It was difficult to find exact answer to our query while studying from notes or webpages. Finding precise information while reading news, collecting data for presentations and research purposes was tedious task.

What it does

AI will help to solve above mentioned problems. AIP-S³ is based on Information Retrieval concept. It helps users to find appropriate information from multiple PDF files and multiple web page links. User can upload PDFs or links of website and then user will ask query, AIP-S³ will retrieve information from files or webpages related to user's query.

How we built it

  • Programming language -Python
  • Preprocessing - NLTK and SpaCy
  • Demonstration - Streamlit

Challenges we ran into

  • Extraction of textual data (Textraction) from PDFs and webpages.
  • Function for Semantic Similarity.
  • Reducing processing time.

Accomplishments that we're proud of

We're proud of - we successfully overcome all of the challenges that came across and modified our model into its flawless working version and this project will be useful for getting useful insights for huge amount of textual data!

What we learned

  • Real life application of NLP techniques.
  • Semantic Similarity.
  • Web Scraping.

What's next for AI Powered - Smart Search System (AIP-S³)

  • AIP-S³ will be available as extension for browsers so user can ask question on webpages and files itself.
  • AIP-S³ will have more preprocessing steps to increase accuracy of retrieval.

Built With

Share this project: