OLAP operators for social network analysis
- PDF / 6,521,565 Bytes
- 28 Pages / 595.276 x 790.866 pts Page_size
- 18 Downloads / 218 Views
(0123456789().,-volV)(0123456789().,-volV)
OLAP operators for social network analysis Maha Ben Kraiem1
•
Mohamed Alqarni2 • Jamel Feki2 • Franck Ravat3
Received: 4 April 2019 / Revised: 12 September 2019 / Accepted: 19 October 2019 Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract The multidimensional data model and implementations of social networks come with a set of specific constraints, such as missing data, reflexive relationship on fact instance. However, the conventional OLAP operators and existing models do not provide solutions for handling those specificities. Therefore, we should invest further efforts to extend these operators to take into consideration the specificities of multidimensional modeling of tweets as well as their manipulation. Face to this issue, we propose, in this paper, new OLAP operators that enhance existing solutions for OLAP analyses involving a reflexive relationship on the fact instances and dealing with missing values on dimension members. For each OLAP operator, we suggest a user-oriented definition as an algebraic formalization, along with an implementation algorithmic. Keywords Social network analysis OLAP operators Null-Drilldown Null-Rollup Null-Select FDrilldown FRollup
1 Introduction In the last decade, the data warehouse (DW) has been the backbone of decision support systems for more than 2 decades and widely accepted and used across the globe in a variety of applications. Contributions of the research community in the data-warehousing field, complimented by advancement in the relevant hardware technology, have matured these systems in managing huge volumes of data and providing their access with matchless efficiency to applications and decision-makers.
& Maha Ben Kraiem [email protected] Mohamed Alqarni [email protected] Jamel Feki [email protected] Franck Ravat [email protected] 1
MIRACL Laboratory, University of Sfax, Airport Road Km 4, P.O. Box. 1088, 3018 Sfax, Tunisia
2
University of Jeddah, CCSE, Jeddah, Saudi Arabia
3
IRIT, University of Toulouse, 118, Route de Narbonne, 31069 Toulouse Cedex 9, France
Data warehousing has proven its success on structured data and pre-established relationships among this data, thereby achieving less performance in dealing with huge volumes of data. The OLAP technology in a data warehouse performs aggregated-oriented analyses from multiple dimensions of interest. Social media is yet another interesting domain producing much more large data volumes that trap the attention of research and business communities. There is growing interests in gaining insights to the way social networks operate, their users behave, engage in conversations, express their opinions and influence others. This involves performing aggregations across conventional and unconventional dimensions in social media data. Furthermore, businesses can largely benefit from this new resource and market of social media. Provided underlying technology and systems of data warehousing can partially solve the challenge
Data Loading...
