出版社：Academy & Industry Research Collaboration Center (AIRCC)
摘要：In this paper focus is on developing a hashtag recommendation system for an online socialnetwork application with a Peer-to-Peer infrastructure motivated by BestPeer++ architecture andBATON overlay structure. A user may invoke a recommendation procedure while writing thecontent. After being invoked, the recommendation procedure returns a list of candidatehashtags, and the user may select one hashtag from the list and embed it into the content. Theproposed approach uses Latent Dirichlet Allocation (LDA) topic model to derive the latent orhidden topics of different content. LDA topic model is a well-developed data mining algorithmand generally effective in analysing text documents with different lengths. The topic model isused to identify the candidate hashtags that are associated with the texts in the published contentthrough their association with the derived hidden topics.The experiments for evaluating the recommendation approach were fed with the tweetspublished in Twitter. Hit-rate of recommendation is considered as an evaluation metricfor ourexperiments. Hit-rate is the percentage of the selected or relevant hashtags contained incandidate hashtags. Our experiment results show that the hit-rate above 50% is observed whenwe use a method of recommendation approach independently. Also, for the case that bothsimilar user and user preferences are considered at the same time, the hit-rate improved to 87%and 92% for top-5 and top-10 candidate recommendations respectively.
关键词：Bestpeer; baton; Hashtag; topic model; hit-rate and peer-to-peer networks.
