An Investigation of Deep Learning Systems for Suicide Risk Assessment

Morales, Michelle; Dey, Prajjalita; Theisen, Thomas; Belitz, Daniel; Chernova, Natalia

doi:10.18653/v1/w19-3023

Cited by 19 publications

(14 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…His findings show CNN to outperform other phenol-typing algorithms on the prediction of 10 phenotypes. Morales et al [35] showed the strength of CNN and LSTM models for a suicide risk assessment presenting the results for a novelly tested personality and tone features. Bhat et al [36] and [37] highlighted CNN's performance over other approaches to identify the presence of suicidal tendencies among adolescents.…”

Section: Background and Related Workmentioning

confidence: 99%

Detection of Suicide Ideation in Social Media Forums Using Deep Learning

et al. 2019

View full text Add to dashboard Cite

Suicide ideation expressed in social media has an impact on language usage. Many at-risk individuals use social forum platforms to discuss their problems or get access to information on similar tasks. The key objective of our study is to present ongoing work on automatic recognition of suicidal posts. We address the early detection of suicide ideation through deep learning and machine learning-based classification approaches applied to Reddit social media. For such purpose, we employ an LSTM-CNN combined model to evaluate and compare to other classification models. Our experiment shows the combined neural network architecture with word embedding techniques can achieve the best relevance classification results. Additionally, our results support the strength and ability of deep learning architectures to build an effective model for a suicide risk assessment in various text classification tasks.

show abstract

Section: Background and Related Workmentioning

confidence: 99%

Detection of Suicide Ideation in Social Media Forums Using Deep Learning

et al. 2019

View full text Add to dashboard Cite

show abstract

“…We leverage two type of models: 1) Convolutional Neural Network (CNN) and 2) pretrained language models, which are used in the relevant shared task (Zirikly et al, 2019). The teams that participated in the shared task demonstrated that CNN is effective for the risk classification task (Morales et al, 2019). Also, ASU (Ambalavanan et al, 2019) shows fine-tuning pre-trained language model is highly effective.…”

Section: Methodsmentioning

confidence: 99%

Suicidal Risk Detection for Military Personnel

Park¹,

Park²,

Ahn³

et al. 2020

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

View full text Add to dashboard Cite

We analyze social media for detecting the suicidal risk of military personnel, which is especially crucial for countries with compulsory military service such as the Republic of Korea. From a widely-used Korean social Q&A site, we collect posts containing military-relevant content written by active-duty military personnel. We then annotate the posts with two groups of experts: military experts and mental health experts. Our dataset includes 2,791 posts with 13,955 corresponding expert annotations of suicidal risk levels, and this dataset is available to researchers who consent to research ethics agreement. Using various finetuned state-of-the-art language models, we predict the level of suicide risk, reaching .88 F1 score for classifying the risks.

show abstract

“…Transformers are typically trained on single messages or pairs of messages, at a time. Since we are tuning towards a human-level task, we label each user's message with their human-level attribute and treat it as a standard document-level task (Morales et al, 2019). Since we are interested in relative differences in performance, we limit each user to at most 20 messages -approximately the median number of messages, randomly sampled, to save compute time for the fine tuning experiments.…”

Section: Transformer Representationsmentioning

confidence: 99%

Empirical Evaluation of Pre-trained Transformers for Human-Level NLP: The Role of Sample Size and Dimensionality

Ganesan¹,

Matero²,

Ravula³

et al. 2021

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langua

View full text Add to dashboard Cite

In human-level NLP tasks, such as predicting mental health, personality, or demographics, the number of observations is often smaller than the standard 768+ hidden state sizes of each layer within modern transformer-based language models, limiting the ability to effectively leverage transformers. Here, we provide a systematic study on the role of dimension reduction methods (principal components analysis, factorization techniques, or multi-layer auto-encoders) as well as the dimensionality of embedding vectors and sample sizes as a function of predictive performance. We first find that fine-tuning large models with a limited amount of data pose a significant difficulty which can be overcome with a pre-trained dimension reduction regime. RoBERTa consistently achieves top performance in humanlevel tasks, with PCA giving benefit over other reduction methods in better handling users that write longer texts. Finally, we observe that a majority of the tasks achieve results comparable to the best performance with just 1 12 of the embedding dimensions.

show abstract

An Investigation of Deep Learning Systems for Suicide Risk Assessment

Abstract: This work presents the systems explored as part of the CLPsych 2019 Shared Task. More specifically, this work explores the promise of deep learning systems for suicide risk assessment.

Cited by 19 publications

References 4 publications

Detection of Suicide Ideation in Social Media Forums Using Deep Learning

Detection of Suicide Ideation in Social Media Forums Using Deep Learning

Suicidal Risk Detection for Military Personnel

Empirical Evaluation of Pre-trained Transformers for Human-Level NLP: The Role of Sample Size and Dimensionality

Contact Info

Product

Resources

About