Chemical Search

This minimalistic website seamlessly checks if HP's internal database contains the chemicals that potential customers search, and allows analytics with the search user's pattern.

Comment 1

Inspiration

Google's search engine inspired us to create this minimalistic search for chemicals.

What it does

This application matches chemicals searched by the user to HP's internal database and updates the number of times each chemical is searched. It also creates a database for chemicals that have been searched, but the exact search did not exist in HP's internal database.

How we built it

We built it using React as our front-end, Django as our back-end and MS SQL as our database. Queries from the website would be translated to queries on MS SQL to return the results for the user. We used flask to create the link for the website to access the database.

Challenges we ran into

To create the internal database, we struggled to obtain the tables from the entire database, and there were a lot of formatting issues but managed to fix the issue using the right python libraries to convert PDF to Tables.

Accomplishments that we're proud of

Managed to put together an entire website for the first time!

What we learned

Each of us picked up new programming languages during this project and learnt how to bring them all together. We learnt to focus on our strengths and integrate different programming concepts to create a single website.

What's next for Chemical Search

Chemical Search will be able to accept input of any type from the users, by creating a list from the file input. The search may then carry on per normal using the list. Furthermore, there will be an analytics page using chart.js for users to view trends of the chemical queries made previously. The analytics would be as follows:

  1. Line graph to denote the number of searches over time
  2. Number of searches over the past 6 months of the current top 3 most searched chemicals.
  3. With further knowledge of chemicals, we would be able to group the chemicals into different relevant categories during the data cleaning part of the internal data (polyesters, etc). We think it would be interesting to have a pie-chart that shows the number of searches for chemicals in each category, rather than a pie-chart with individual chemicals.

Built With

Share this project:

Updates