Depression Detection via Harvesting Social Media: A Multimodal Dictionary Learning Solution

Shen, Guangyao; Jiang, Jia; Nie, Liqiang; Feng, Fuli; Zhang, Cunjun; Hu, Tianrui; Chua, Tat-Seng; Zhu, Wenwu

doi:10.24963/ijcai.2017/536

Cited by 232 publications

(225 citation statements)

References 18 publications

Supporting

Mentioning

222

Contrasting

Unclassified

Order By: Relevance

“…Beyond a simple review of social media posts, social media analytics can be employed to analyze the public data in ways unbeknown to the original author. While the use of predictive analytics in social media-such as using social media posts to detect depression (Shen et al 2017)-have begun to emerge, organizations and society need to be vigilant to ensure that the power relations do not become so unbalanced and unfair that job applicants are no longer considered stakeholders.…”

Section: Resultsmentioning

confidence: 99%

Cybervetting job applicants on social media: the new normal?

Jacobson

Gruzd

2020

Ethics Inf Technol

View full text Add to dashboard Cite

With the introduction of new information communication technologies, employers are increasingly engaging in social media screening, also known as cybervetting, as part of their hiring process. Our research, using an online survey with 482 participants, investigates young people's concerns with their publicly available social media data being used in the context of job hiring. Grounded in stakeholder theory, we analyze the relationship between young people's concerns with social media screening and their gender, job seeking status, privacy concerns, and social media use. We find that young people are generally not comfortable with social media screening. A key finding of this research is that concern for privacy for public information on social media cannot be fully explained by some "traditional" variables in privacy research. The research extends stakeholder theory to identify how social media data ethics should be inextricably linked to organizational practices. The findings have theoretical implications for a rich conceptualization of stakeholders in an age of social media and practical implications for organizations engaging in cybervetting.

show abstract

Section: Resultsmentioning

confidence: 99%

Cybervetting job applicants on social media: the new normal?

Jacobson

Gruzd

2020

Ethics Inf Technol

View full text Add to dashboard Cite

show abstract

“…Such data could help in sentiment analysis and provide insights to sudden aberrations in the personality traits of the user as reflected in one's posts. In [37], authors leverage social media platforms to detect depression by harnessing the social media data. They categorize the tweets gathered from Twitter API into depression and non-depression data.…”

Section: Depression Quotient Detection In Textmentioning

confidence: 99%

Multi-level Attention Network using Text, Audio and Video for Depression Prediction

Ray

Kumar

Reddy

et al. 2019

Proceedings of the 9th International on Audio/Visual Emotion Challenge and Workshop

View full text Add to dashboard Cite

Depression has been the leading cause of mental-health illness worldwide. Major depressive disorder (MDD), is a common mental health disorder that affects both psychologically as well as physically which could lead to loss of lives. Due to the lack of diagnostic tests and subjectivity involved in detecting depression, there is a growing interest in using behavioural cues to automate depression diagnosis and stage prediction. The absence of labelled behavioural datasets for such problems and the huge amount of variations possible in behaviour makes the problem more challenging. This paper presents a novel multi-level attention based network for multimodal depression prediction that fuses features from audio, video and text modalities while learning the intra and inter modality relevance. The multi-level attention reinforces overall learning by selecting the most influential features within each modality for the decision making. We perform exhaustive experimentation to create different regression models for audio, video and text modalities. Several fusions models with different configurations are constructed to understand the impact of each feature and modality. We outperform the current baseline by 17.52% in terms of root mean squared error.

show abstract

“…It expresses the mining of users' reviews [5 & 9] about product/ topic online as a social network or blogger. The recent research introduces the challenges [10] of sentiment meaning understanding and sentiment parsing sentences and words. Writer's review is a major criterion for the quality of services enhancement to grant deliverables.…”

Section: A Sentiment Analysismentioning

confidence: 99%

“…There are difference between classification [10] and clustering [11] for sentiment ‗text. Sentiment Classification is a type of learning models for identifying data classes and pattern recognition based on labeled data.…”

Section: B Sentiment Classification and Clusteringmentioning

confidence: 99%

SentiNeural: A Depression Clustering Technique for Egyptian Women Sentiments

El-Din¹,

Hamed²,

Eldeen³

2019

IJACSA

View full text Add to dashboard Cite

Online Sentiments Analysis is a trending research domain of study which is based on natural language processing, artificial intelligence, and computational linguistics. Negation sentiments usually are not included in sentiment's analysis process. The depression analysis can be improved by negative sentiments processing. The negation sentiments may contribute to classify the depression problems and its causes. The proposed clustering technique can detect female sentiments from the sentiment's text through cause's classification, and the written sentiment style. The combination of sentiment analysis and neural network is a promising solution for creating a new clustering algorithm. According to Egypt Independent Journal in 2018, 7% of Egyptians suffer from mental illness reported by the Public Health Ministry in Egypt. But the real statistics is more than the mentioned percentage which causes major social problems such as divorce, avoiding responsibilities, or nonmarriage. This paper will address the real statistics and cluster the depression causes and social status for each sentiment. Online women sentiments are the essential focus of this research. The proposed technique consists of two algorithms clustering for user's sex and classification algorithm for causes and responsibilities of women. The proposed clustering algorithm can recognize automatically for the sentiments user sex (females or males) and the level of depression automatically. The neural network clustering approach will produce accurate analysis results. The hardness of depression analysis implicitly and explicitly demonstrated in the different classifications for sentiments. This paper introduces a new technique for clustering sentiments and evaluating Egyptian women depression based on social sentiments.

show abstract

Depression Detection via Harvesting Social Media: A Multimodal Dictionary Learning Solution

Cited by 232 publications

References 18 publications

Cybervetting job applicants on social media: the new normal?

Cybervetting job applicants on social media: the new normal?

Multi-level Attention Network using Text, Audio and Video for Depression Prediction

SentiNeural: A Depression Clustering Technique for Egyptian Women Sentiments

Contact Info

Product

Resources

About