Twigraph: Discovering and Visualizing Influential Words between Twitter Profiles

📝 Abstract

The social media craze is on an ever increasing spree, and people are connected with each other like never before, but these vast connections are visually unexplored. We propose a methodology Twigraph to explore the connections between persons using their Twitter profiles. First, we propose a hybrid approach of recommending social media profiles, articles, and advertisements to a user.The profiles are recommended based on the similarity score between the user profile, and profile under evaluation. The similarity between a set of profiles is investigated by finding the top influential words thus causing a high similarity through an Influence Term Metric for each word. Then, we group profiles of various domains such as politics, sports, and entertainment based on the similarity score through a novel clustering algorithm. The connectivity between profiles is envisaged using word graphs that help in finding the words that connect a set of profiles and the profiles that are connected to a word. Finally, we analyze the top influential words over a set of profiles through clustering by finding the similarity of that profiles enabling to break down a Twitter profile with a lot of followers to fine level word connections using word graphs. The proposed method was implemented on datasets comprising 1.1 M Tweets obtained from Twitter. Experimental results show that the resultant influential words were highly representative of the relationship between two profiles or a set of profiles

💡 Analysis

The social media craze is on an ever increasing spree, and people are connected with each other like never before, but these vast connections are visually unexplored. We propose a methodology Twigraph to explore the connections between persons using their Twitter profiles. First, we propose a hybrid approach of recommending social media profiles, articles, and advertisements to a user.The profiles are recommended based on the similarity score between the user profile, and profile under evaluation. The similarity between a set of profiles is investigated by finding the top influential words thus causing a high similarity through an Influence Term Metric for each word. Then, we group profiles of various domains such as politics, sports, and entertainment based on the similarity score through a novel clustering algorithm. The connectivity between profiles is envisaged using word graphs that help in finding the words that connect a set of profiles and the profiles that are connected to a word. Finally, we analyze the top influential words over a set of profiles through clustering by finding the similarity of that profiles enabling to break down a Twitter profile with a lot of followers to fine level word connections using word graphs. The proposed method was implemented on datasets comprising 1.1 M Tweets obtained from Twitter. Experimental results show that the resultant influential words were highly representative of the relationship between two profiles or a set of profiles

📄 Content

Twigraph: Discovering and Visualizing Influential Words between Twitter Profiles
Dhanasekar S[1] and Sudharshan Srinivasan[1] 1 Chennai, India dhanasekar312213@gmail.com Abstract. The social media craze is on an ever increasing spree, and people are connected with each other like never before, but these vast connections are visu- ally unexplored. We propose a methodology Twigraph to explore the connections between persons using their Twitter profiles. First, we propose a hybrid approach of recommending social media profiles, articles, and advertisements to a user. The profiles are recommended based on the similarity score between the user profile, and profile under evaluation. The similarity between a set of profiles is investigated by finding the top influential words thus causing a high similarity through an Influence Term Metric for each word. Then, we group profiles of var- ious domains such as politics, sports, and entertainment based on the similarity score through a novel clustering algorithm. The connectivity between profiles is envisaged using word graphs that help in finding the words that connect a set of profiles and the profiles that are connected to a word. Finally, we analyze the top influential words over a set of profiles through clustering by finding the similarity of that profiles enabling to break down a Twitter profile with a lot of followers to fine level word connections using word graphs. The proposed method was implemented on datasets comprising 1.1 M Tweets obtained from Twitter. Ex- perimental results show that the resultant influential words were highly repre- sentative of the relationship between two profiles or a set of profiles. Keywords: Twitter, Clustering, Profile Modeling, Profile Similarity, Multiple profiles connectivity 1 Introduction The important characteristic of a successful social media is its large, engaged user base. Hence, every social media tries to improve its user base. Twitter is one such popular social media site providing microblogging service that has been an important repre- sentative of people’s personal opinion in the past decade [1]. People use Twitter to share and seek information ranging from gossips to the news [26,27], as its range of connectivity far greater than any other medium. Now Twitter has around 317 million users worldwide and about 500 million tweets posted per day. Though it has tons of information with monumentally large user-base, it is practically impossible for a user to find fellow users who share a common interest manually. There is a need for an 2 efficient user suggestion system that can group users with similar interests. An auto- mated suggestion system [3] helps a user to find other users with similar interests, thus acquiring and sharing knowledge about a particular domain.
Now after an efficient recommendation system is built, a user develops his follower list. This list is built gradually or radically depending on the user’s status, and popularity resulting in the accumulation of the followers. These followers would have followed the user based on his nature of the user’s tweets. By nature, here we mean the topics used in the tweets. If the user is a guitarist and tweets were highly concentrated on acoustics, electrics and the brands of guitar, the followers of that user would probably have these topics in the majority. But when the user is a worldwide popular celebrity or politician, the nature of tweets may span several topics ranging from philosophy to cinema. Hence the followers of such a user may have followed that user for a range of topics found in his tweets. Though this is obvious, what if there is a way to find the important or influential words between a user and his follower group causing a person to be a follower. This method called Twigraph would enable to visualize the connec- tivity across profiles through words and vice versa (connectivity across words through profiles). To summarize, • We take approximately 3000 tweets of various users of domains like sports, poli- tics, philosophy and education from Twitter. We also take a large number of news and advertisement articles available online. Subsequently, we analyze, pre-process and store them efficiently.
• A profile under evaluation (user profile) is chosen, and top profiles similar to that of the user profile based on his nature of tweets are found (Explained in the upcom- ing sections).Article and advertisement suggestions are also made. • Then, we analyze the top influential words between a profile and the gradually evolving user group (user profile and his followers) using Influence Term Metric (ITM) and a variant of clustering algorithm (proposed in Section 6).

The paper is organized as follows. Section 2 deals with the related works about the usage of Twitter as a social media data set in performing various tasks like document clustering and topic modeling. Section 3 talks about the coll

This content is AI-processed based on ArXiv data.