Natural language querying on business documents:

This is placeholder text. To change this content, double-click on the element and click Change Content.

The system should be able to not only show relevant documents but also create an effective summary of results. Following steps are involved in building the application

1. Effective extraction of sections and tables from pdf files

2. Extracting metadata using NLP

3. Fine tuning language models like bert etc on domain specific data for QA

4. Summarization models for creating search summaries

5. Table QA- Answer queries on tables

6. Performance tuning for model inference. Response time should not exceed 2 sec.

7. Show results with previews on UI

Power in Numbers

Academic Concentration

Bachelor's/ Masters

Project Area

Python, NLP, ML, React, Pytorch/Tensorflow

(in weeks)