Data Management and Analytics Professional Development Group

Webinars

• AI in Food Safety Research: A Focus on Algorithms and Their Applications Part 4 Nov 6, 2023

  Organized by IAFP's Data Management and Analytics PDG and Student PDG

  In the fast-moving field of artificial intelligence, new algorithms are reported practically every day. This rapid evolution, combined with the voluminous body of literature, can make understanding how to apply AI for food safety research very challenging. In this webinar series, the participants will be introduced to fundamental machine learning algorithms used in food safety research. By the end of the session, participants will have an overview of each models’ mathematical foundation, evaluation metrics, and limitations. Insight on the models’ applications in food safety will be gained via ‘case studies’ from invited speakers’ research. Participants will have the opportunity to ask questions at the end of the session.

  Part 4: Large Language Models for Foodborne Outbreak Tracking/Epidemiology

  Model(s): Large Language Models (LLMs)

  Topics:

  Talk 1: Developing LLMs for food recall prediction and safety enhancement: Georgios Makridis, University of Piraeus, Greece

  Talk 2: Probabilistic models for locating the source of foodborne outbreaks: Abigail Horne, Massachusetts Institute of Technology, USA

  In this webinar, participants will learn about machine learning models used for language and textual analysis in foodborne outbreak epidemiology. Through case-study, we will review the how to select, optimize, evaluate, and apply large language models for tracking and source attribution of foodborne outbreaks.

  Outcomes: By the end of the webinar, participants will gain an understanding of:

  • How large language models work.
  • How to select and fine-tune a LLM for downstream tasks.”
  • How the models are currently being used in food safety.

  Presenters

  • TBD
  • Claire Zoellner, Moderator
•  AI in Food Safety Research: A Focus on Algorithms and Their Applications Part 3 Oct 30, 2023

  Organized by IAFP's Data Management and Analytics PDG and Student PDG

  In the fast-moving field of artificial intelligence, new algorithms are reported practically every day. This rapid evolution, combined with the voluminous body of literature, can make understanding how to apply AI for food safety research very challenging. In this webinar series, the participants will be introduced to fundamental machine learning algorithms used in food safety research. By the end of the session, participants will have an overview of each models’ mathematical foundation, evaluation metrics, and limitations. Insight on the models’ applications in food safety will be gained via ‘case studies’ from invited speakers’ research. Participants will have the opportunity to ask questions at the end of the session.

  Part 3: Neural Networks for Biosensor Image Analysis

  Model(s): RNN,CNN, MLP

  Topics:

  Talk 1: Developing an AI-enabled biosensing pipeline for pathogen detection: Jiyoon Yi, Michigan State University, USA

  Talk 2: Analyzing optical imaging data to quantify foodborne pathogens: Luyao Ma, University of Florida, USA

  In this webinar, participants will learn about machine learning models used for image analysis in food safety research. Through case-study, we will review how to select, optimize, evaluate, and apply neural networks for the analysis of diagnostic imaging used to detect foodborne pathogens.

  Outcomes: By the end of the webinar, participants will gain an understanding of:

  • Traditional machine learning versus deep learning.
  • How neural networks work.
  • Different types of neural networks.
  • How neural networks are currently being used in food safety research.

   

  Presenters

  • Jiyoon Yi, Presenter Michigan State University
  • Luyao Ma, Presenter Florida State University
  • Zhenjiao Du, Moderator
• AI in Food Safety Research: A Focus on Algorithms and Their Applications Part 2 Oct 23, 2023

  Organized by: Data Management and Analytics PDG and Student PDG

  In the fast-moving field of artificial intelligence, new algorithms are reported practically every day. This rapid evolution, combined with the voluminous body of literature, can make understanding how to apply AI for food safety research very challenging. In this webinar series, the participants will be introduced to fundamental machine learning algorithms used in food safety research. By the end of the session, participants will have an overview of each models’ mathematical foundation, evaluation metrics, and limitations. Insight on the models’ applications in food safety will be gained via ‘case studies’ from invited speakers’ research. Participants will have the opportunity to ask questions at the end of the session.

  Part 2: Machine Learning Models for Detecting Adulterants and Food Fraud

  Model(s): tSNE (t-distributed Stochastic Neighbor Embeding), PCA (Principal Component Analysis), LDA (Linear Discriminant Analysis)

  Topics: 

  Talk 1: Leveraging dimensionality reduction for genomic biomarker discovery

  Talk 2: Comparing models for validating the origins of meat floss: Kuwat Triyana, Universitas Gadjah Mada, Indonesia

  In this webinar, participants will learn about unsupervised machine learning models used to study the spoilage and quality of food products. Through case-study, we will review how to select, optimize, evaluate, and apply algorithms such as t-Stochastic Neighbor Embedding (tSNE), Principal Component Analysis (PCA), and linear discriminant analysis (LDA) in food safety research.

  Outcomes: By the end of the webinar, participants will gain an understanding of:

  1. Unsupervised vs Supervised machine learning.

  2. How to select features for unsupervised machine learning algorithms.

  3. The advantages and limitations of each algorithm.

  4. How the models are currently being used in food safety research.

  Presenters

  • Talk 1: TBD
  • Kuwat Triyana, Presenter Universitas Gadjah Mada, Indonesia
  • Harsimran Kapoor, Moderator
• AI in Food Safety Research: A Focus on Algorithms and Their Applications Part 1 Oct 16, 2023

  Organized by: IAFP's Data Management and Analytics PDG and Student PDG

  In the fast-moving field of artificial intelligence, new algorithms are reported practically every day. This rapid evolution, combined with the voluminous body of literature, can make understanding how to apply AI for food safety research very challenging. In this webinar series, the participants will be introduced to fundamental machine learning algorithms used in food safety research. By the end of the session, participants will have an overview of each models’ mathematical foundation, evaluation metrics, and limitations. Insight on the models’ applications in food safety will be gained via ‘case studies’ from invited speakers’ research. Participants will have the opportunity to ask questions at the end of the session.

  Part 1: Regression and Classification for Predicting Microbial Responses

  Model(s): Logistic Regression, Support Vector Machine, Random Forest Models

  Topics:

  Talk 1: Evaluating algorithms for monitoring E. coli in surface water: Yakov Pachepsky, USDA-ARS, USA

  Talk 2: Comparing machine learning algorithms for genome-based classification of Salmonellaenterica disease severity: Shraddha Karanth/ Abani K. Pradhan, University of Maryland, USA

  Exploring the predictive capability of advanced machine learning in identifying severe disease phenotype in Salmonella enterica. https://doi.org/10.1016/j.foodres.2021.110817

  In this webinar, participants will learn about machine learning models often used for classifying and quantitatively predicting quantitative microbial responses of foodborne pathogens. Through case-study, we will cover how to select, optimize, evaluate, and apply logistic regression, support vector machine, and random forest models to food safety research.

  Outcomes: By the end of the webinar, participants will gain an understanding of:

  • Each models’ advantages and limitations.
  • What are hyperparameters and how to optimize each model.
  • How to compare the models’ performance using evaluation metrics.
  • How the models are currently being used in food safety research

  Presenters

  • Yakov Pachepsky, Presenter USDA-ARS
  • Matthew Stocker USDA-ARS
  • Shraddha Karanth, Presenter University of Maryland
  • Abani K. Pradhan, Presenter University of Maryland
  • Olivia Haley and Manreet Bhullar, Moderators