Classifying Semantic Relations in Bioscience Texts

By Barbara Rosario, Marti A. Hearst. Annual Meeting of the Association for Computational Linguistics, 2004

309 citations recorded by the source corpus.

A crucial step toward the goal of automatic extraction of propositional information from natural language text is the identification of semantic relations between constituents in sentences. We examine the problem of distinguishing among seven relation types that can occur between the entities "treatment" and "disease" in bioscience text, and the problem of identifying such entities. We compare five generative graphical models and a neural network, using lexical, syntactic, and semantic features, finding that the latter help achieve high classification accuracy.

Subfields: Artificial Intelligence, Molecular Biology

View Paper: https://www.semanticscholar.org/paper/dc34d70f8b0a7dd35725f03463221705eb67cefb

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Molecular Biology

Citations in the corpus

BANNER: An Executable Survey of Advances in Biomedical Named Entity Recognition (2007, methodology): “Results of comparing BANNER against existing freely-available software, using 5x2 crossvalidation on the disease mentions from the BioText disease/treatment corpus [10].”
Biomedical text mining and its applications in cancer research (2013, unknown): “BioText [178–184] BioText was initially constructed by 1000 randomly selected MEDLINE abstracts from the results of a query on the term yeast.”
UTD: Determining Relational Similarity Using Lexical Patterns (2012, background): “…has examined and elaborated upon a wide variety of semantic relations between concepts along with techniques for automatically discovering pairs of concepts for which a relation holds (Bejar et al., 1991; Stephens and Chen, 1996; Rosario and Hearst, 2004; Khoo and Na, 2006; Girju et al., 2009).”
CollaboNet: collaboration of deep neural networks for biomedical named entity recognition (2018, background): “Biomedical named entity recognition (BioNER) is an essential building block of many downstream text mining applications such as extracting drug-drug interactions [8] and disease-treatment relations [9].”
Special Issue Introduction: Semantic Role Labeling: An Introduction to the Special Issue (2008, unknown): “Related work on classifying the semantic relations in noun phrases has also been encouraging for NLP tasks (Moldovan et al. 2004; Rosario and Hearst 2004).”
Open Information Extraction from the Web (2007, background): “[Bunescu and Mooney, 2005], maximum-entropy models [Kambhatla, 2004], graphical models [Rosario and Hearst, 2004; Culotta et al. , 2006], and co-occurrence statistics [Lin and Pantel, 2001; Ciaramita et al. , 2005] over small, domain-speciﬁc corpora and limited sets of relations.”
MeInfoText: associated gene methylation and cancer information from text mining (2008, methodology): “The data used for the cancer name evaluation can be retrieved at [30,31].”
Frontiers of biomedical text mining: current progress (2007, background): “…both in terms of semantic resources such as proposition banks [123, 124] that take us beyond the binary representations that have characterized most of the early work on biomedical relation extraction [125] and in terms of work that targets binary relations of increasing granularity [126].”