CEASE, a Corpus of Emotion Annotated Suicide notes in English

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 1618–1626 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 1618 CEASE, a Corpus of Emotion Annotated Suicide notes in English Soumitra Ghosh, Asif Ekbal, Pushpak Bhattacharyya Department of Computer Science and Engineering Indian Institute of Technology Patna, Patna, India {1821cs05, asif, pb}@iitp.ac.in Abstract A suicide note is usually written shortly before the suicide, and it provides a chance to comprehend the self-destructive state of mind of the deceased. From a psychological point of view, suicide notes have been utilized for recognizing the motive behind the suicide. To the best of our knowledge, there are no openly accessible suicide note corpus at present, making it challenging for the researchers and developers to deep dive into the area of mental health assessment and suicide prevention. In this paper, we create a fine-grained emotion annotated corpus (CEASE) of suicide notes in English, and develop various deep learning models to perform emotion detection on the curated dataset. The corpus consists of 2393 sentences from around 205 suicide notes collected from various sources. Each sentence is annotated with a particular emotion class from a set of 15 fine-grained emotion labels, namely (forgiveness, happiness peacefulness, love, pride, hopefulness, thankfulness, blame, anger, fear, abuse, sorrow, hopelessness, guilt, information, instructions). For the evaluation, we develop an ensemble architecture, where the base models correspond to three supervised deep learning models, namely Convolutional Neural Network (CNN), Gated Recurrent Unit (GRU) and Long Short Term Memory (LSTM).We obtain the highest test accuracy of 60.17%, and cross-validation accuracy of 60.32%. Keywords: Emotion Analysis, Suicide notes, Deep learning 1. Introduction Suicide, a serious public well-being concern and a ma- jor cause of death worldwide, can be marked as the death caused by self-directed harmful behaviour with any intent to end one’s life. Moments before committing this hor- rendous act, usually an individual’s mind is flooded with a range of unpleasant emotions. (Pestian et al., 2008; Pestian et al., 2010; Duch et al., 2008). A suicide note is a signif- icant asset when attempting to evaluate a patient’s risk of attempting suicide repeatedly. The suicide note furnishes us with the firsthand information about that individual’s specific personality status and mind rationale (Pestian and Matykiewicz, 2008). Suicide notes may serve some infor- mative purpose and may have a remedial role in helping the surviving relatives to understand the reason behind suicide. An understanding of the messages contained inside suicide notes could be helpful for suicide aversion programs. An- alyzing the inward feelings uncovered in the suicide note may assist us with identifying individuals who conceivably have suicide ideation (Xu et al., 2012), and thus prevent the misery from happening. Modelling the emotions present in such notes may help health experts in surveying suicide hazard, by contrasting the model with writings written by at-risk subjects, such as mental patients or online content producers. The success of suicide prevention, a significant general well-being concern worldwide, relies on satisfac- tory …

Content truncated to 500 words