Document clustering with committees

Pantel, Patrick; Lin, Dekang

doi:10.1145/564376.564412

Cited by 117 publications

(54 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Pantel (2005) created semantic vectors for each word in WordNet by disambiguating contexts which appeared with different senses of a word. The building of semantic vectors is described in (Pantel 2003). WordNet's hierarchy was used to propagate contexts where words may appear throughout the network.…”

Section: Updating Wordnetmentioning

confidence: 99%

Evaluation of automatic updates of Roget’s Thesaurus

Kennedy

Śzpakowicz

2014

JLM

View full text Add to dashboard Cite

Thesauri and similarly organised resources attract increasing interest of Natural Language Processing researchers. Thesauri age fast, so there is a constant need to update their vocabulary. Since a manual update cycle takes considerable time, automated methods are required. This work presents a tuneable method of measuring semantic relatedness, trained on Roget's Thesaurus, which generates lists of terms related to words not yet in the Thesaurus. Using these lists of terms, we experiment with three methods of adding words to the Thesaurus. We add, with high confidence, over 5500 and 9600 new word senses to versions of Roget's Thesaurus from 1911 and 1987 respectively. We evaluate our work both manually and by applying the updated thesauri in three NLP tasks: selection of the best synonym from a set of candidates, pseudo-word-sense disambiguation and SAT-style analogy problems. We find that the newly added words are of high quality. The additions significantly improve the performance of Roget's-based methods in these NLP tasks. The performance of our system compares favourably with that of WordNet-based methods. Our methods are general enough to work with different versions of Roget's Thesaurus.

show abstract

Section: Updating Wordnetmentioning

confidence: 99%

Evaluation of automatic updates of Roget’s Thesaurus

Kennedy

Śzpakowicz

2014

JLM

View full text Add to dashboard Cite

show abstract

“…Once the frequency vectors are collected, the pointwise mutual information vectors are generated since experiments showed that they produce much higher quality clusters than by using the term frequency vectors C(e) [10].…”

Section: There Is a List Of Standard Dependency Relations Listed In [mentioning

confidence: 99%

“…Mutual information vector MI(e) = (mi e1 , mi e2 , …, mi em ) for each element e is generated, where mi ef is the pointwise mutual information between element e and feature f, which is defined (1) as [10]:…”

Section: Stage 3: Generating the Pointwise Mutual Information Vectorsmentioning

confidence: 99%

“…In the clustering by committee algorithm, any similarity metric can be applied [10]. The similarity between two elements e i and e j will be computed using the cosine coefficient.…”

Section: Stage 4: Measuring Similarity Between Feature Vectorsmentioning

confidence: 99%

“…The cosine coefficient model is a common similarity model because of its simplicity (2). However, it is only applicable with numerical features and therefore it suits the introduced application and is defined as [10].…”

Section: Stage 4: Measuring Similarity Between Feature Vectorsmentioning

confidence: 99%

See 2 more Smart Citations

Road Map Approach to Automatic Topic Detection of Diaries

Ismail¹

2013

IJMLC

View full text Add to dashboard Cite

Abstract-The famous question on social networks "what is on your mind?" urges many of us to convert their thoughts and feelings into diaries. Detecting the topic of diaries is an interesting task to know people' interest. Many automatic methods have been introduced. The method used in this paper depends on preprocessing the diary words to generate a feature vector for each word. Then, the senses of each word are detected from the diary context by using CBC (clustering by committee) algorithm. CBC can avoid discovering duplicate senses and discover the less frequent senses of a word. The sense with the highest score is selected for each word. To detect the topic of the diary, the whole discovered senses are translated into concepts using the hierarchical concept taxonomy Wordnet. Concepts are traversed from bottom up to reach the most generalized concepts that express the topic of the diary. The previous approaches studied in this paper depend on the idea that the more frequent a word is used, the more important it is. The approach presented differs in that it incorporates the semantic dependencies among words because straightforward word counting misses the important concepts in the single document. This model can be used to explore people interests from their writings and can serve in e-marketing.

show abstract

Exploiting corpus‐related ontologies for conceptualizing document corpora

Zheng

Borchert

Kim

2009

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

As a greater volume of information becomes increasingly available across all disciplines, many approaches, such as document clustering and information visualization, have been proposed to help users manage information easily. However, most of these methods do not directly extract key concepts and their semantic relationships from document corpora, which could help better illuminate the conceptual structures within given information. To address this issue, we propose an approach called "Clonto" to process a document corpus, identify the key concepts, and automatically generate ontologies based on these concepts for the purpose of conceptualization. For a given document corpus, Clonto applies latent semantic analysis to identify key concepts, allocates documents based on these concepts, and utilizes WordNet to automatically generate a corpus-related ontology. The documents are linked to the ontology through the key concepts. Based on two test collections, the experimental results show that Clonto is able to identify key concepts, and outperforms four other clustering algorithms. Moreover, the ontologies generated by Clonto show significant informative conceptual structures.

show abstract

Document clustering with committees

Cited by 117 publications

References 10 publications

Evaluation of automatic updates of Roget’s Thesaurus

Evaluation of automatic updates of Roget’s Thesaurus

Road Map Approach to Automatic Topic Detection of Diaries

Exploiting corpus‐related ontologies for conceptualizing document corpora

Contact Info

Product

Resources

About