Impact summary

The research article titled "Depression Detection from Social Network Data Using Machine Learning Techniques," published in collaboration with Victoria and Swinburne University, has made a significant impact by influencing Canadian mental health policy. Cited in the Canadian Government Agency for Drugs and Technologies in Health (CADTH) policy document, "Artificial Intelligence and Machine Learning in Mental Health Services: An Environmental Scan," this research has contributed to shaping national strategies on the application of artificial intelligence (AI) and machine learning (ML) in mental health services. Furthermore, the research has also influenced technological developments, being cited in the U.S. patent "METHOD AND SYSTEM FOR CONDUCTING SURVEY RELATING TO URINATION, AND NON-TRANSITORY COMPUTER-READABLE RECORDING MEDIUM" (US20210265027A1 application filed in 2021), highlighting its relevance to the broader field of digital health innovations.

Research and engagement activities leading to impact

Rationale: Social networks provide rich data reflecting users' moods, feelings, and sentiments, offering a unique opportunity to analyze mental health indicators. The research aimed to explore the potential of machine learning techniques in detecting depression from social network data, specifically Facebook, to provide scalable and accurate solutions for mental health issues.

Research Work and Engagement Activities: The study developed a machine learning-based method to analyze depression symptoms from publicly available Facebook data. The research was conducted in collaboration with external collaborators, including collaborators from Victoria and Swinburne University and was published in the top-tier journal Health Information Science and Systems (SJR Q1, cite score 11.3, percentile 90%, impact factor 5.4) in 2018.


End-users Involved: The research has been utilised by policy makers, particularly the CADTH, to guide the development of AI and ML applications in mental health services across Canada. This relationship highlights the relevance of academic research in informing public policy.

Relevant Inputs: The study leveraged publicly available Facebook data, psycholinguistic analysis tools, and machine learning algorithms to develop and validate the proposed depression detection method.

Research outputs associated with the impact

The key output of this research was the article titled "Depression Detection from Social Network Data Using Machine Learning Techniques," published in 2018 in Health Information Science and Systems (SJR Q1, cite score 11.3, percentile 90%, impact factor 5.4). The article has garnered significant academic attention with 427 citations according to Google Scholar and 5.64 FWCI (as per Scopus), making it the second most-cited paper in that journal. This high citation count reflects the paper's influence and relevance within the academic and professional communities.

Researcher involvement

As the co-lead researcher and 2nd author, I played a central role in conceptualizing and designing the study, developing the machine learning models, and conducting the data analysis. The research benefited from the collaboration with experts at IUT, Victoria and Swinburne Universities, who contributed to the methodological rigor and provided additional insights into the data interpretation and psycholinguistic analysis.

Outcomes of research leading to impact

The research has been implemented in practice by being cited in a significant policy document by CADTH titled "Artificial Intelligence and Machine Learning in Mental Health Services: An Environmental Scan." This document serves as a foundation for understanding the current landscape of AI and ML applications in mental health across Canada, guiding future research, development, and policy-making efforts in the field. The findings from our study have informed the report's objectives, particularly in understanding the trends, applications, and key players in AI-driven mental health initiatives. The influence of this research on policy development signifies its broader impact beyond the academic community.

Beneficiaries of the impact

Primary Beneficiaries: The Canadian Government, particularly the CADTH, which has used the research to shape national policies on AI and ML in mental health services.

Secondary Beneficiaries: The broader mental health community, including patients and healthcare providers, also benefits indirectly from the improved AI-driven diagnostic tools and frameworks informed by this research.

Reach: The research has impacted national policy in Canada, with potential implications for AI and ML applications in mental health globally.

Details of the impact achieved

Significance: The citation of this research in the CADTH policy document underscores its significance in informing national health policies. It highlights the importance of machine learning techniques in advancing mental health care and the role of academic research in shaping public health strategies.

Evidence: The citation of our article in the CADTH report, "Artificial Intelligence and Machine Learning in Mental Health Services: An Environmental Scan", as one of the 92 scholarly outputs that informed its conclusions, provides concrete evidence of the research's impact on policy development. Additionally, the high citation count of the article further validates its influence within the academic community and beyond. Moreover, the article's influence extends to technological innovations, as evidenced by its citation in the U.S. patent "METHOD AND SYSTEM FOR CONDUCTING SURVEY RELATING TO URINATION, AND NON-TRANSITORY COMPUTER-READABLE RECORDING MEDIUM" (US20210265027A1, application filed in 2021).
Impact date2021
Category of impactPublic policy Impact
Impact levelInternational

Countries where impact occurred

  • Canada

Sustainable Development Goals

  • SDG 3: Good Health and Well-Being