Link prediction on Twitter

Abstract

With over 300 million active users, Twitter is among the largest online news and social networking services in existence today. Open access to information on Twitter makes it a valuable source of data for research on social interactions, sentiment analysis, content diffusion, link prediction, and the dynamics behind human collective behaviour in general. Here we use Twitter data to construct co-occurrence language networks based on hashtags and based on all the words in tweets, and we use these networks to study link prediction by means of different methods and evaluation metrics. In addition to using five known methods, we propose two effective weighted similarity measures, and we compare the obtained outcomes in dependence on the selected semantic context of topics on Twitter. We find that hashtag networks yield to a large degree equal results as all-word networks, thus supporting the claim that hashtags alone robustly capture the semantic context of tweets, and as such are useful and suitable for studying the content and categorization. We also introduce ranking diagrams as an efficient tool for the comparison of the performance of different link prediction algorithms across multiple datasets. Our research indicates that successful link prediction algorithms work well in correctly foretelling highly probable links even if the information about a network structure is incomplete, and they do so even if the semantic context is rationalized to hashtags.

Linking Physicians to Medical Research Results via Knowledge Graph Embeddings and Twitter

“…Classic link prediction methods on social media use graph properties of the social network or NLP feature of nodes to predict links between entities. For example, [11] is base solely on graph features and [3] uses a similar technique for the social networks in healthcare. Meanwhile, [1] uses common words to cluster and rank nodes and based on that predicts the closely-ranked nodes to be connected.…”

Primjena šume slučajnih stabala za predviđanje ishoda šahovske partije reprezentirane kompleksnom mrežom

“…Nadalje, značajke, odnosno, vektori značajki koji se koriste u postupcima učenja šume slučajnih stabala (Breiman, 2001) temelje se na mjerama kompleksnih mreža (Newman, 2018). Kompleksne mreže primjenjuju se na raznim područjima, poput predviđanja novih veza u društvenim mrežama (Martinčić-Ipšić et al, 2017), određivanja ključnih riječi u tekstu (Beliga et al, 2015;Beliga et al, 2016), modeliranja jezika , analize koautorstva (Meštrović i Grubiša, 2015) i sličnog, dok je šahovska igra samo mjestimično modelirana kroz formalizme kompleksnih mreža, i to u radu (Farren et al, 2013), stoga je to upravo i predmet istraživanja. U radu se istražuju mogućnosti modeliranja šahovske igre kroz formalizam kompleksnih mreža, s ciljem dobivanja boljeg uvida u razvoj šahovske partije, te uvida u način na koji informacije dobivene iz strukture kompleksnih mreža utječu na predviđanje konačnog ishoda partije.…”

Network topology inference using information cascades with limited statistical knowledge

“…Theoretical guarantees and recovery conditions are provided in the references [26], [28], [29]. Similar research work also includes link prediction in social networks such as Twitter [30], [31]. Link prediction refers to inferring the future relationships from nodes in the complex network based on the observed network structure and node attributes.…”