Bilingual News Clustering Using Named Entities and Fuzzy Similarity

Montalvo, Soto; Martínez, Raquel; Casillas, Arantza; Fresno, Víctor

doi:10.1007/978-3-540-74628-7_16

Cited by 17 publications

(9 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on the basic information of the documents and additional information from Wikipedia, the documents are clustered in each language. There are also some studies on multilingual document clustering such as English-Spanish [14][15][16] , English-Japanese 17 and English-Bulgarian 18 . However there is not much study bilingual clustering towards English and Malay documents 19 .…”

Section: Related Workmentioning

confidence: 99%

Clustering Bilingual Documents Using Various Clustering Linkages Coupled with Different Proximity Measurement Techniques

et al. 2015

View full text Add to dashboard Cite

With the rich data on the web, a documents clustering task for monolingual documents is insufficient in order to produce an efficient information retrieval system. A Multilingual Document Clustering (MDC) had been introduced and it is one of the most popular trends in the area of natural language processing (NLP). In this paper, the effects of applying different clustering linkages coupled with different proximity measurements on the clustering bilingual Malay-English documents in parallel are investigated. A Hierarchical Agglomerative Clustering (HAC) has been implemented and applied in clustering bilingual Malay-English documents. Several different linkages are used in the HAC method that includes Single, Complete, Centroid and Average linkages. Not only that, the cosine similarity and the extend Jaccard coefficient are also applied in order to investigate a proper proximity measurement that can be coupled with the different type of clustering linkages used for clustering bilingual news articles written in English and Malay. The HAC method coupled with the average linkage can be considered to produce reasonable clustering results even though the average DBI is a bit high. Now only that, the study also shows that the extend Jaccard coefficient proximity measurement can produce a better clustering results compared to the cosine similarity.

show abstract

Section: Related Workmentioning

confidence: 99%

Clustering Bilingual Documents Using Various Clustering Linkages Coupled with Different Proximity Measurement Techniques

et al. 2015

View full text Add to dashboard Cite

show abstract

“…Also, for the same reason this dataset contains a large number of unrelated articles. In [15], the authors stated that NEs play an important role in news documents. They wanted to exploit that characteristic by considering them as the only distinguishing features of the documents.…”

Section: Experiments 1: Nes In Speechmentioning

confidence: 99%

Discovering Links between Political Debates and Media

Jurić

Hollink²,

Houben

2013

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Politics and media are heavily intertwined and both play a role in the discussion on policy proposals and current affairs. However, a dataset that allows a joint analysis of the two does not yet exist. In this paper we take the first step by discovering links between parliamentary debates in a political dataset and newspaper articles in a media dataset. Our approach consists of 3 steps. We first discover topics discussed in the debates. Second, we query a newspaper archive for relevant articles using a combination of debate elements: dates, actors, topics, and named entities of the debates. Finally, we discover links, represent them in RDF, and make them available for download. An evaluation of various versions of this approach shows that the topic detection adds to the quality of the discovered links, as well as the use of the semantic structure of the debate, such as headers and a division into smaller events.

show abstract

“…Friburger et al (2002) and Montalvo et al (2007) point out the effect of named entity recognition in improving monolingual and bilingual document clustering respectively. So we extract named entities through named entities recognition tools released by Stanford University, and then calculate the similarity between the two named entity sets as a feature according to Equation (7).…”

Section: Bilingual Named Entities Similaritymentioning

confidence: 99%

Minimum Error Rate Training for Bilingual News Alignment

Wang

Liu

Sun

2013

Lecture Notes in Computer Science

View full text Add to dashboard Cite

News articles in different languages on the same event are invaluable for analyzing standpoints and viewpoints in different countries. The major challenge to identify such closely related bilingual news articles is how to take full advantage of various information sources such as length, translation equivalence and publishing date. Accordingly, we propose a discriminative model for bilingual news alignment, which is capable of incorporating arbitrary information sources as features. Chinese word segmentation, Part-of-speech tagging and Named Entity Recognition technologies are used to calculate the semantic similarities between words or text as feature values. The feature weights are optimized using the minimum error rate training algorithm to directly correlate training objective to evaluation metric. Experiments on Chinese-English data show that our method significantly outperforms two strong baseline systems by 12.7% and 2.5%, respectively.

show abstract

Bilingual News Clustering Using Named Entities and Fuzzy Similarity

Cited by 17 publications

References 12 publications

Clustering Bilingual Documents Using Various Clustering Linkages Coupled with Different Proximity Measurement Techniques

Clustering Bilingual Documents Using Various Clustering Linkages Coupled with Different Proximity Measurement Techniques

Discovering Links between Political Debates and Media

Minimum Error Rate Training for Bilingual News Alignment

Contact Info

Product

Resources

About