User:Nkunkle/sandbox

My name is Nicole Kunkle and I am a Masters Student at the University of Pittsburgh.

Article:

The emergence of social media has provided sociologists with a new way of studying social phenomenon. Social media networks, such as Facebook and Twitter, are increasingly being mined for research. For example, Twitter data is easily available to researchers through the Twitter API. Twitter provides researchers with demographic data, time and location data, and connections between users. From these data, researchers gain insight into user moods and how they communicate with one another. Furthermore, social networks can be graphed and visualized.

Using large datasets, like those obtained from Twitter can be challenging. First of all, researchers have to figure out how to store this data effectively in a database. Several tools are at their disposal. Since large data sets can be unwieldly and contain numerous types of data (i.e. photos, videos, GIF images), most researchers choose to store their data in non-relational databases, such as MongoDB and Hadoop. Processing and querying this data is an additional challenge. However, there are several options available to researchers. One common option is to use a querying language, such as Hive, in conjunction with Hadoop to perform queries on large datasets.

The internet and social media have allowed sociologists to study how controversial topics are discussed over time--otherwise known as Issue Mapping. Sociologists can search social networking sites (i.e. Facebook or Twitter) for posts related to a hotly-debated topic and then parse through and analyze the text. Sociologists can then use a number of easily accessible tools to visualize this data, such as MentionMapp or Twitter Streamgraph. MentionMapp shows how popular a hashtag is and Twitter Streamgraph depicts how often certain words are paired together and how their relationship changes over time.