The Unified Medical Language System: An Informatics Research Collaboration

Humphreys, Betsy L.; Lindberg, Donald A. B.; Schoolman, Harold M.; Barnett, G. Octo

doi:10.1136/jamia.1998.0050001

Cited by 374 publications

(183 citation statements)

References 14 publications

Supporting

Mentioning

179

Contrasting

Unclassified

Order By: Relevance

“…Here, the UMLS [15] is a readily available resource to provide syntactic and semantic information. For example, Liu et al [16] focused on ambiguous abbreviations.…”

Section: Information Sourcesmentioning

confidence: 99%

“…We chose the UMLS [15] Semantic Network as our external knowledge source and tested which portions of this network help disambiguate words automatically. In considering a sentence containing an ambiguity, we use the symbolic representation of that sentence in the UMLS Semantic Network [24] and do not use the actual words surrounding the ambiguous term.…”

Section: External Knowledge Sourcementioning

confidence: 99%

See 1 more Smart Citation

Effects of information and machine learning algorithms on word sense disambiguation with small datasets

Leroy

Rindflesch

2005

International Journal of Medical Informatics

View full text Add to dashboard Cite

Summary Current approaches to word sense disambiguation use (and often combine) various machine learning techniques. Most refer to characteristics of the ambiguity and its surrounding words and are based on thousands of examples. Unfortunately, developing large training sets is burdensome, and in response to this challenge, we investigate the use of symbolic knowledge for small datasets. A naïve Bayes classifier was trained for 15 words with 100 examples for each. Unified Medical Language System (UMLS) semantic types assigned to concepts found in the sentence and relationships between these semantic types form the knowledge base. The most frequent sense of a word served as the baseline. The effect of increasingly accurate symbolic knowledge was evaluated in nine experimental conditions. Performance was measured by accuracy based on 10-fold cross-validation. The best condition used only the semantic types of the words in the sentence. Accuracy was then on average 10% higher than the baseline; however, it varied from 8% deterioration to 29% improvement. To investigate this large variance, we performed several followup evaluations, testing additional algorithms (decision tree and neural network), and gold standards (per expert), but the results did not significantly differ. However, we noted a trend that the best disambiguation was found for words that were the least troublesome to the human evaluators. We conclude that neither algorithm nor individual human behavior cause these large differences, but that the structure of the UMLS Metathesaurus (used to represent senses of ambiguous words) contributes to inaccuracies in the gold standard, leading to varied performance of word sense disambiguation techniques.

show abstract

“…Here, the UMLS [15] is a readily available resource to provide syntactic and semantic information. For example, Liu et al [16] focused on ambiguous abbreviations.…”

Section: Information Sourcesmentioning

confidence: 99%

Section: External Knowledge Sourcementioning

confidence: 99%

Effects of information and machine learning algorithms on word sense disambiguation with small datasets

Leroy

Rindflesch

2005

International Journal of Medical Informatics

View full text Add to dashboard Cite

show abstract

“…[48][49][50][51] One major obstacle to cross-source information retrieval is that the same information is often expressed differently in different vocabularies used by the various systems and there is no universal biomedical vocabulary. Knowing that to dictate the use of a single vocabulary is not realistic, the UMLS circumvents this problem by creating links between the terms in different vocabularies.…”

Section: Unified Medical Language System (Umls)mentioning

confidence: 99%

Knowledge Representation and Ontologies

Fung

Bodenreider

2012

Health Informatics

View full text Add to dashboard Cite

Ontologies have become important tools in biomedicine, supporting critical aspects of both health care and biomedical research, including clinical research [1]. Some even see ontologies as integral to science [2]. Unlike terminologies (focusing on naming) and classification systems (developed for partitioning a domain), ontologies define the types of entities that exist, as well as their interrelations. And while knowledge bases generally integrate both definitional and assertional knowledge, ontologies focus on what is always true of entities, i.e., definitional knowledge [3]. In practice, however, there is no sharp distinction between these kinds of artifacts and 'ontology' has become a generic name for a variety of knowledge sources with important differences in their degree of formality, coverage, richness and computability [4]. In this chapter, we focus on those ontologies of particular relevance to clinical research. After a brief introduction to ontology development and knowledge representation, we present the characteristics of some of these ontologies. We then show how ontologies are integrated in and made accessible through knowledge repositories, and illustrate their role in clinical research. Ontology developmentOntology development has not yet been formalized to the same extent as, say, database development has, and there is still no equivalent for ontologies to the entity-relationship model. However, ontology development is guided by fundamental ontological distinctions and supported by the formalisms and tools for knowledge representation that have emerged over the past decades. Several top-level ontologies provide useful constraints for the development of domain ontologies and one the most recent trends is increased collaboration among the creators of ontologies for coordinated development.In: Richesson RL, Andrews JE, editors. Clinical research informatics. London: Springer-Verlag; 2012. p. 255-275. 2 Important ontological distinctionsA small number of ontological distinctions inherited from philosophical ontology provide a useful framework for creating ontologies. The first distinction is between types and instances. Instances correspond to individual entities (e.g., my left kidney, the patient identified by 1234), while types represent the common characteristics of sets of instances (e.g., a kidney is a bean-shaped, intraabdominal organ -properties common to all kidneys) [5]. Instances are related to the corresponding types by the relation instance of. For example, my left kidney is an instance of kidney. (It must be noted that most biomedical ontologies only represent types in reference to which the instances recorded in patient records and laboratory notebooks can be annotated). Another fundamental distinction is between continuants and occurrents [6]. While continuants exist (endure) through time, occurrents go through time in phases. Roughly speaking, objects (e.g., a liver, an endoscope) are continuants and processes (e.g., the flow of blood through the mitral valve) are continuants. O...

show abstract

“…Finally, the event extractor finds the binary relation using the syntactic information of a given sentence, co-occurrence statistics between two named entities, and pattern information of an event verb. General medical term was trained with UMLS meta-thesaurus [12] and the biological entity and its interaction was trained with GENIA [13] corpus. The underlying NLP approaches for named entity recognition are based on the system of Hwang et al [14] and Lee et al [15] with collaborations.…”

Section: Interaction Extractionmentioning

confidence: 99%

PubMiner: Machine Learning-Based Text Mining System for Biomedical Information Mining

Eom

Zhang

2004

Artificial Intelligence: Methodology, Systems, and Applications

View full text Add to dashboard Cite

Abstract. PubMiner, an intelligent machine learning based text mining system for mining biological information from the literature is introduced. PubMiner utilize natural language processing and machine learning based data mining techniques for mining useful biological information such as protein-protein interaction from the massive literature data. The system recognizes biological terms such as gene, protein, and enzymes and extracts their interactions described in the document through natural language analysis. The extracted interactions are further analyzed with a set of features of each entity which were constructed from the related public databases to infer more interactions from the original interactions. An inferred interaction from the interaction analysis and native interaction are provided to the user with the link of literature sources. The evaluation of system performance proceeded with the protein interaction data of S.cerevisiae (bakers yeast) from MIPS and SGD.

show abstract

The Unified Medical Language System: An Informatics Research Collaboration

Cited by 374 publications

References 14 publications

Effects of information and machine learning algorithms on word sense disambiguation with small datasets

Effects of information and machine learning algorithms on word sense disambiguation with small datasets

Knowledge Representation and Ontologies

PubMiner: Machine Learning-Based Text Mining System for Biomedical Information Mining

Contact Info

Product

Resources

About