Skip to content

Latest commit

 

History

History
29 lines (25 loc) · 2.17 KB

File metadata and controls

29 lines (25 loc) · 2.17 KB

Datasets

Tabular dataset: Financial information (CSV format)

  • tax_data.csv: This is a structured dataset containing financial records for various entities. It includes columns for:
    • Taxpayer Type (e.g., Non-Profit, Partnership, Individual, Corporation).
    • Financial Metrics (Income, Deductions, Taxable Income, Tax Owed).
    • Contextual Data (State, Tax Year, Transaction Date, Income Source, Deduction Type).

Unstructured financial reports: (PDF format)

  • i1040gi.pdf: This is the official IRS Instructions for Form 1040 (and 1040-SR) for the 2023 tax year. It contains:
    • Detailed filing requirements and standard deduction tables.
    • Line-by-line instructions for completing the federal income tax return.
    • Specifics on tax credits, such as the Child Tax Credit and Earned Income Credit.
    • Administrative information, including IRS office addresses for mailing returns.
  • usc26@118-78.pdf: This document contains the text of Title 26 of the United States Code, also known as the Internal Revenue Code. It includes:
    • Legal statutes governing federal tax law.
    • Comparison tables relating provisions of the 1954 Code to the 1986 Code.
    • Statutory notes and amendments, such as references to the Health Care Portability and Accountability Act.

Financial presentations: (PPT format)

  • MIC_3e_Ch11.ppt: This is a PowerPoint presentation, likely Chapter 11 from a financial textbook or report. Based on your task description, it serves as the "Financial presentation" data source. Note that the preview text for this file is currently garbled, which is common for binary presentation files, but it is intended to provide the unstructured presentation data for your chatbot.

Task:

Create a working demo of a chatbot that can:

  • Ingest and process the provided datasets.
  • Perform semantic search across all data sources.
  • Answer user queries about financial information contained in the datasets.
  • Implement a solution that combines vector-based and graph-based approaches for efficient information retrieval and relationship understanding.
  • Utilize any free, publicly available tools and libraries of your choice.