2017 IEEE/ACS 14th International Conference on Computer Systems and Applications (AICCSA) (2017)
Oct. 30, 2017 to Nov. 3, 2017
Social networks such as Facebook, LinkedIn, Twitter, YouTube, etc. have revolutionized the ways of interacting and exchanging information on the Internet. They provide the users with the ability to interact frequently and share variety of digital content with each other. Unstructured data exchanged over the social networks represent an important source of information, even it their characteristics make them difficult to analyze. Online analytical processing (OLAP)is a powerful primitive for social media data analysis. However, OLAP tools face major challenges in manipulating unstructured data generated from social media. In this paper, we attempt to extend the established OLAP technology to allow multidimensional analysis of social media unstructured data. We suggest a new multidimensional model called "Unified Multidimensional Social Media Data Model" dedicated to the OLAP of social media and specially Twitter, Facebook and YouTube. Our model is generic and dynamic, that is not limited to a set of determined social networks. It can be extended by other social networks or with other enrichement data using dynamic discovery of multidimensional concepts. It provides the possibility to analyze social networks users and their published posts and comments according to device source, geographic locations and temporal axes.
data analysis, data mining, data models, social networking (online)
H. Dabbechi, N. Haddar, M. B. Abdallah and K. Haddar, "A Unified Multidimensional Data Model from Social Networks for Unstructured Data Analysis," 2017 IEEE/ACS 14th International Conference on Computer Systems and Applications (AICCSA), Hammamet, Tunisia, 2018, pp. 415-422.