Dr. Jamell Dacon

Computer Science · Health AI · Program Leadership

Dr. Jamell Dacon Data Scientist, AI Researcher & Technical Program Leader

Dr. Dacon is a Ph.D. data scientist advancing NLP, large language models, and trustworthy machine learning for socially consequential domains, including health disparities and clinical decision support, and learning analytics. He pairs this research with a record of leading cross-functional technical teams and complex portfolios.

Explore the research Academic CV TPM Resume
Morgan State UniversityMINDS Lab DirectorCEAMLS Research FacultyGoogle Research Visiting Faculty / ML Engineer
Dr. Jamell Dacon
CRA x Microsoft Trustworthy AI Research Fellow

Primary research pillar

Health AI & Trustworthy
Machine Learning

Dr. Dacon develops human-centered AI methods for sociodemographic and clinical settings where reliability, fairness, and context matter as much as model performance.

01

Clinical decision support

Personalized comorbidity management for non-communicable diseases (NCDs), alongside cohort discovery for more useful clinical insights.

Health AI & Biomedical Systems Search related work
02

LLM evaluation

Studying supervision and evaluation of language models under distribution shift, with attention to dependable use in high-stakes workflows.

Trustworthy NLP, LLMs & AI Safety Search related work
03

Health equity & sociodemographic bias

Examining how population context and representation shape model behavior, with an emphasis on identifying and reducing inequitable outcomes.

Computational Social Science & Fairness Search related work
Research supported through NIH AIM-AHEADNCATSChan Zuckerberg InitiativeNovartis / TMCF

Explore MINDS Lab research and projects

Beyond the lab

Technical Program Management
& Industry Leadership

Dr. Dacon translates ambitious technical goals into coordinated delivery, bringing research, engineering, and stakeholder priorities into alignment across academia, industry, and government.

01 / BIG TECH & FORTUNE 500

Google Research / Kimberly-Clark

Research & Machine Learning Scientist. Audited lexical diversity across 1M+ hours of multimodal video data at 200% faster processing speeds, and built automated ML pipelines predicting supply chain OTIF failures with over 90% accuracy.

02 / NATIONAL SECURITY

U.S. Army DEVCOM

Chemical Biological Center · Lead Data Scientist and project consultant, supporting technical validation, delivery, and engineering process standardization for defense work.

03 / RESEARCH DIRECTORSHIP

MINDS Lab & Morgan State

Assistant Professor of Computer Science and Director of the MINDS Lab, with research faculty responsibilities at CEAMLS and leadership across multidisciplinary technical portfolios.

Program leadership capabilities

  • End-to-end lifecycle delivery
  • Cross-functional teams (5-20+ Staff)
  • Executive advisory for 100+ leaders
  • Joint-venture & contract governance
  • Agile / Scrum delivery
  • Enterprise System Architecture
  • Data Pipelines

Research output

Ideas, evidence
& impact

Dr. Dacon’s published conference papers, journal articles, and preprints are organized by research area. Supplied paper and publisher links are attached to the corresponding titles.

Oral / spotlight presentation · Poster presentationGoogle Scholar profile

Author markers: * co-first author; † undergraduate researcher mentored by Dr. Dacon; ‡ graduate researcher mentored by Dr. Dacon. Acceptance rates and presentation labels are shown as supplied.

01

Health AI & Biomedical Systems

Conference proceedings

  1. Optimizing Insulin Dosing for Type 1 Diabetes with Thyroid Dysfunction Using Q-Learning: A Personalized Approach to Chronic Disease Management

    Jamell Dacon, Chukwulenyeudo Uwaeme†, Chukwuemeka Obasi†, Oluwasegun Soji-John†, Oluwatobi Olajide†, Iyinoluwa Ayodele†, Oluwajomiloju King†, Chelsea Minard†, Michael Mosuro†, Marissa Savage†, Obaloluwa Wojuade†, Nicholaus Somerville†, Mikayla Brown†, Abdulai Thomas Hallowell†, Nyah Nunnally†.

    AAAI 2026 · Oral · Best Paper Runner-Up · Acceptance rate 26%

  2. Manifold-Informed Cohort Discovery (MICD): A Framework for Uncovering Latent Risk Signals in Imbalanced Healthcare Data

    Jamell Dacon, Chelsea Minard†, Chukwuemeka Obasi†, Michael Mosuro†, Iyinoluwa Ayodele†, Oluwasegun Soji-John†, Oluwatobi Olajide†, Chukwulenyeudo Uwaeme†.

    AAAI 2026 · Poster · Acceptance rate 26%

  3. BIO-DQNA: Meta-Learning and Contrastive Reinforcement Learning for Personalized Comorbidity Management in Type 1 Diabetes and Hypertension

    Jamell Dacon, Ricky Gole‡, Anuva Nuzhat†, Holy Agyei†, Obaloluwa Wojuade†, Mikayla Brown†.

