Inspiration
We want drip. You want drip. We all want drip. But you know what would be more epic than having drip? Having s u s t a i n a b l e drip. With the growing demand of obtaining drip, fast fashion companies have been producing massive volumes of clothing every year. However, this massive overhaul in production has led an abundance of textile waste, environmental pollution, and the exploitation of many workers. To help shoppers become more aware of these issues, we've developed a Chrome extension that can do research for the shopper and keep them informed of a company's decisions regarding sustainability and social issues.
What it does
Juniper empowers online shoppers to make environmentally and socially conscious purchases. Our extension gathers articles about a company's social/environmental impact and analyzes them with Google Cloud's Natural Language Processing AI to deliver consumers immediate and actionable insights. Furthermore, we highlight the specific areas where a company is unethical, and we offer more sustainable and socially conscious alternatives for your purchase.
How we built it
- Wrote a web-scraping script to gather relevant articles regarding a specific company's environmental and social impacts
- Fed results into Google Cloud's NLP AI to perform sentiment analysis and categorization on the articles' content
- Developed a Chrome extension that automatically recognizes the company of interest and performs the web-scraping, sentiment analysis, and categorization
- Designed UI that displays the information concisely while still highlighting all important insights
Challenges we ran into
- NLP ran too slow synchronously, worked out how to run it asynchronously to improve performance
- Originally used News API to get articles, realized that the API was not generating a robust set of articles for our NLP, shifted to our own web-scraping approach which yielded much better results
- Being able to combine all individual scripts/program into one seamless process
- Never used NLP, Node.js, or Google Cloud
- Never developed Chrome Extension
- Very unfamiliar with JavaScript
Accomplishments that we're proud of
-It kinda works :)
What we learned
- Learned about asynchronous processing for NLP
- Integrating JavaScript and Python
- How to use Google Cloud NLP
- Basic NLP Principles
- Building a basic server
What's next for Juniper
- Currently focusing on fashion industry, however concept is applicable to more general product space
- Looking to create a more robust alternatives system
- Optimize code efficiency
Log in or sign up for Devpost to join the conversation.