Multi-class SVM for relation extraction from clinical reports

Anne-Lyse Minard (1) , Anne-Laure Ligozat (1) , Brigitte Grau (1)

1 LIMSI - Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur

Information extraction in specialized texts raises different problems related to the kind of searched information. In this paper, we are interested in relation identification between some concepts in medical reports, a task that was evaluated in the i2b22010 challenge. As relations are expressed in natural language with a great variety of forms, we proceeded to sentence analysis by extracting features that enable all together to identify a relation and we modeled this task as a multi-class classification based on an SVM, each type of relation representing a class. We will present the selection of the features used by our system and an error analysis. This approach allowed us to obtain an F-measure of 0.70, classifying the system among the best systems.

Domaines: Informatique [cs]; Informatique et langage [cs.CL]

Anne-Lyse Minard, Anne-Laure Ligozat, Brigitte Grau. Multi-class SVM for relation extraction from clinical reports. Recent Advances in Natural Language Processing, Jan 2011, Hissar, Bulgaria. ⟨hal-02282084⟩