Deriving concept hierarchies from text

Sanderson, Mark; Croft, Bruce

doi:10.1145/312624.312679

Cited by 381 publications

(159 citation statements)

References 12 publications

Supporting

Mentioning

146

Contrasting

Unclassified

Order By: Relevance

“…We tested four different methods: 1) the classic subsumption method [8,13], mentioned in section 2.3.1 (labelled S); 2) the original Klink algorithm, as described in [11] (labelled K); 3) a first version of Klink-2, with the ability of integrating multiple relationships, but not addressing ambiguous keywords (labelled KR); 4) the final version of Klink-2, with also the ability to detect and split ambiguous keywords in contextual mode (labelled K2); The co-occurrence graph derived from Scopus was enriched by exploiting the cooccurrences on Google Scholar and Wikipedia, as described in [11]. KR and K2 used six statistical relationships computed on the Scopus dataset, i.e.…”

Section: Discussionmentioning

confidence: 99%

“…There are a variety of approaches for learning taxonomies or ontologies, including natural language processing [21], clustering techniques [22], statistical methods [13], and methods based on spreading activation [19]. Text2Onto [21] is a popular system for learning ontologies, which represents the learned ontological structures in a probabilistic ontology model and uses natural language processing techniques.…”

Section: Discussionmentioning

confidence: 99%

“…However, as we discussed in [11], this is a limited solution and therefore our model 12 , which builds on the BIBO ontology 13 , uses a richer set of relationships: 1) skos:broaderGeneric. This is used when we have solid evidence that a topic is a sub-area of another one -e.g., "linked data" is a sub-area of "sematic web".…”

Section: Data Modelmentioning

confidence: 99%

“…A classical way to infer a hierarchical relationship between two entities, which can occur in a set of documents, is the subsumption method [13]. According to this approach, term x subsumes term y if P(x|y) ≥ α and P(y|x) < 1, with α usually set to 0.8.…”

Section: Hierarchical Relationship Indicatorsmentioning

confidence: 99%

“…A variation of Klink was also used in the field of recommender systems to improve significantly the performance of a state of the art content-based recommender [12]. However, both Klink and similar solutions -e.g., [8,13,14], suffer from a number of limitations. First, they only consider the graph of co-occurrences between keywords [11] and/or direct semantic relationships [12], thus ignoring relevant indirect statistical and semantic relationships -e.g., the situation where two topics are related to the same conferences or associated to the same standards, knowledge which can improve the robustness and the performance of a solution, especially in the presence of noisy data.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Klink-2: Integrating Multiple Web Sources to Generate Semantic Topic Networks

Osborne

Motta

2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. The amount of scholarly data available on the web is steadily increasing, enabling different types of analytics which can provide important insights into the research activity. In order to make sense of and explore this large-scale body of knowledge we need an accurate, comprehensive and up-todate ontology of research topics. Unfortunately, human crafted classifications do not satisfy these criteria, as they evolve too slowly and tend to be too coarsegrained. Current automated methods for generating ontologies of research areas also present a number of limitations, such as: i) they do not consider the rich amount of indirect statistical and semantic relationships, which can help to understand the relation between two topics -e.g., the fact that two research areas are associated with a similar set of venues or technologies; ii) they do not distinguish between different kinds of hierarchical relationships; and iii) they are not able to handle effectively ambiguous topics characterized by a noisy set of relationships. In this paper we present Klink-2, a novel approach which improves on our earlier work on automatic generation of semantic topic networks and addresses the aforementioned limitations by taking advantage of a variety of knowledge sources available on the web. In particular, Klink-2 analyses networks of research entities (including papers, authors, venues, and technologies) to infer three kinds of semantic relationships between topics. It also identifies ambiguous keywords (e.g., "ontology") and separates them into the appropriate distinct topics -e.g., "ontology/philosophy" vs. "ontology/semantic web". Our experimental evaluation shows that the ability of Klink-2 to integrate a high number of data sources and to generate topics with accurate contextual meaning yields significant improvements over other algorithms in terms of both precision and recall.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Data Modelmentioning

confidence: 99%

Section: Hierarchical Relationship Indicatorsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Klink-2: Integrating Multiple Web Sources to Generate Semantic Topic Networks

Osborne

Motta

2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

Automatic thesaurus generation for Chinese documents

Tseng

2002

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

This article reports an approach to automatic thesaurus construction for Chinese documents. An effective Chinese keyword extraction algorithm is first presented. Experiments showed that for each document an average of 33% keywords unknown to a lexicon of 123,226 terms could be identified by this algorithm. Of these unregistered words, only 8.3% of them are illegal. Keywords extracted from each document are further filtered for term association analysis. Association weights larger than a threshold are then accumulated over all the documents to yield the final term pair similarities. Compared to previous studies, this method speeds up the thesaurus generation process drastically. It also achieves a similar percentage level of term relatedness.

show abstract

Observing users, designing clarity: A case study on the user‐centered design of a cross‐language information retrieval system

Petrelli

Beaulieu

Sanderson

et al. 2004

J. Am. Soc. Inf. Sci.

Self Cite

View full text Add to dashboard Cite

This report presents a case study of the development of an interface for a novel and complex form of document retrieval: searching for texts written in foreign languages based on native language queries. Although the underlying technology for achieving such a search is relatively well understood, the appropriate interface design is not. A study involving users from the beginning of the design process is described, and it covers initial examination of user needs and tasks, preliminary design and testing of interface components, building, testing, and refining the interface, and, finally, conducting usability tests of the system. Lessons are learned at every stage of the process, leading to a much more informed view of how such an interface should be built.

show abstract

Deriving concept hierarchies from text

Cited by 381 publications

References 12 publications

Klink-2: Integrating Multiple Web Sources to Generate Semantic Topic Networks

Klink-2: Integrating Multiple Web Sources to Generate Semantic Topic Networks

Automatic thesaurus generation for Chinese documents

Observing users, designing clarity: A case study on the user‐centered design of a cross‐language information retrieval system

Contact Info

Product

Resources

About