    IEEE BIBM 2025 · Oral · Acceptance rate 19%

  4. Diagnosing and Mitigating Modality Dominance in Multi-Omic Antidepressant Response Prediction

    Ricky Gole‡, Oluwasegun Soji-John†, Gabriel Chambers†, Sharon Iweha†, Peyton Blake†, and Jamell Dacon.

    IEEE BIBM 2026 · Oral · Acceptance rate 19%

  5. Towards Data-Driven Diabetes Care: Identifying Key Biomarkers and Risk Factors for Type 2 Diabetes through AI Models

    Mikayla Brown†, Oluwatobi Olajide†, Michael Mosuro†, Okikioluwa Popoola†, Iyinoluwa Ayodele†, Nyah Nunnally†, Obaloluwa Wojuade†, Oluwatomiwa Baruwa†, Nicholaus Somerville-Edordu†, Abimbola Ologurun†, and Jamell Dacon.

    Society for Epidemiologic Research (SER) 2025 Mid-Year Meeting · Poster · NSEA 3rd Place Award · Non-archival

  6. Exploring Socioeconomic and Demographic Factors in Coronary Artery Disease: Using AI and Knowledge Graphs to Identify Healthcare Inequities

    Chelsea Minard†, Chukwuemeka Obasi†, Michael Mosuro†, Iyinoluwa Ayodele†, Oluwasegun Soji-John†, Oluwatobi Olajide†, and Jamell Dacon.

    SER 2025 Mid-Year Meeting · Poster · Non-archival

  7. Improving Hypertension Prediction and Management through AI: A Focus on Socioeconomic, Environmental, and Demographic Influences

    Chukwuemeka Obasi†, Chelsea Minard†, Oluwasegun Soji-John†, Thomas Hallowell†, Okikioluwa Popoola†, Andrew Bikay†, Oluwatomiwa Baruwa†, Abimbola Ologurun†, and Jamell Dacon.

    SER 2025 Mid-Year Meeting · Poster · Non-archival

Journal article

  1. Topological data analysis across the evolution of flowering plants reveals a core gene expression backbone that defines plant form and function

    Sourabh Palande, Joshua A.M. Kaste, Miles D. Roberts, Kenia Segura Abá, Carly Claucherty, Jamell Dacon, Rei Doko, Thilani B. Jayakody, Hannah R. Jeffery, Nathan Kelly, Andriana Manousidaki, Hannah M. Parks, Emily M. Roggenkamp, Ally M. Schumacher, Jiaxin Yang, Sarah Percival, Jeremy Pardo, Aman Y. Husbands, Arjun Krishnan, Beronda L. Montgomery, Elizabeth Munch, Addie M. Thompson, Alejandra Rougon-Cardoso, Daniel H. Chitwood, and Robert VanBuren.

    PLOS Biology 2023

Preprint

  1. The Topological Shape of Gene Expression across the Evolution of Flowering Plants

    Sourabh Palande, Joshua A.M. Kaste, Miles D. Roberts, Kenia Segura Abá, Carly Claucherty, Jamell Dacon, Rei Doko, Thilani B. Jayakody, Hannah R. Jeffery, Nathan Kelly, Andriana Manousidaki, Hannah M. Parks, Emily M. Roggenkamp, Ally M. Schumacher, Jiaxin Yang, Sarah Percival, Jeremy Pardo, Aman Y. Husbands, Arjun Krishnan, Beronda L. Montgomery, Elizabeth Munch, Addie M. Thompson, Alejandra Rougon-Cardoso, Daniel H. Chitwood, and Robert VanBuren.

    bioRxiv 2023

02

Trustworthy NLP, LLMs & AI Safety

Conference proceedings

  1. Detecting and Mitigating Convergent Failures in Clinical LLM Ensembles

    Ricky Gole‡* and Jamell Dacon*.

    IEEE International Conference on Data Mining (ICDM) 2026 · Oral · Acceptance rate 19.94%

  2. Beyond Accuracy: A Type-Aware Taxonomy and Evaluation Framework for Prompted Hallucination in Forensic Multimodal LLMs

    Opeyemi Adeniran, Jamell Dacon, Derrick Cook, Temitope Ajibola, Kelechi Nwachukwu, Peter Taiwo, and Kofi Nyarko.

    IEEE ICDM 2026 · Oral · Acceptance rate 19.94%

  3. Prescriptive Persistence: Quantifying the Breakdown in Human-AI Pedagogical Co-Regulation in ELL Writing Feedback

    DeBoris Leonard‡*, Ricky Gole‡*, and Jamell Dacon.

    ACM Conference on Learning at Scale (L@S) 2026 · Oral · Acceptance rate 23%

  4. Semantic Gap in Behavioral Embeddings: Why Linear Methods Fail for Educational RAG in Mathematics

    Ricky Gole‡ and Jamell Dacon.

