Toward discovery support systems: A replication, re-examination, and extension of Swanson's work on literature-based discovery of a connection between Raynaud's and fish oil

Gordon, Michael D.; Lindsay, Robert K.

doi:10.1002/(sici)1097-4571(199602)47:2<116::aid-asi3>3.0.co;2-1

Cited by 119 publications

(94 citation statements)

References 8 publications

(9 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two entities are considered to have a relationship if they co-occur with a common entity. [11] [12][36] [13]The above two approaches often do not employ natural language processing (NLP) techniques and focus on a few target entities, such as "fish oil" and "Raynaud's disease in the work by Swanson, Lindsay and Gorden [38] [11]. Rules-based techniques on the other hand heavily reply on NLP techniques to identify both syntactic and semantic units in the text.…”

Section: Introductionmentioning

confidence: 80%

A Verb-Centric Approach for Relationship Extraction in Biomedical Text

Sharma

Swaminathan

Yang

2010

2010 IEEE Fourth International Conference on Semantic Computing

View full text Add to dashboard Cite

Advances in biomedical technology and research have resulted in a large number of research findings, which are primarily published in unstructured text such as journal articles. Text mining techniques have been thus employed to extract knowledge from such data. In this article we focus on the task of identifying and extracting relations between bioentities such as green tea and breast cancer. Unlike previous work that employs heuristics such as co-occurrence patterns and handcrafted syntactic rules, we propose a verb-centric algorithm. This algorithm identifies and extracts the main verb(s) in a sentence; therefore, it does not require the usage of predefined rules or patterns. Using the main verb(s) it then extracts the two involved entities of a relationship. The biomedical entities are identified using a dependence parse tree by applying syntactic and linguistic features such as preposition phrases and semantic role analysis. The proposed verb-centric approach can effectively handle complex sentence structures such as clauses and conjunctive sentences. We evaluate the algorithm on several datasets and achieve an average F-score of 0.905, which is significantly higher than that of previous work.

show abstract

Section: Introductionmentioning

confidence: 80%

A Verb-Centric Approach for Relationship Extraction in Biomedical Text

Sharma

Swaminathan

Yang

2010

2010 IEEE Fourth International Conference on Semantic Computing

View full text Add to dashboard Cite

show abstract

“…For example, Gordon and Lindsay (1996, 1998, 2001 used tf * idf to extract concepts from texts. Moreover, UMLS and MeSH are widely used as the semantic tool to identify concepts.…”

Section: Related Researchmentioning

confidence: 99%

A compound correlation model for disjoint literature‐based knowledge discovery

Huang

He²,

Yang³

et al. 2012

Aslib Proceedings

View full text Add to dashboard Cite

Purpose -The algorithm of disjoint literature-based knowledge discovery provides a convenient, efficient and effective auxiliary method for scientific research. Based on an analysis of Swanson's A-B-C model of disjoint literature-based knowledge discovery and Gordon's intermediate literature theory, this paper seeks to propose a more comprehensive compound correlation model for disjoint literature-based knowledge discovery. Design/methodology/approach -A new algorithm of vector space model (VSM) based disjoint literature-based knowledge discovery is designed to implement the compound correlation model. Findings -The validity tests showed that this new model not only simulated both of Swanson's early and well-known discoveries of Raynaud's disease-fish oil and migraine-magnesium connections successfully, but also applied to knowledge discovery in the agricultural economics literature in the Chinese language. Research limitations/implications -Although the workload was reduced to the minimum under the compound correlation model compared with other algorithms and models, part of the work needed some manual intervention in the process of disjoint literature-based knowledge discovery with the VSM-based compound correlation model. Practical implications -The algorithm was capable of knowledge discovery with a large-scale dataset and had an advantage in identifying a series of hidden connections among a set of literatures. Therefore, application of the model might be extended to more fields. Originality/value -Traditional two-step knowledge discovery procedures were integrated into the model, which contained open and closed disjoint literature-based knowledge discovery.

show abstract

“…He finally was able to establish the logical link between the bibliographically disparate areas by searching for keywords and terms extracted from titles and abstracts in the Medline database. Ever since, his findings could be confirmed by other scholars [GORDON & LINDSAY, 1996] and extended to other medical areas [SWANSON & SMALHEISER, 1997]. In addition, various methodologies have been proposed to improve such knowledge discovery processes, for instance by structured keywords [KAJIKAWA & AL., 2006] or natural language processing (NLP) [WEEBER & AL., 2001].…”

Section: Information Science and Knowledge Generationmentioning

confidence: 99%

“…The latter tend to describe knowledge flows that not necessarily need to be connected with similar contents [HARTER & AL., 1993;STERNITZKE & BERGMANN, 2009]. Some scholars even envision the emerge of advanced text-mining tools that, one day, will enable scientists to come up with new discoveries and create new knowledge out of publicly available databases [GORDON & LINDSAY, 1996].…”

Section: Information Science and Knowledge Generationmentioning

confidence: 99%

Patents and publications as sources of novel and inventive knowledge

Sternitzke

2009

Scientometrics

View full text Add to dashboard Cite

This paper briefly reviews the knowledge-generation process and explores to what degree technical and scientific knowledge from prior art anticipates novelty or the inventive step of an invention. Inventions are novel if they have not been described (in the public) before, and they are inventive if the technical solution was non-obvious to a skilled person in the field. We employ a novel approach of patent citation analysis to investigate this phenomenon. Since in this context common approaches of such citation analysis are biased (usually, citations are neither exhaustive nor relevant in their entirety), we focus on examination reports of European patent applications and the references given therein. Our findings reveal that particularly technical knowledge comprised in patents serves as a source of novelty, while scientific knowledge frequently stems from multiple scientific papers and accounts for the inventive step. In addition, it is found that in many cases scientific knowledge is of commercial relevance and therefore constitutes more than general background information that aids the technical knowledge generation process.

show abstract

Toward discovery support systems: A replication, re-examination, and extension of Swanson's work on literature-based discovery of a connection between Raynaud's and fish oil

Cited by 119 publications

References 8 publications

A Verb-Centric Approach for Relationship Extraction in Biomedical Text

A Verb-Centric Approach for Relationship Extraction in Biomedical Text

A compound correlation model for disjoint literature‐based knowledge discovery

Patents and publications as sources of novel and inventive knowledge

Contact Info

Product

Resources

About