Title: RecomAlly: dynamic ally recommendation on Twitter based on rhetorical structure theory and valence shifters

Authors: A. Sai Charan; Vegesna S.M. Srinivasavarma; Rajesh Eswarawaka

Addresses: Chennai, India ' Department of Computing Technologies, SRMIST, Chennai, India ' Department of AI and ML, AMC Engineering College, Bangalore, India

Abstract: Microblogs like Twitter emerged as a significant means for instantaneous information sharing on the web and forming the communities of similar interests by recommending the correlated information like tiny URL's, hashtags, friends or ally's, etc. to the target users. Existing systems do not capture the dynamic change in the user's interest over the time while recommending the potential user to the target user. Also, while computing the content similarity of the tweets, the existing systems just instinctively considers the total number of matching words in the two tweets, without considering the semantic similarity or discourse relation between them. In this work, we propose a dynamic and personalised ally recommendation system that computes the user's interests dynamically and considers the larger matching of semantic orientation of tweets on common topics which are computed based on content similarity of tweets using rhetorical structure theory.

Keywords: Twitter ally recommendation; rhetorical structure theory; RST; valence shifters; recommendation; hashtag similarity.

DOI: 10.1504/IJESDF.2023.131958

International Journal of Electronic Security and Digital Forensics, 2023 Vol.15 No.4, pp.348 - 358

Received: 30 Jun 2022
Accepted: 05 Oct 2022

Published online: 05 Jul 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article