    International Conference on Educational Data Mining (EDM) 2026 · Oral · Acceptance rate 23%

  5. Evaluating and Mitigating Inherent Linguistic Bias of African American English through Inference

    Jamell Dacon, Haochen Liu, and Jiliang Tang.

    29th International Conference on Computational Linguistics (COLING) 2022 · Oral · Acceptance rate 28.1% · Spotlight Talk at UMASS

  6. Towards a Deep Multi-layered Dialectal Language Analysis: A Case Study of African-American English

    Jamell Dacon.

    Workshop on Bridging Human-Computer Interaction and NLP (NAACL) 2022 · Oral

  7. Detecting Harmful Online Conversational Content towards LGBTQ+ Individuals

    Jamell Dacon, Harry Shomer, Shaylynn L.A. Crum-Dacon, and Jiliang Tang.

    Queer in AI Workshop (NAACL) 2022 · Oral · Spotlight Talk

  8. Using Inference to Mitigate Linguistic Bias Against African American English

    Jamell Dacon, Haochen Liu, and Jiliang Tang.

    8th International Conference on Computational Social Science (IC²S²) 2022 · Oral · Non-archival · Spotlight Talk · Acceptance rate 24.1%

  9. Does Gender Matter? Towards Fairness in Dialogue Systems

    Haochen Liu, Jamell Dacon, Wenqi Fan, Hui Liu, Zitao Liu, and Jiliang Tang.

    28th International Conference on Computational Linguistics (COLING) 2020 · Oral · Spotlight Talk · Acceptance rate 33.4%

03

Computational Social Science & Fairness

Conference proceedings

  1. Beyond Race and Gender: A Look at Sociodemographic Biases Toward Persons with Disabilities

    Jamell Dacon and Jiliang Tang.

    9th International Conference on Computational Social Science (IC²S²) 2023 · Oral · Non-archival · Acceptance rate 27%

  2. Can We Identify and Dismantle “ISMs” that Plague Our Society: An Online Approach

    Jamell Dacon and Jiliang Tang.

    9th IC²S² 2023 · Poster · Non-archival · Acceptance rate 27%

  3. Examining Word Representations between #BlackLivesMatter Movement and its Counter Protests: 2013 to 2020

    Jamell Dacon and Jiliang Tang.

    8th IC²S² 2022 · Poster · Non-archival · Acceptance rate 24.1%

  4. Understanding African American English on a Token-Level Beyond Accuracy

    Jamell Dacon.

    8th IC²S² 2022 · Poster · Non-archival · Acceptance rate 24.1%

  5. Detecting and Examining Gender Bias in News Articles

    Jamell Dacon and Haochen Liu.

    7th IC²S² 2021 · Poster · Non-archival

  6. Does Gender Matter in the News? Detecting and Examining Gender Bias in News Articles

    Jamell Dacon and Haochen Liu.

    Companion Proceedings of the 30th International Web Conference (WWW) 2021 · Oral · Acceptance rate 22%

Journal article

  1. Link and Interaction Polarity Predictions in Signed Networks

    Tyler Derr, Zhiwei Wang, Jamell Dacon, and Jiliang Tang.

    Social Network Analysis and Mining (SNAM) 2020

Preprints

  1. Are You Worthy of My Trust? A Socioethical Perspective on the Impacts of Widespread AI Systems on the Environment and Human Society

    Jamell Dacon.

    Manuscript · 2024

  2. What Truly Matters? Using Linguistic Cues for Analyzing the #BlackLivesMatter Movement and its Counter Protests: 2013 to 2020

    Jamell Dacon and Jiliang Tang.

    Manuscript · 2021 · Keynote at UNF

Soft skills & leadership

Leadership that
moves teams forward

Dr. Dacon combines technical judgment with clear communication, inclusive mentorship, and steady coordination across disciplines and institutions.

01 / PEOPLE

Mentorship & team development

Builds multidisciplinary teams, mentors undergraduate and graduate researchers, and creates the conditions for contributors to grow into ownership.

02 / COMMUNICATION

Executive communication

Turns complex technical decisions into concise briefings and actionable roadmaps for more than 100 corporate, academic, and government leaders.

03 / COLLABORATION

Cross-functional facilitation

Aligns researchers, engineers, agency partners, and stakeholders around shared outcomes, clear responsibilities, and workable decisions.

04 / INCLUSIVE PRACTICE

Equity-minded leadership

Brings attention to representation, accessibility, and disparate impact in both AI research and the teams designing high-stakes systems.

05 / DELIVERY

Strategic judgment & accountability

Balances scope, schedule, risk, and quality while maintaining transparent governance across complex technical portfolios and joint ventures.

06 / ADAPTABILITY

Calm execution under change

Guides teams through shifting constraints and evolving evidence, keeping decisions grounded in mission, user needs, and delivery realities.