WebIntelligence final project in collaboration with Pau Nonell. Data obtained directly from both Twitter and Youtube APIs.
In the context of Web Inteligence final project, both me and Pau were really interested in understanding the affectance that real-events could have in social network user behaviors. Having both of us lived in Taiwan for a while and considering that Taiwan is a really interesting geopolitical hot location, Taiwan was chosen as the main topic of the project.
Taiwan is considered as a geopolitica hot spot as it presents a de-facto independent status, but its not accepted as an independent country worldwide. This has leaded this little island of the southeast Asia region to such a paradoxical scenario: It is not considered as an independent country on its own (only 11 minor countries worldwide recognize it as an independent country), however, it presents a total independence from china (economically, politically and socially).
- This has made us wonder what can we obtain from social networks regarding this political conflict.
- Some extra studies will be performed to see how real-life events influence social networks and correlate with its users behavior.
To understand this behavior, some libraries have been used to detect the positivity of tweets and the polarity of the sentiments associated to them (such as NLTK -Natural language toolkit- or textblob). Some interesting conclusions came out as outputs of this project:
- The social media analysis have confirmed that Taiwan reactions are highly related with China and their political conflict. However, many other interactions with real time successes have been observed as well, such as the Myanmar political situation.
- Real Time-events present immediate consequences and influences on social networks. This has been observed in several different political events that took place during the period of study and presented inmediat repercussion on twitter.
- Random events such as earthquakes or the political bans on some specific goods exports can be monitorized using social networks.
- When some specific topic suddnely increases its presence on social networks, usually it is an indicator that something related to it has happened. Furthermore, worthy information can be directly obtained when observing the interaction of users with such event on social networks.
- The main conclusion from the Youtube analysis is thatt the more subjective a comment is, the more polarized it is as well.