Inspiration
The inspiration for this project was to connect myself with the community more. Due to the difficulty to find authentic local businesses especially during the COVID-19 pandemic where it is harder to support these businesses.
What it does
The algorithm determines the key words and compares it to the catalog present. The business whose values align with your search are the ones that are present. The program picks up words and sentences such as "local vegan restaurant" and will find you the best match. This will not only make it easier to find businesses in your communities but also follow good consumer practice.
How I built it
The system takes descriptions from all the restaurants in the catalog and compares it to the description in the search using count vectorizer. The item with the highest index is the one that will be presented. The html front end search bar sends the value to the JavaScript and then to the python backend. The python backend catalogs the businesses and sends it back to the JavaScript and then html.
Challenges I ran into
The biggest challenge was the front end design of exchanging information from python to JavaScript and back into a formattable html. This was solved using the eel library in order to transport functions between languages.
Accomplishments that I'm proud of
The thing that I am proud of the most was making the python backend spit out the correct business, before, the system would just say some random business. The system was later correctly built so it could output the right businesses.
What I learned
I learned that, 2 things, 1. front end is not as complicated as it seems and there are a lot of resources for these things and 2. that things that seemed impossible at first quickly become possible with hard work.
What's next for test
The next thing for Local Business Finder is to improve the front end and back end design. The front end could be exported to multiple platforms such as mobile and making the platform much more accessible for people. The back end will be optimized to find better results as well as make the process of the local finding automated which was not done during the time of this hackathon.
Built With
- css
- dreamweaver
- html
- html5
- javascript
- nltk
- pycharm
- python
- sklearn




Log in or sign up for Devpost to join the conversation.