Latent interest-topic model

Kawamae, Noriaki

doi:10.1145/1871437.1871521

Cited by 15 publications

(8 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…5. In general, with more topics, the Topic Coherence score increases while KL-Divergence decreases, which is in accordance with results in [21,32]. We found that when is larger than 15, topics became more and more similar with each other (average KL-Divergence < 2.5 …”

Section: Effects Of Number Of Topicssupporting

confidence: 89%

“…It only provides a measure of how well the model fits the data. Thus, we choose two evaluation metrics, Topic Coherence and KL-Divergence, which directly evaluate our framework on topic interpretability and topic distinctiveness [21,32]. We also report statistical significance of improvements of our framework calculated based on paired t-test.…”

Section: Objective Evaluationmentioning

confidence: 99%

“…Another important metric for topic models is topic distinctiveness [21,32]. We want to evaluate how distinctive the discovered topics are.…”

Section: Kl-divergencementioning

confidence: 99%

“…We want to evaluate how distinctive the discovered topics are. To measure the distinctiveness, we use KL-Divergence as in [21,32]. Since KL-Divergence is asymmetric, we compute its values between all pairs of topics and average them to get the average KL-Divergence.…”

Section: Kl-divergencementioning

confidence: 99%

See 3 more Smart Citations

Discovering coherent topics using general knowledge

Chen

Mukherjee

Liu

et al. 2013

Proceedings of the 22nd ACM International Conference on Conference on Information &Amp; Knowledge Management - CIKM '13

106

View full text Add to dashboard Cite

Topic models have been widely used to discover latent topics in text documents. However, they may produce topics that are not interpretable for an application. Researchers have proposed to incorporate prior domain knowledge into topic models to help produce coherent topics. The knowledge used in existing models is typically domain dependent and assumed to be correct. However, one key weakness of this knowledge-based approach is that it requires the user to know the domain very well and to be able to provide knowledge suitable for the domain, which is not always the case because in most real-life applications, the user wants to find what they do not know. In this paper, we propose a framework to leverage the general knowledge in topic models. Such knowledge is domain independent. Specifically, we use one form of general knowledge, i.e., lexical semantic relations of words such as synonyms, antonyms and adjective attributes, to help produce more coherent topics. However, there is a major obstacle, i.e., a word can have multiple meanings/senses and each meaning often has a different set of synonyms and antonyms. Not every meaning is suitable or correct for a domain. Wrong knowledge can result in poor quality topics. To deal with wrong knowledge, we propose a new model, called GK-LDA, which is able to effectively exploit the knowledge of lexical relations in dictionaries. To the best of our knowledge, GK-LDA is the first such model that can incorporate the domain independent knowledge. Our experiments using online product reviews show that GK-LDA performs significantly better than existing state-of-the-art models.

show abstract

Section: Effects Of Number Of Topicssupporting

confidence: 89%

Section: Objective Evaluationmentioning

confidence: 99%

“…Another important metric for topic models is topic distinctiveness [21,32]. We want to evaluate how distinctive the discovered topics are.…”

Section: Kl-divergencementioning

confidence: 99%

Section: Kl-divergencementioning

confidence: 99%

See 2 more Smart Citations

Discovering coherent topics using general knowledge

Chen

Mukherjee

Liu

et al. 2013

Proceedings of the 22nd ACM International Conference on Conference on Information &Amp; Knowledge Management - CIKM '13

106

View full text Add to dashboard Cite

show abstract

“…Standard topic models do not consider this key piece of information. Although there are extensions to consider authors [37], persona [31] and interest [24], none of them are suitable for considering the pair structure.…”

Section: Jte-p: Encoding Pair Structuresmentioning

confidence: 99%

Mining contentions from discussions and debates

Mukherjee

Liu

2012

Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

Social media has become a major source of information for many applications. Numerous techniques have been proposed to analyze network structures and text contents. In this paper, we focus on fine-grained mining of contentions in discussion/debate forums. Contentions are perhaps the most important feature of forums that discuss social, political and religious issues. Our goal is to discover contention and agreement indicator expressions, and contention points or topics both at the discussion collection level and also at each individual post level. To the best of our knowledge, limited work has been done on such detailed analysis. This paper proposes three models to solve the problem, which not only model both contention/agreement expressions and discussion topics, but also, more importantly, model the intrinsic nature of discussions/debates, i.e., interactions among discussants or debaters and topic sharing among posts through quoting and replying relations. Evaluation results using real-life discussion/debate posts from several domains demonstrate the effectiveness of the proposed models.

show abstract