The rapid growth of Social Media Networks (SMN) initiated a new era for data analytics. We use various data mining and machine learning algorithms to analyze different types of data generated within these complex networks, attempting to produce usable knowledge. When engaging in descriptive analytics, we utilize data aggregation and mining techniques to provide an insight into the past or present, describing patterns, trends, incidents etc. and try to answer the question “What is happening or What has happened”. Diagnostic analytics come with a pack of techniques that act as tracking/monitoring tools aiming to understand “Why something is happening or Why it happened”. Predictive analytics come with a variety of forecasting techniques and statistical models, which combined, produce insights for the future, hopefully answering “What could happen”. Prescriptive analytics, utilize simulation and optimization methodologies and techniques to generate a helping/support mechanism, answering the question “What should we do”. In order to perform any type of analysis, we first need to identify the correct sources of information. Then, we need APIs to initialize data extraction. Once data are available, cleaning and preprocessing are performed, which involve dealing with noise, outliers, missing values, duplicate data and aggregation, discretization, feature selection, feature extraction, sampling. The next step involves analysis, depending on the Social Media Analytics (SMA) task, the choice of techniques and methodologies varies (e.g. similarity, clustering, classification, link prediction, ranking, recommendation, information fusion). Finally, it comes to human judgment to meaningfully interpret and draw valuable knowledge from the output of the analysis step. This chapter discusses these concepts elaborating on and categorizing various mining tasks (supervised and unsupervised) while presenting the required process and its steps to analyze data retrieved from the Social Media (SM) ecosystem.
CITATION STYLE
Koukaras, P., & Tjortjis, C. (2019). Social Media Analytics, Types and Methodology. In Learning and Analytics in Intelligent Systems (Vol. 1, pp. 401–427). Springer Nature. https://doi.org/10.1007/978-3-030-15628-2_12
Mendeley helps you to discover research relevant for your work.