We describe our approach for creating a system able to detect emotions in suicide notes. Motivated by the sparse and imbalanced data as well as the complex annotation scheme, we have considered three hybrid approaches for distinguishing between the different categories. Each of the three approaches combines machine learning with manually derived rules, where the latter target very sparse emotion categories. The first approach considers the task as single label multi-class classification, where an SVM and a CRF classifier are trained to recognise fifteen different categories and their results are combined. Our second approach trains individual binary classifiers (SVM and CRF) for each of the fifteen sentence categories and returns the union of the classifiers as the final result. Finally, our third approach is a combination of binary and multi-class classifiers (SVM and CRF) trained on different subsets of the training data. We considered a number of different feature configurations. All three systems were tested on 300 unseen messages. Our second system had the best performance of the three, yielding an F1 score of 45.6% and a Precision of 60.1% whereas our best Recall (43.6%) was obtained using the third system.
PDF (579.75 KB PDF FORMAT)
RIS citation (ENDNOTE, REFERENCE MANAGER, PROCITE, REFWORKS)
BibTex citation (BIBDESK, LATEX)
It's a great experience publishing with Biomedical Informatics Insights. I am particularly impressed with the in-depth and constructive comments provided by the reviewers within such a short time-frame. The typesetting was not only prompt, but most importantly, effective. In fact, this was among the very few publication experiences that I have had when no correction was needed in the author proofs. I highly recommend Biomedical Informatics Insights to both readers and prospective ...
All authors are surveyed after their articles are published. Authors are asked to rate their experience in a variety of areas, and their responses help us to monitor our performance. Presented here are their responses in some key areas. No 'poor' or 'very poor' responses were received; these are represented in the 'other' category.See Our Results
Copyright © 2013 Libertas Academica Ltd (except open access articles and accompanying metadata and supplementary files.)