<?xml version="1.0" encoding="UTF-8"?><xml><records><record><source-app name="Biblio" version="6.x">Drupal-Biblio</source-app><ref-type>13</ref-type><contributors><authors><author><style face="normal" font="default" size="100%">Kim Luyckx</style></author><author><style face="normal" font="default" size="100%">Frederik Vaassen</style></author><author><style face="normal" font="default" size="100%">Peersman, Claudia</style></author><author><style face="normal" font="default" size="100%">Walter Daelemans</style></author></authors></contributors><titles><title><style face="normal" font="default" size="100%">Fine-Grained Emotion Detection in Suicide Notes: A Thresholding Approach to Multi-Label Classification</style></title><secondary-title><style face="normal" font="default" size="100%">Presented at the 22th Meeting of Computational Linguistics in the Netherlands (CLIN 2012), Tilburg, The Netherlands</style></secondary-title><short-title><style face="normal" font="default" size="100%">Fine-Grained Emotion Detection in Suicide Notes</style></short-title></titles><keywords><keyword><style  face="normal" font="default" size="100%">emotion detection</style></keyword><keyword><style  face="normal" font="default" size="100%">multi-label classification</style></keyword><keyword><style  face="normal" font="default" size="100%">probability estimates</style></keyword><keyword><style  face="normal" font="default" size="100%">thresholds</style></keyword></keywords><dates><year><style  face="normal" font="default" size="100%">2012</style></year><pub-dates><date><style  face="normal" font="default" size="100%">20/01/2012</style></date></pub-dates></dates><urls><related-urls><url><style face="normal" font="default" size="100%">http://www.cnts.ua.ac.be/sites/default/files/poster-suicide-v2.pptx</style></url></related-urls></urls><abstract><style face="normal" font="default" size="100%">We present a system to automatically identify emotion-carrying sentences in suicide notes and to detect the specific fine-grained emotion conveyed. With this system, we competed in Track 2 of the 2011 Medical NLP Challenge,14 where the task was to distinguish between fifteen emotion labels, from guilt, sorrow, and hopelessness to hopefulness and happiness.
Since a sentence can be annotated with multiple emotions, we designed a thresholding approach that enables assigning multiple labels to a single instance. We rely on the probability estimates returned by an SVM classifier and experimentally set thresholds on these probabilities. Emotion labels are assigned only if their probability exceeds a certain threshold and if the probability of the sentence being emotion-free is low enough. We show the advantages of this thresholding approach by comparing it to a naïve system that assigns only the most probable label to each test sentence, and to a system trained on emotion-carrying sentences only.</style></abstract></record></records></xml>