The Invisible bird
Nowadays, people spend tons of hours on social media per day, with or without noticing how their opinions and decision-making process have been shaped in an insidious way. In response to such phenomenon, we think it would be interesting to build a mechanism to remind people of the opinion bubbles they live in, by allowing them to physically interact with one of the social media behemoths. In our case, we decided to use Twitter as our target platform to make people think outside of their thought cages. The goal is to empower every one of us with an accessible and human-friendly tool to gauge and understand how we collectively think about the world and the issues around us, without dealing with the arcane, intimidating details behind most advanced technologies nowadays.
Of all the tools that have built the advancement of human society, words are by far the most powerful. Words can capture empathy, deliver joy or they can be destructive, used to create pain. Today, words overwhelm our networked society, flooding us in the form of 34 gigabytes of information each day. However, we can only absorb a sliver of that information when we create our opinions, thus keeping our perspectives narrow. We have applied a machine-learning algorithm called sentiment analysis to identify emotional patterns in the words we use online. We analyzed tweets written by people around the world, sentence by sentence and word by word, in hopes of creating a way for people to engage with different points of view, and even challenge their own from the positive to the negative. By showing the polarity of opinions before deciding what to believe as well as exposing to the invisible thought cages we’ve built around ourselves, we get to see just how long we’ve been trapped.
The majority of people seldom interacts with large public’s data, either because of the lack of appropriate tool, or because of the information bubble we’re unwittingly encapsulated in. Hence, we thought it would be meaningful to hand in the power of AI, especially those that are usually exploited by large corporates only, back to the hands of everyone, and observe whether it would break people out of the bubbles and change their understanding. We utilized the aggregated analysis data of Twitter’s attitude towards different topics overtime. In tandem of allowing people to interact with the data and guess the general stances, we summarize and update the sentiment analysis results on different topics regularly. Also, statistics regarding how people react to different topics is recorded, such as the total number of people interacted, overall correct rate of guessing for each topic, etc.