ABSTRACT
The exponential rise in usage of social area network platforms such as Facebook, Twitter, LinkedIn and e-mail make connections between people demand huge amount of data. Millions of social media users generate huge amounts of data every day. The collection and maintenance of data from different social sources are not an easy task. Machine learning (ML) concepts can help to automate, collect, maintain and manipulate the data without human intervention. The role of ML on social media services include security, branding to target the audience, media quality enhancement, data handling and automation.
It is pertinent to study the interaction of ML techniques in social network analysis and to focus on practical applications such as sentiment analysis, health diagnosis, persuasive identification and interpersonal relationships. The above-mentioned applications are implemented by using rule-based algorithms (RBA).
RBA is a learning classifier system that acts as a supervised ML system model for social media analysis. RBA is a unique algorithm and it has a flexible set of features to solve the problems in ML. The key idea of rule-based system is based on the human experts' knowledge in a specific domain and transferred into the computer system through coding. The rules are framed as if-then-else statements. Thus, ML plays an important role in social area network analysis with the help of RBA.
