A Computational Approach to Feature Extraction for Identification of Suicidal Ideation in Tweets

Sawhney, Ravinder Singh; Manchanda, Prachi; Singh, Raj Mohan; Aggarwal, Swati

doi:10.18653/v1/p18-3013

Cited by 69 publications

(35 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…He introduced TF-IDF matrices and singular vector decompositions for them. Sawhney et al [27] improved the performance of Random Forest (RF) classifier for identification of suicide ideation in tweets. Logistic regression classification algorithms applied in Aladag et al [28] showed promising results in detecting suicidal content with 80-92% accuracy rate.…”

Section: Background and Related Workmentioning

confidence: 99%

See 1 more Smart Citation

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%

“…The count of each word is used to create a feature vector for a further document summarization [78]. Statistical features [27] are extracted from the posts to encompass the number of tokens, words, sentences and their length.…”

Section: Baselinementioning

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

“…Moreover, text-based judgments will fail to identify users who do not choose to explicitly share suicidal or depressive feelings online 19 . Finally, in many cases, the data collection and judgment process has been conducted on postings from designated, suicide-related forums, such as "suicide watch" on Reddit 14,20 , thus limiting their applicability to more natural, everyday settings.…”

mentioning

confidence: 99%

Deep neural networks detect suicide risk from textual facebook posts

Ophir

Tikochinski

Asterhan

et al. 2020

Sci Rep

View full text Add to dashboard Cite

Detection of suicide risk is a highly prioritized, yet complicated task. Five decades of research have produced predictions slightly better than chance (AUCs = 0.56–0.58). In this study, Artificial Neural Network (ANN) models were constructed to predict suicide risk from everyday language of social media users. The dataset included 83,292 postings authored by 1002 authenticated Facebook users, alongside valid psychosocial information about the users. Using Deep Contextualized Word Embeddings for text representation, two models were constructed: A Single Task Model (STM), to predict suicide risk from Facebook postings directly (Facebook texts → suicide) and a Multi-Task Model (MTM), which included hierarchical, multilayered sets of theory-driven risk factors (Facebook texts → personality traits → psychosocial risks → psychiatric disorders → suicide). Compared with the STM predictions (0.621 ≤ AUC ≤ 0.629), the MTM produced significantly improved prediction accuracy (0.697 ≤ AUC ≤ 0.746), with substantially larger effect sizes (0.729 ≤ d ≤ 0.936). Subsequent content analyses suggested that predictions did not rely on explicit suicide-related themes, but on a range of text features. The findings suggest that machine learning based analyses of everyday social media activity can improve suicide risk predictions and contribute to the development of practical detection tools.

show abstract

“…en el idioma inglés (Sawhney et al, 2018;Carson et al, 2019;Leiva & Freire, 2017) o en dos idiomas español-inglés (Cook et al, 2016), originando una necesidad creciente de contar con recursos de análisis de textos de causas del suicidio. Esto abre una ventana de desafíos y retos para llevar a cabo procesamiento automático de textos en español.…”

Section: Trabajo Relacionadounclassified

“…Esto abre una ventana de desafíos y retos para llevar a cabo procesamiento automático de textos en español. Además, algunos trabajos revisados (Reyes-Ortiz & Bravo, 2018;Sawhney et al, 2018;Wicentowski & Sydes, 2012;Luyckx et al, 2012;Liakata et al, 2012;Poulin et al, 2014;Carson et al, 2019), se enfocan, solamente, en la identificación o clasificación del suicidio, mientras que nuestro trabajo añade la extracción y análisis de las causas del mismo. Por lo tanto, además de aportar una solución al problema de la extracción automática de causas del suicidio, brinda un panorama sobre los marcadores lingüísticos causales que son característicos de este tipo de textos en español y almacena las causas del suicidio reportadas en notas periodísticas.…”

Section: Trabajo Relacionadounclassified

Extracción y análisis de las causas de suicidio a través de marcadores lingüísticos en reportes periodísticos

Ortiz¹,

Tovar²

2020

Linguamática

View full text Add to dashboard Cite

El análisis automático de información(textos) sobre el suicidio se ha convertido en un reto para el campo de investigación en lingüística computacional, cada vez más, son necesarias herramientas que ayuden a disminuir las tasas de suicidios, por ejemplo, extraer las causas para apoyar en su detección temprana. Los aspectos lingüísticos en los textos en Español, tales como frases clave o partes de la oración, pueden ayudar en dicha tarea. Por ello, en este artículo se presenta un enfoque computacional para la extracción y análisis de causas a partir de cabeceras de reportes periodísticos sobre el suicidio en español. La tarea de extracción automática de causas de suicidio es llevada a cabo mediante marcadores lingüísticos basados en verbos, conectores, preposiciones y conjunciones. Por su parte, el análisis de las causas de suicidio es realizado en dos enfoques: a) un análisis centrado en frases verbales y nominales, estudiando la presencia de la negación; b) un análisis centrado en la frecuencia de los unigramas y bigramas de palabras. Ambos análisis muestran resultados prometedores, los cuales son útiles para conocer los motivos de los suicidios reportados en México en un periodo determinado. Finalmente, se obtiene una colección de 581 causas del suicidio.

show abstract

A Computational Approach to Feature Extraction for Identification of Suicidal Ideation in Tweets

Cited by 69 publications

References 25 publications

Detection of Suicide Ideation in Social Media Forums Using Deep Learning

Detection of Suicide Ideation in Social Media Forums Using Deep Learning

Deep neural networks detect suicide risk from textual facebook posts

Extracción y análisis de las causas de suicidio a través de marcadores lingüísticos en reportes periodísticos

Contact Info

Product

Resources

About