- 期刊
- OpenAccess
Using Ensemble Models to Classify the Sentiment Expressed in Suicide Notes
並列摘要
In 2007, suicide was the tenth leading cause of death in the U.S. Given the significance of this problem, suicide was the focus of the 2011 Informatics for Integrating Biology and the Bedside (i2b2) Natural Language Processing (NLP) shared task competition (track two). Specifically, the challenge concentrated on sentiment analysis, predicting the presence or absence of 15 emotions (labels) simultaneously in a collection of suicide notes spanning over 70 years. Our team explored multiple approaches combining regular expression-based rules, statistical text mining (STM), and an approach that applies weights to text while accounting for multiple labels. Our best submission used an ensemble of both rules and STM models to achieve a micro-averaged F1 score of 0.5023, slightly above the mean from the 26 teams that competed (0.4875).
延伸閱讀
- 陳煥杰（2012）。Building a Model to Predict The Occurrence of Posttraumatic Seizure〔碩士論文，臺北醫學大學〕。華藝線上圖書館。https://doi.org/10.6831/TMU.2012.00002
- Read, J., Velldal, E., & Øvrelid, L. (2012). Labeling Emotions in Suicide Notes: Cost-Sensitive Learning with Heterogeneous Features. Biomedical Informatics Insights, (), 99-103. https://doi.org/10.4137/BII.S8930
- Wicentowski, R., & Sydes, M. R. (2012). Emotion Detection in Suicide Notes using Maximum Entropy Classification. Biomedical Informatics Insights, (), 51-60. https://doi.org/10.4137/BII.S8972
- Liakata, M., Kim, J. Y., Saha, S., astings, J., & Schuhmann, D. R. (2012). Three Hybrid Classifiers for the Detection of Emotions in Suicide Notes. Biomedical Informatics Insights, (), 175-184. https://doi.org/10.4137/BII.S8967
- Roberts, K., & Harabagiu, S. M. (2012). Statistical and Similarity Methods for Classifying Emotion in Suicide Notes. Biomedical Informatics Insights, (), 195-204. https://doi.org/10.4137/BII.S8958