Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM67629
Volume:
10 |
10 |
Oct 2026
Pages:
1-5
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Department of Computer Science and Engineering
East West Institute of Technology
Department of Computer Science and Engineering
East West Institute of Technology
Department of Computer Science and Engineering
East West Institute of Technology
Department of Computer Science and Engineering
East West Institute of Technology
Department of Computer Science and Engineering
East West Institute of Technology
Department of Computer Science and Engineering
East West Institute of Technology
Preparing an income tax return involves more than calculating tax. A taxpayer has to identify relevant income sources, understand deductions, select an appropriate return form, provide supporting documents, and interpret the final result. This paper presents a web-based tax preparation framework for Indian individual taxpayers that combines deterministic tax computation with conversational assistance and document-based information extraction. The system maintains a structured taxpayer profile and uses rule-based services for tax calculation and return selection, while the AI assistant is used for natural-language interaction and explanation. Assessment Year 2026–27 is currently supported, with return selection covering ITR-1 to ITR-4 within the implemented scope and explicit handling of unsupported conditions. The system also includes assessment-year-specific knowledge retrieval for the conversational layer. Uploaded documents are processed to produce candidate information that must be reviewed before being added to the taxpayer profile. The implementation uses Next.js and TypeScript on the frontend and FastAPI, Python, SQLAlchemy, and PostgreSQL on the backend. The design keeps numerical tax decisions separate from language generation so that calculations remain reproducible and easier to verify.
KEYWORDS
This paper presented a hybrid framework for personal-ized income tax preparation that combines deterministic tax computation, rule-based return selection, document-assisted information extraction, assessment-year knowledge retrieval, and conversational AI. The central design principle is to use AI where natural-language interaction is useful while keeping numerical tax decisions and return eligibility within explicit and testable software rules. The resulting system provides a structured workflow in which users can enter or extract tax information, review their profile, understand their tax calculation, identify an appropriate return form, and prepare a return document within the supported scope. The architecture also provides a foundation for extend-ing assessment-year rules, improving document intelligence, expanding return schedules, and strengthening knowledge-grounded conversational assistance.
[1] S. B. J., A. V. K., V. Daimane, C. Shetty, R. Patil, and S. Sethi, “TaxCraft: Machine Learning-Powered Automated Tax Assistant for Personalized Financial Empowerment,” in Proc. 3rd IEEE Int. Conf. Knowledge Engineering and Communication Systems (ICKECS), 2025, doi: 10.1109/ICKECS65700.2025.11035348. [2] R. Kumar, P. Devi, M. K. Mishra, and S. S. Chauhan, “Artificial Intelligence in Goods and Services Network (GSTN) Application: Scope of Implementation,” in Proc. 1st Int. Conf. Advances in Computer Science, Electrical, Electronics, and Communication Technologies (CE2CT), 2025, doi: 10.1109/CE2CT64011.2025.10941311. [3] R. Kumar, R. K. Malhotra, S. Pandey, A. Gehlot, I. Gautam, and S. Chamola, “Role of Artificial Intelligence in Input Tax Credit Reconciliation,” in Proc. 3rd Int. Conf. Pervasive Computing and Social Networking (ICPCSN), 2023, doi: 10.1109/ICPCSN58827.2023.00086. [4] P. Bharati, V. D. Hanchate, M. D, S. Gani, and S. L. Shakkeera, “The Secure Framework to Develop Income Tax Fraud Detection using AI-ML Techniques,” Journal of Emerging Technologies and Innovative Research, vol. 11, no. 1, pp. b1–b8, Jan. 2024. [5] N. Naveen Kumar, R. Sridhar, U. Uday Prasanna, and G. Priyanka, “Tax Management in the Digital Age: A TAB Algorithm-based Approach to Accurate Tax Prediction and Planning,” in Proc. Int. Conf. Inventive Computation Technologies (ICICT), 2023, pp. 909–913. [6] Y. Tan, B. Wu, J. Cao, and B. Jiang, “LLaMA-UTP: Knowledge-Guided Expert Mixture for Analyzing Uncertain Tax Positions,” IEEE Access, vol. 13, pp. 90637–90648, 2025, doi: 10.1109/ACCESS.2025.3571502. [7] Income Tax Department, Government of India, “Salaried Indi-viduals for AY 2026-27: Returns and Forms Applicable,” 2026. [Online]. Available: https://www.incometax.gov.in/iec/foportal/help/ individual/return-applicable-1 [8] Income Tax Department, Government of India, “File ITR-1 (Sahaj) Online User Manual,” 2026. [Online]. Available: https://www.incometax. gov.in/iec/foportal/newdownloads/itr/itr-UM [9] Income Tax Department, Government of India, “Downloads: Income Tax Returns,” 2026. [Online]. Available: https://www.incometax.gov.in/iec/ foportal/downloads/income-tax-returns [10] D. S., “TaxWise: AI-Powered Tax Filing Application,” GitHub repository, 2026. [Online]. Available: https://github.com/Diganths2005/aaa
Publication Date
10/2/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
1
Total views since publication
PDF Downloads
6
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...