Semantic Tagging Using Topic Models Exploiting Wikipedia Category Network

Abstract

In this paper we propose a probabilistic topic model that incorporates DBpedia knowledge into the topic model for tagging Web pages and online documents with topics discovered in them. Our method is based on integration of the DBpedia hierarchical category network with statistical topic models where DBpedia categories are considered as topics. We have conducted extensive experiments on two different datasets to demonstrate the effectiveness of our method.

Novel semantic tagging detection algorithms based non-negative matrix factorization

“…In [24] the authors developed a correlated tag learning (CTL) model for the semi-structured corpora based on the topic model to enable the construction of the correlation graph among tags via a logistic normal participation process. In [25] the authors proposed a probabilistic topic model that incorporates DBpedia knowledge into the topic model for tagging web pages and online documents with topics discovered. The model learns the probability distribution of each category over the words using the statistical topic models considering the prior knowledge from Wikipedia about the words, and their associated probabilities in various categories [25].…”

“…In [25] the authors proposed a probabilistic topic model that incorporates DBpedia knowledge into the topic model for tagging web pages and online documents with topics discovered. The model learns the probability distribution of each category over the words using the statistical topic models considering the prior knowledge from Wikipedia about the words, and their associated probabilities in various categories [25]. In [26] the authors proposed Wikipedia-category-concept mention latent Dirichlet allocation (WCM-LDA), which not only models the relationship between words and topics, but also utilizes the concept and category knowledge of entities to model the semantic relation of entities and topics.…”

“…Due to the semantic text representation approach specially WSD in addition the semantic similarity, the conventional topic models are not effectively to capture semantic topics, which leads to the poor performance in [33] X √ Schneider et al [22] X X Zhao et al [23] √ X Chen et al [32] X √ Hong et al [34] X X Peng et al [47] X X Xu et al [26] X X Li et al [24] √ X Allahyari et al [25] √ X Izquierdo et al [48] √ X Viegas et al [35] √ (word embedding) √ semantic topic learning. To tackle this problem, we introduce semantic topic modelling SNNMF.…”

Semantic Tagging-Based Document Retrieval Using Non-Negative Matrix Factorization

“…Secondly, as described above, the tags were a set of semantic topic distributions, which were learned from the plain text, and so the correlations should be modeled from the semantic level, while only considering the co-occurrences was not enough. Allahyari, et al (2016). Proposed a probabilistic topic model that incorporates DBpedia Knowledge into the topic model for tagging Web pages and online documents with topics discovered.…”

“…Proposed a probabilistic topic model that incorporates DBpedia Knowledge into the topic model for tagging Web pages and online documents with topics discovered. They learn the probability distribution of each category over the words using the statistical topic models taking into account the prior knowledge from Wikipedia about the words, and their associated probabilities in various categories [2]. There were two main limitations in this approach.…”