Grasping the emotions people display through text can be really important data for the companies they address. We found this really interesting and saw a lot of potential in using this program to help companies ease their social media presence. We also really wanted to learn and implement google cloud APIs and use beautiful soup for web-scraping so this project seemed like a perfect fit.
What it does
Our program collects data from websites about customer's experience with flights and uses natural language processing to determine their sentiment towards the airways they used.
How we built it
We used beautiful soup to scrape data from websites and then used google's API for natural language processing to retrieve the sentiment value correlated to the site's data. We then implemented flags to retrieve more specific data about the customer's experiences.
Challenges we ran into
We did not have the right update for python when downloading the API from Google. This wasted a lot of our time as it was not clear that this was the reason we were not being able to proceed.
Accomplishments that we're proud of
We are content that we could integrate web scraping effectively with our NLP commands.
What we learned
We learnt that understanding and using a API at first may not be easy but after using it a few times, the process becomes significantly easier!
What's next for Sentiment Scraper
We hope to contact Facebook and Twitter so we can access their APIs and retrieve data from them too. Additionally, we want to include more flags that the user can choose when running our program - allowing them to access more information that the NLP provides.