As a standard method, surveys were used to collect large empirical data on the field of social sciences in the last 50-70 years. However, surveys are not the only way to gather large amount of data. One of the new data sources is the digital footprint of the society. Social media is one of the most interesting data source of the digital world.
Facebook (FB) is the biggest social media site in the word and it is growing continuously. There were over 2.07 billion monthly active FB users in 2017 (Q3), and 1.37 billion daily active users. In every minute 510 000 comments are posted and around 300k status are updated. An extremely huge amount of data arises continuously but it is rarely exploited by social scientist. To change this, we have to solve two problems: how to collect social media data, and how to analyse it effectively.
Collection of social media data is hard, especially in the case of Facebook. Our project will take an initial step here, by conducting an experimental FB study. Based on this study we could answer many important methodological questions which are inevitable if we would like to use FB as a social science research tool.
It is also not straightforward how to analyse FB data effectively. Here we have to deal with a huge amount of unstructured text data. To understand this data, we need to use the newest text analytics methods, and we need to develop new techniques and analytics strategies to reach our objectives. Our project will make future contribution on this field too.
The reserach is supported by NKFI, unnder the grant agreement: 12899. The principal Investigator of the research is Dr. Zoltán Kmetty.