IEEEnspired

👉 Easy. Efficient. Excellent. 👈

Comment

✍️ IEEEspiration

Tired of scrolling through lengthy documents and straining your eyes to find particularly long chemical names you can't pronounce? Then IEEENspired is the creative solution that is here to help!

Stemmed from the need to streamline documentation for hazardous chemicals from various clients.

❓ What it does:

✔️ 1. Generates a lookup table from the most updated HP General Specification for Environment (GSE) document online.

✔️ 2. Bundles it together with a search engine for easy lookup of chemical names and their data.

✔️ 3. Presents all relevant data as a text for easy copy-pasting onto the excel-sheet. (No more screenshots!)

❤️ How we built it:

With love using Python3 & NodeJS.

Check out the GitHub repo here!

🥺 Challenges we ran into:

Translating the client's document was tough since we did not have access to Google's Translation API and other open-source solutions such as LibreTranslate fell short in translating the chemical compounds accurately.

😊 Accomplishments that we're proud of:

  1. Making use of OCR techniques to process a lengthy PDF document.
  2. Structuring the data for easy reference through the use of Elastic Search.

📘 What we learnt:

  1. How to convert a PDF document with tables into an acceptable format such as CSV/JSON.
  2. How to lookup and find the latest PDF document given a specific title.
  3. How to use Elastic Search to allow the end-user to more easily find the chemical compound name.

🤔 What's next for IEENspired?

We wish to further build upon this model if we do receive a source of funds to leverage the Google Translate API and fully automate the process of:

  1. Uploading the client's document.
  2. Translating the text in-place.
  3. Using PyTesseract/Other OCR methods to extract out the data from all tables within the sheet.
  4. Pushing them onto an excel sheet.
  5. Using the names in the cells, call on our program IEEEnspired to automatically fill in the adjacent cell with all relevant data.
  6. Export everything as a single excel sheet for the client.
  7. Profit!

Built With

Share this project:

Updates