Detecting and Analyzing Suicidal Ideation on Social Media Using Deep Learning and Machine Learning Models

Aldhyani, Theyazn H. H.; Alsubari, Saleh Nagi; Alshebami, Ali Saleh; Alkahtani, Hasan; Ahmed, Zeyad A. T.

doi:10.3390/ijerph191912635

Cited by 47 publications

(16 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This information can assist in detecting potential high-risk situations or individuals who may require timely attention or specialized care. For example, as the help-seeker’s suicide risk is closely correlated with their expressed sentiments [ 27 ], using ChatGPT to automatically analyze sentiments could potentially detect suicidal ideation at early stages and facilitate the counselor’s intervention accordingly.…”

Section: Discussionmentioning

confidence: 99%

Efficacy of ChatGPT in Cantonese Sentiment Analysis: A Comparative Study (Preprint)

FU,

Hsu,

Chan

et al. 2023

Journal of Medical Internet Research

View full text Add to dashboard Cite

Background Sentiment analysis is a significant yet difficult task in natural language processing. The linguistic peculiarities of Cantonese, including its high similarity with Standard Chinese, its grammatical and lexical uniqueness, and its colloquialism and multilingualism, make it different from other languages and pose additional challenges to sentiment analysis. Recent advances in models such as ChatGPT offer potential viable solutions. Objective This study investigated the efficacy of GPT-3.5 and GPT-4 in Cantonese sentiment analysis in the context of web-based counseling and compared their performance with other mainstream methods, including lexicon-based methods and machine learning approaches. Methods We analyzed transcripts from a web-based, text-based counseling service in Hong Kong, including a total of 131 individual counseling sessions and 6169 messages between counselors and help-seekers. First, a codebook was developed for human annotation. A simple prompt (“Is the sentiment of this Cantonese text positive, neutral, or negative? Respond with the sentiment label only.”) was then given to GPT-3.5 and GPT-4 to label each message’s sentiment. GPT-3.5 and GPT-4’s performance was compared with a lexicon-based method and 3 state-of-the-art models, including linear regression, support vector machines, and long short-term memory neural networks. Results Our findings revealed ChatGPT’s remarkable accuracy in sentiment classification, with GPT-3.5 and GPT-4, respectively, achieving 92.1% (5682/6169) and 95.3% (5880/6169) accuracy in identifying positive, neutral, and negative sentiment, thereby outperforming the traditional lexicon-based method, which had an accuracy of 37.2% (2295/6169), and the 3 machine learning models, which had accuracies ranging from 66% (4072/6169) to 70.9% (4374/6169). Conclusions Among many text analysis techniques, ChatGPT demonstrates superior accuracy and emerges as a promising tool for Cantonese sentiment analysis. This study also highlights ChatGPT’s applicability in real-world scenarios, such as monitoring the quality of text-based counseling services and detecting message-level sentiments in vivo. The insights derived from this study pave the way for further exploration into the capabilities of ChatGPT in the context of underresourced languages and specialized domains like psychotherapy and natural language processing.

show abstract

Section: Discussionmentioning

confidence: 99%

Efficacy of ChatGPT in Cantonese Sentiment Analysis: A Comparative Study (Preprint)

FU,

Hsu,

Chan

et al. 2023

Journal of Medical Internet Research

View full text Add to dashboard Cite

show abstract

“…• Pre-processed CSSR dataset converted to vector using TF-IDF and Word2Vec vectorizers [12,41,48]. • Feature vector is feed into various machine learning classifier which is trained to determine suicide intensity label in depression post.…”

Section: Methodsmentioning

confidence: 99%

“…In NLP features are converted to corresponding vectors referred as embedding. Typical word embedding approaches TF-IDF, Word2Vec, CNN-BiLSTM [12,17,30,48]. It is observed that Text based expression depicts mental illness more clearly compared to other features and is dominant among researchers to detect mental issues effectively.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Exploratory analysis of suicidal tendency in depression investigation social media post

