Uncorrected Author Proof Journal of Intelligent & Fuzzy Systems xx (20xx) x–xx DOI:10.3233/JIFS-201036 IOS Press 1 Finding influential users in social networks based on novel features & link-based analysis 1 2 3 Shahid Iqbal a , Hikmat Ullah Khan a,∗ , Umar Ishfaq a , Muhammad Alghobiri b and Saqib Iqbal c 4 a Department of Computer Science, COMSATS University Islamabad, Wah Campus, Wah Cantt, Pakistan 5 b Department of Information Systems, College of Computer Science, King Khalid University, Abha, Saudi Arabia 6 c Department of Software Engineering, Al-Ain University of Science and Technology, Al-Ain, UAE 7 Abstract. The social web appears to enrich human lives by providing effective applications for online social interactions. Microblogs are one of the most important applications of the social Web. The Microbloggers who influence the social com- munity users through their content in the form of tweets are known as the influential microbloggers. The identification of such influential microbloggers has vast applications in advertising, online marketing, corporate communication, informa- tion dissemination, etc. This paper investigates the problem of identifying influential microbloggers by proposing MIPPLA (Model to identify Influential using Productivity, Popularity and Link Analysis) model which integrates the modules of Pro- ductivity and Popularity. The Productivity module considers a micro-blogger’s activity and the Popularity module identifies a microbloggers influence in an online social community. In addition, we modify the classic PageRank by utilizing the Twitter features such as retweet, mention, and reply for ranking the influential users. The proposed approaches are evaluated using real-world social networks. The results prove that the MIPPLA model efficiently identifies and ranks the top influential users in an effective manner as compared to the existing techniques. 8 9 10 11 12 13 14 15 16 17 18 Keywords: Social web, online social networks, microblogs, influential users, big data, data mining 19 1. Introduction 20 The social web, also known as Web 2.0, refers to 21 social media platforms that people use to interact with 22 each other. The social web platforms include blogs, 23 forums, wikis including e-commerce portals. Statis- 24 tics show that more than 4.15 billion people are using 25 internets across the globe. It is expected that there will 26 be approximately 2.77 billion social network users 27 around the world in 2019, up from 3.19 billion in 2018 28 [1]. Social network analysis is a study of exploring 29 social structures using network and graph theories. 30 ∗ Corresponding author. Hikmat Ullah Khan, Department of computer science, COMSATS University Islamabad, Wah Cam- pus, Pakistan. E-mail: hikmat.ullah@ciitwah.edu.pk. Such graphs describe network structures in term of 31 nodes and the connections between the nodes, known 32 as edges. 33 Nowadays, online social networks such as Twitter, 34 Facebook, Google + have become part of our daily 35 life for interaction as well as sharing our views and 36 opinions. Microblogging is a short form of blogs 37 and messages. A user can share limited content 38 such as text messages, pictures, video, or audio to 39 friends/followers. The widely used microblogging 40 sites include Twitter, Instagram, Tumbler, etc. Twitter 41 is one of the most famous microblog services which 42 provides free social networking platform to more than 43 330 Million active users per month contributing 500 44 Million tweets per day [2]. A Twitter user can pub- 45 lish text messages, called tweets of length 140 since 46 ISSN 1064-1246/20/$35.00 © 2020 – IOS Press and the authors. All rights reserved