Vibes

Inspiration

Our main aim was to use AI with a social purpose. We wanted to get close to people in a simple way, so we thought that we could approach an AI utility to something as common as Twitter.

What it does

It implements Everis AI ('everisMoriarty') to estimate the "positivity" or "negativity" of a certain Twitter user or hashtag, based on the lastest tweets he or she wrote or was mentioned at. We thought it was quite interesting to see the popularity and social acceptance of certain users, as well as harassing behaviours. Also, the most promising application which we have in mind is to study the positivity of different hashtags in order to see how the population reacts to it, this may be usefull when we are trying to search for good social policies . This extension has also great potential as we can restrict the search using Twitter's API with options like selecting especific localizations to make better fitting social policies in those places, selecting specific range of ages or languages to select smaller samples and therefore make more specific "models" for them .

How we built it

We developed a Chrome extension for Twitter based on Javascript. The code uses the given API from Everis, the Twitter API, and other implementations specifical for extensions (Chrome Platform API's). We used the sentiment recognition from 'EverisMoriarty' implemented in javascript to get 'Vibes'

Challenges we ran into

We had serious trouble implementing the extension in Chrome due to lack of desired packages. We overcame this using the Browserify to implement everything in one Bundle. In addition, we had some issues due to the sentiment recognition of the Tweets as it was quite slow and if many Tweets were analized it wouldn't be a functional extension.

Also, we had serious trouble implementing a configuration menu for our extension.

Accomplishments that we're proud of

We overcame many challenges during the building of this extension. We tried our best and, in general, we made what we wanted to do. We are proud of building a Chrome plug-in that is able to read Tweets from desired people or hashtags and then give a rating or "vibe". Also, we are very proud of doing so because we started this weekend with basic knowledge on javascript and it was the first time we were using any API.

What we learned

We learned to implement an extension in Chrome using javascript. We learned to use javascript and html at a basic level. Also, we learned to modify an external 'html' with the implementation of the mentioned plug-in. Finally, we learned to use several API's in order to make more functional and try to extract the maximum profit of 'Vibes'.

What's next for Vibes

We are looking forward to implement a configuration menu to 'Vibes' in order to make more specific Tweet searchs to adress what we think might be a good way to extract data from the society in order to stop racism, online harassment and even to do more easy to make specific social policies that adress people in need for it. It can also be used to track the "popularity" of a certain Twitter user as a function of time and therefore with a big enough data set we could estimate which declarations have made the rest of Twitter to react the way the made.

Built With

• javascript