financial services on AWS," SRC Publ. J. AI Cloud Comput., vol. 2, no. 1, Aug. 2024. [22] Amazon Web Services, "AWS well-architected framework: Microservices for FinTech," AWS Whitepaper, Seattle, WA, USA, 2025.
[23] Reserve Bank of India, "Handbook of statistics on the Indian economy 2022–23," RBI, Mumbai, 2023.
[24] IDRBT, "Indian payments dataset for research," IDRBT, Hyderabad, 2022.
[25] Intuit Inc., "Mint: Personal finance product overview," 2025. [Online]. https://mint.intuit.com
[26] YNAB LLC, "You Need A Budget: Methodology and features," 2025. [Online]. https://www.youneedabudget.com
[27] Reserve Bank of India, "Account aggregator framework: Guidelines for regulated entities," RBI Circular, Mumbai, May 2021.
[28] P. Kairouz et al., "Advances and open problems in federated learning," Found. Trends Mach. Learn., vol. 14, no. 1–2, pp. 1–210, Jun. 2021.
[29] V. Sanh et al., "DistilBERT, a distilled version of BERT," arXiv:1910.01108, Oct. 2019.
[30] A. Das et al., "Long-term forecasting with TiDE: Time-series dense encoder," Trans. Mach. Learn. Res., 2024.
[31] H. Touvron et al., "Llama 2: Open foundation and fine-tuned chat models," arXiv:2307.09288, Jul. 2023.
