Product

Our final product is a thermometer that can be hung up on a wall and felt throughout the day to learn the current Twitter sentiment of the displayed politician. The python script will randomly choose a politician from a pre-programmed list at a given time of day, currently it is 12:00 pm, and find the current sentiment of the chosen politician. In the next iteration the user would be able to change the list of politicians that are cycled through and also set what time the politician will be chosen. As mentioned before, the Sentiment140 API was the planned API because of it's capability to analyze multiple tweets at one time using a twitter specific sentiment analysis algorithm but we moved to the DatumBox API because the Sentiment140 API was not reliable. The DatumBox API was not as robust as we were not able to analyze multiple tweets at a time forcing us to use the basic sentiment analysis rather than the twitter specific. We used the Maker IFTT API to send the sentiment value and the name of the politician to the photon board. The casing was designed using SolidWorks and 3D printed using the NVbots printer. We designed the casing in two parts, the face and the base, in order to allow us to open the case and access the circuit if there was ever a need. In order to mimic the red mercury of an actual thermometer the face contains space for three red LEDs that light up when the sentiment is positive and turn off when the sentiment is negative. A screen was also added when we made the decision to rotate through a list of politicians so that a user would know what politician was being analyzed.