DepressionOne@LT-EDI-ACL2022: Using Machine Learning with SMOTE and Random UnderSampling to Detect Signs of Depression on Social Media Text.

Dowlagar, Suman; Mamidi, Radhika

doi:10.18653/v1/2022.ltedi-1.45

Cited by 3 publications

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Algorithms can struggle to adequately learn patterns and V J Biomed Phys Eng properties of the minority class due to insufficient instances. To mitigate this, SMOTE synthetically generates new minority class examples by interpolating between existing minority data points in feature space [42,43]. Augmenting the minority class via oversampling improves class balance and enhances model performance on the rare class.…”

Section: Smotementioning

confidence: 99%

Supervised Learning Algorithm Comparison in Discharge Status Prediction of Trauma Patients: Empirical Evaluation

2024

J Biomed Phys Eng

View full text Add to dashboard Cite

Background: By analyzing information from trauma centers, hospitals can identify crucial performance indicators that affect budgets and present growth opportunities, potentially leading to lower mortality rates and improved health status indicators.Objective: This study aims to determine the best-supervised algorithm for diagnosing the discharge status of trauma patients. Material and Methods:This retrospective study used the data, collected by the Kashan Trauma Registry from March 2018 to February 2019. Several supervised algorithms, including Naive Bayes, Logistic Regression, Support Vector Machine, Random Forest, and K-Nearest Neighbors, have been evaluated for predicting the discharge status of trauma patients. The performance metrics of accuracy, precision, recall, and F-measure were used. The hold-out technique was applied to train the data. Results:The Random Forest algorithm had the best performance among the other algorithms. The best accuracy, precision, recall, and F-measure for Gini index were 84/2%, 79/7%, 78/3%, and 76.4%, and for information gain were 84.6%, 79.6%, 76.8%, and 76/20%, respectively. Conclusion:The results of this research showed that the supervised algorithms, with proper parameter settings, can help diagnose the discharge status of trauma patients. In addition, data balancing can help improve the performance of the algorithms. However, this claim cannot be generalized because it depends on the type of algorithm and the values of the parameters.

show abstract

Section: Smotementioning

confidence: 99%

Supervised Learning Algorithm Comparison in Discharge Status Prediction of Trauma Patients: Empirical Evaluation

2024

J Biomed Phys Eng

View full text Add to dashboard Cite

show abstract

Explainable Deep Learning for Mental Health Detection from English and Arabic Social Media Posts

Kumar,

Kumari,

Pradhan

2023

ACM Trans. Asian Low-Resour. Lang. Inf. Process.

View full text Add to dashboard Cite

The research communities have recently begun exploring the detection of depression through social media, making it a relatively new development in this field. Some research has been done to identify depressive signs from English social media postings, while little work has been done for the Arabic posts. This paper proposes a BERT and Bi-LSTM pipeline for the identification of depressive signs from Arabic social media posts. A fine-tuned RoBERTa-based model is also proposed for English social media posts for depressive state identification. Along with the proposed model, seven conventional machine learning and eight deep learning models are also explored for the identification of depressive signs from Arabic and English social media posts. The performance of the proposed model is validated on two Arabic datasets and one English dataset. The proposed BERT and Bi-LSTM pipeline achieved state-of-the-art performance with an F 1 -score of 1.00 and 0.82 for two different Arabic datasets, whereas the proposed fine-tuned RoBERTa achieved a F 1 -score of 0.60 which is comparable in identifying depressive sign from English social media posts. The majority of the suggested deep learning models are end-to-end, which necessitates a greater explanation for their success. An explainable AI-based model may enhance decision-making, transparency, and interpretability. Therefore, this research identifies where the suggested system learned well and where it failed in recognition of the depression signs that can help in future developments in the field of depression detection.

show abstract

Depression Classification based on Automatic Ontology Generation and Natural Language Processing (Preprint)

Goga,

Vasilateanu,

Popa

et al. 2023

Preprint

View full text Add to dashboard Cite

BACKGROUND Depression is one of the most common mental disorders and one that has a great potential to affect people mentally, physically, and socially. Unfortunately, a large number of people either do not have access to treatment or avoid seeking help. In this context, many platforms have emerged to provide a space for discussion and support where users can interact anonymously. OBJECTIVE A supporting objective of the research was to create a natural language processing tool for classifying depression to obtain the corpus necessary for ontology creation. METHODS This study presents the results of our research on classifying depression in the specific community of religious people by analyzing texts posted on social media using semantic techniques, such as comparative analysis of texts from users using ontologies. RESULTS There were no significant differences in similarities between the various categories examined in this study. This suggests that, with respect to the characteristics under consideration, the categories did not exhibit substantial variations. Related to the comparison between religious individuals diagnosed with depression and the general depressed population, our analysis indicated a noteworthy distinction. Specifically, the classification based on words utilized by the religious group highlighted the significance of the religious dimension in the context of depression. CONCLUSIONS This research underscores the potential of language analysis on social media platforms as a tool for gaining insights into the experience of depression. Additionally, the study has shed light on the distinctive role that religion can play in the coping mechanisms of religious individuals facing depression.

show abstract

DepressionOne@LT-EDI-ACL2022: Using Machine Learning with SMOTE and Random UnderSampling to Detect Signs of Depression on Social Media Text.

Cited by 3 publications

References 6 publications

Supervised Learning Algorithm Comparison in Discharge Status Prediction of Trauma Patients: Empirical Evaluation

Supervised Learning Algorithm Comparison in Discharge Status Prediction of Trauma Patients: Empirical Evaluation

Explainable Deep Learning for Mental Health Detection from English and Arabic Social Media Posts

Depression Classification based on Automatic Ontology Generation and Natural Language Processing (Preprint)

Contact Info

Product

Resources

About