We wanted to learn and use about APIs at our time here at AthenaHacks. While using Twitter, we thought of an idea that we could use on a daily basis.
What it does
Our project uses the Twitter API to allow a user to filter out their feed by specific keywords or categories(tweets/retweets). It also allows people to find the most commonly used keyword in their feed.
How we built it
We obtained tweets from the Twitter API by use of a Twitter screen name and then used Python as we felt it worked best with our direction. After accessing information about the tweets through the API, we used various data structures, mainly lists (or arrays), to parse through all the tweets.
Challenges we ran into
We had trouble figuring out what topic to dive into as we had various ideas however the platforms we needed weren't available to us for use. In addition, we had trouble working with the various fields of the tweet object from Twitter as we weren't very comfortable with various web APIs. In addition, we needed to adjust the tweets due to the emojis integrated in the wording that weren't compatible with our IDE. In order to fix this, we used the ASCII table and account for letters/images that were not in the table.
Accomplishments that we're proud of
We're proud that we were able to access and parse through the data from the Twitter API while also adjusting our project goals when stuff didn't necessarily work the way we wanted it to.
What we learned
We learned to better plan out an idea for a project before coming to a Hackathon, such as downloading and having an understanding of the technologies we would be working with. We also learned how to access information from API's that do not involve JSON file formatting. Furthermore, we learned that different APIs have different fields and parameters, which must be researched before diving right into the code.
What's next for CtrlTwitter
We would like to make better use of the tweet object and find a way to check date/time for tweets pertaining only to a user and their friends, as opposed to all of Twitter. We also would add location as a possible parameter for filtering out tweets.