MOBIN,

MRIDHA,

MAHMUD

2024

Preprint

View full text Add to dashboard Cite

Depression and suicide are interconnected. A depressive condition influences the chance of suicide. Though, to what extent of depression level triggers suicidal attempt is a scrutiny. The traditional interview-based clinical diagnostic approach is not effective for predicting a depressed person’s psychiatric status. Since patients often prefer not to disclose their emotions, they are reluctant to seek help from psychotherapists or doctors. Social media’s footprint often sees people discussing things that are not disclosed publicly. Emotionally distressed people often disclose themselves, seek empathy, and reveal physiological states. Hence, it has become a valuable source of research for depression and suicidal behaviors. This study uses social media datasets for exploratory data analysis to estimate the degree of suicidal thoughts within a depressed person’s post. The objective is to determine if a depressed individual has a suicidal tendency, determine the degree of intensity, or do the opposite. This study presents an unsupervised feature analysis using the topic model approach followed by supervised classification to quantify significance. Latent suicidal intents,cross-topic co-occurrence patterns, and dominant high-impact keywords of suicide are revealed from unsupervised Latent Dirichlet Allocation (LDA) modeling. Furthermore, supervised machine learning classifier models are applied to determine the severity of suicide tendencies. To get the best results, cutting-edge text embedding vectorization techniques and machine learning estimators are applied. Statistical measurements depict the degree of suicide intensity within the depressed label post. From the analysis, it is revealed that suicidal tendency among people with depression is extremely high. The depressed person’s post showed 60% similarities in various categories of suicidal intensity.

show abstract

“…To evaluate the performances of the RF and XGBoost algorithms for classifying the participant assessments on the 5-Item FRAIL, CHS, and SOF indices into robust, pre-frail, and frail, we employed the common evaluation indicators for ML classification: Accuracy ( Equation 7 ), Precision ( Equation 8 ), Recall ( Equation 9 ), and F1 score ( Equation 10 ): ( 35 ).…”

Section: Methodsmentioning

confidence: 99%

Physical frailty identification using machine learning to explore the 5-item FRAIL scale, Cardiovascular Health Study index, and Study of Osteoporotic Fractures index

Yang,

Chen,

Yang

et al. 2024

Front. Public Health

View full text Add to dashboard Cite

BackgroundPhysical frailty is an important issue in aging societies. Three models of physical frailty assessment, the 5-Item fatigue, resistance, ambulation, illness and loss of weight (FRAIL); Cardiovascular Health Study (CHS); and Study of Osteoporotic Fractures (SOF) indices, have been regularly used in clinical and research studies. However, no previous studies have investigated the predictive ability of machine learning (ML) for physical frailty assessment. The aim was to use two ML algorithms, random forest (RF) and extreme gradient boosting (XGBoost), to predict these three physical frailty assessment models.Materials and methodsQuestionnaires regarding demographic characteristics, lifestyle habits, living environment, and physical frailty assessment were answered by 445 participants aged 60 years and above. The RF and XGBoost algorithms were used to assess their scores for the three physical frailty indices. Furthermore, feature importance and Shapley additive explanations (SHAP) were used to determine the important physical frailty factors.ResultsThe XGBoost algorithm obtained higher accuracy for predicting the three physical frailty indices; the areas under the curve obtained by the XGBoost algorithm for the 5-Item FRAIL, CHS, and SOF indices were 0.84. 0.79, and 0.69, respectively. The feature importance and SHAP of the XGBoost algorithm revealed that systolic blood pressure, diastolic blood pressure, age, and body mass index play important roles in all three physical frailty models.ConclusionThe XGBoost algorithm has a more accurate predictive rate than RF across all three physical frailty assessments. Thus, ML can be a useful tool for the early detection of physical frailty.

show abstract

Detecting and Analyzing Suicidal Ideation on Social Media Using Deep Learning and Machine Learning Models

Cited by 47 publications

References 42 publications

Efficacy of ChatGPT in Cantonese Sentiment Analysis: A Comparative Study (Preprint)

Efficacy of ChatGPT in Cantonese Sentiment Analysis: A Comparative Study (Preprint)

Exploratory analysis of suicidal tendency in depression investigation social media post

Physical frailty identification using machine learning to explore the 5-item FRAIL scale, Cardiovascular Health Study index, and Study of Osteoporotic Fractures index

Contact Info

Product

Resources

About