SWQL: A new domain-specific language for mining the social Web

Abstract

Social media generate a massive amount of information each day. This information is usually generated by people and may be used for many studies. Social media mining is a growing discipline inside data mining, and the results have proven to be quite revealing. Getting the data to work with might seem to be the easiest task in the process, but it can be very challenging for people without programming knowledge. Thus, there are numerous ways to extract information from social media, such as to use the network API, a Web scraper or specialized libraries. Using a straightforward and a cross social media query language we can hide the complexity of those mechanisms and gather information in a more efficient and easier way. This is because many social media share common elements, so we can create and unify queries to search, find and extract information from those platforms. In this paper, we propose a domain-specific query language, specially designed to allow developers or domain experts to extract data from different social media. With this language we unify and simplify the mechanisms of data extraction from social networks such as Twitter and Facebook. • A DSL specially designed to allow developers or domain experts to extract data from different social media. • The DSL unifies and simplifies the different mechanisms to extract data from social networks. • The DSL facilitates the generation of large datasets that can be reused by researchers in different fields. • The DSL serves as the basis for improving data mining and Big data process.