Computers and Electrical Engineering 86 (2020) 106723 Contents lists available at ScienceDirect Computers and Electrical Engineering journal homepage: www.elsevier.com/locate/compeleceng A machine learning-based approach to enhancing social media marketing  B. Senthil Arasu, B. Jonath Backia Seelan ∗ , N. Thamaraiselvan Department of Management Studies, National Institute of Technology, Tiruchirappalli, India a r t i c l e i n f o Article history: Received 19 December 2019 Revised 28 May 2020 Accepted 28 May 2020 Keywords: Social media marketing Social analytics Artificial intelligence Information sciences Machine learning and WEKA a b s t r a c t Social media (SM) represent beneficial channels for marketers, business promoters and consumers. To acquire continuous revenues and more active customers, key business play- ers should understand the behaviour and purchase preferences of buyers. To predict the buying decisions of purchasers, data about purchase intentions and desires have to be ex- tracted with the help of data mining techniques. The purpose of this paper is to examine social media data analytics using machine learning tools; this new approach for developing a social media marketing strategy employs the Waikato Environment for Knowledge Analy- sis (WEKA). WEKA is compared with other algorithms of interest and found to outperform its peers, especially with regard to parameters such as precision, recall, and F-measure, indicating that WEKA performs better than other approaches. © 2020 Elsevier Ltd. All rights reserved. 1. Introduction The internet is a dominant marketing tool, and it can be used to attract customers, build trustworthiness and extend a product or service’s brand [1]. SM offer platforms where people communicate and collaborate virtually. Users’ thoughts are controlled and influenced by frequent advertisements that they come across on various micro blogging and social media platforms [2]. Business analysts use SM for business exploration, corporate knowledge gathering, and product awareness. The current number of social media users is increasing every day due to their varied browsing interests [3]. Fig. 1 shows how individuals consider social media when making buying decisions. This paper mainly focuses on the possible ways to leverage marketing via SM using various available machine learning techniques to predict customer purchase preferences. The remainder of this paper is organized as follows. Section II presents an overall literature review on the trends in SM plat- forms, and Section III further elaborates the study report on the social data analysis. Section IV describes the proposed ML integrated social media marketing (ML-SMM) approach and analyses its implementation and performance. Finally, Section V concludes the paper by describing the potential advantages and applications of the proposed ML-SMM approach. 2. Literature review Many marketers prefer to use artificial intelligence (AI) to transform data into valuable customer insights. Information gathering is an art [4] that involves identifying the benefits of online marketing for improving information gathering and  This paper is for regular issues of CAEE. Reviews processed and recommended for publication to the Editor-in-Chief by Associate Editor Dr. S. Smys. ∗ Corresponding author. E-mail addresses: arasu@nitt.edu (B.S. Arasu), jonathbaskaran@gmail.com (B.J.B. Seelan), selvan@nitt.edu (N. Thamaraiselvan). https://doi.org/10.1016/j.compeleceng.2020.106723 0045-7906/© 2020 Elsevier Ltd. All rights reserved.