Social Network Analysis for Precise Friend Suggestion for Twitter by Associating Multiple Networks Using ML

Dharmendra Kumar Singh Singh, Nithya N., Rahunathan L., Preyal Sanghavi, Ravirajsinh Sajubha Vaghela, Poongodi Manoharan, Mounir Hamdi, Godwin Brown Tunze

Source Title: International Journal of Information Technology and Web Engineering (IJITWE)17(1)

ISSN: 1554-1045|EISSN: 1554-1053|EISBN13: 9781799894001|DOI: 10.4018/IJITWE.304050

Cite Article Cite Article

MLA

Singh, Dharmendra Kumar Singh, et al. "Social Network Analysis for Precise Friend Suggestion for Twitter by Associating Multiple Networks Using ML." IJITWE vol.17, no.1 2022: pp.1-11. http://doi.org/10.4018/IJITWE.304050

APA

Singh, D. K., Nithya N., Rahunathan L., Sanghavi, P., Vaghela, R. S., Manoharan, P., Hamdi, M., & Tunze, G. B. (2022). Social Network Analysis for Precise Friend Suggestion for Twitter by Associating Multiple Networks Using ML. International Journal of Information Technology and Web Engineering (IJITWE), 17(1), 1-11. http://doi.org/10.4018/IJITWE.304050

Chicago

Singh, Dharmendra Kumar Singh, et al. "Social Network Analysis for Precise Friend Suggestion for Twitter by Associating Multiple Networks Using ML," International Journal of Information Technology and Web Engineering (IJITWE) 17, no.1: 1-11. http://doi.org/10.4018/IJITWE.304050

Export Reference

Favorite Full-Issue Download

View Full Text HTML

View Full Text PDF

Abstract

The main aim in this paper is to create a friend suggestion algorithm that can be used to recommend new friends to a user on Twitter when their existing friends and other details are given. The information gathered to make these predictions includes the user's friends, tags, tweets, language spoken, ID, etc. Based on these features, the authors trained their models using supervised learning methods. The machine learning-based approach used for this purpose is the k-nearest neighbor approach. This approach is by and large used to decrease the dimensionality of the information alongside its feature space. K-nearest neighbor classifier is normally utilized in arrangement-based situations to recognize and distinguish between a few parameters. By using this, the features of the central user's non-friends were compared. The friends and communities of a user are likely to be very different from any other user. Due to this, the authors select a single user and compare the results obtained for that user to suggest friends.