Using Supervised Machine Learning to Classify Customer Input

Slides:



Advertisements
Similar presentations
SANITATION THE FOUNDATION OF FOOD SAFETY
Advertisements

New York State Department of Agriculture and Markets
Collaborative Efforts of Federal, State, and Local Public Health Partners in Foodborne Illness Investigations United States Public Health Commissioned.
Memristor in Learning Neural Networks
A Brief Overview of Neural Networks By Rohit Dua, Samuel A. Mulder, Steve E. Watkins, and Donald C. Wunsch.
Clemson University’s statewide network of Public Service Activities (PSA) is the only state agency that conducts research and technology transfer to support.
Artificial Neural Networks
Machine Learning: Connectionist McCulloch-Pitts Neuron Perceptrons Multilayer Networks Support Vector Machines Feedback Networks Hopfield Networks.
Government Agencies HUM-FNW-3 Unit 4. USDA United States Department of Agriculture Mission Statement We provide leadership on food, agriculture, natural.
Who’s Minding the Store Downunder: The regulator protecting public health and safety. Deon Mahoney Food Standards Australia New Zealand Seattle University.
14-1 Food Safety Regulation and Standards Apply Your Knowledge: Test Your Food Safety Knowledge 1.True or False: The Food and Drug Administration.
Food Safety and Government Regulations Food Animal Quality Assurance Youth Curriculum Guide.
FOOD SAFETY RELATED TOPICS FOR DISCUSSION DURING WEEKLY MEETINGS
1 Webinar on: Establishing a Fully Integrated National Food Safety System with Strengthened Inspection, Laboratory and Response Capacity Sponsored by Partnership.
SAFE Safe Animal Feed Education Program 2003 California Department of Food & Agriculture Agricultural Commodities & Regulatory Services Branch.
U.S. Department of Agriculture Office of Food Safety Food Safety and Inspection Service Winston Felton, D.V.M. Dearborn Circuit Supervisor Madison District.
NACMPI November 15-16, 2005 Risk-Based Inspection Dr. Barbara Masters Administrator, Food Safety and Inspection Service Philip Derfler Assistant Administrator,
Labeling and Program Delivery Division USDA, FSIS, OPPD
April 11, 2008 “Who’s Minding the Store? The Current State of Food Safety and How It Can Be Improved” Conference sponsored by Marler Clark and Stoel Rives.
Food Safety Regulation and Standards
Images shutterstock.com Food Science: An Old but New Subject Chapter 1 Remind Ms J to clock in.
NS 440 LEGAL AND REGULATORY ENVIRONMENT IN FOOD PRODUCTION SPRING YOUNTS DAHL, MS PHD INSTRUCTOR Unit 5: Policy Considerations in Food Regulation.
Appendix B: An Example of Back-propagation algorithm
Chapter 1-How Food Affects Life Factors that Affect the Food Supply.
LINEAR CLASSIFICATION. Biological inspirations  Some numbers…  The human brain contains about 10 billion nerve cells ( neurons )  Each neuron is connected.
Purpose of the USDA  Established in 1862 by President Abraham Lincoln  Back then, more than half of the Nation’s population lived and worked on farms.
United States Department of Agriculture Food Safety and Inspection Service 1 National Advisory Committee on Meat and Poultry Inspection August 8-9, 2007.
Food Safety …From Farm to Table By: Allison Weis
SAFEGUARDING THE FOOD SUPPLY HOW DO WE PROTECT THE FOOD SUPPLY FOR AN ENTIRE NATION?
UNITED STATES DEPARTMENT OF AGRICULTURE FOOD SAFETY AND INSPECTION SERVICE FSIS Directive PRIOR LABELING APPROVAL 1 District Correlation Meeting.
The Food Supply Factors that affect the food supply.
CSC 4510 – Machine Learning Dr. Mary-Angela Papalaskari Department of Computing Sciences Villanova University Course website:
UNITED STATES DEPARTMENT OF AGRICULTURE FOOD SAFETY AND INSPECTION SERVICE Nutrition Labeling of Single- Ingredient Products and Ground or Chopped Meat.
Stakeholder Matrix START IT!. Stakeholder Matrix.
Health Inspections FOOD HANDLING, HACCP, FDA, AND WHAT HEALTH SCORES REALLY MEAN.
Using askFSIS FSIS Directive /7/2015Policy Development Division1 United States Department of Agriculture Food Safety and Inspection Service.
OHIO STATE UNIVERSITY EXTENSION go.osu.edu/AQCA Youth Food Animal Quality Assurance Curriculum Guide Food Safety and Government Relations.
Data Acquisition to Anticipate Foodborne Hazards National Advisory Committee on Meat and Poultry Inspection Sean Altekruse DVM, PhD November 16, 2004.
FACTORS THAT AFFECT THE FOOD SUPPLY AFFECT WHAT YOU CHOOSE WHEN YOU GO TO THE STORE.
Artificial Neural Networks (ANN). Artificial Neural Networks First proposed in 1940s as an attempt to simulate the human brain’s cognitive learning processes.
 Foodservice Standards equals “Quality.”  Standards are established models or examples used to compare quality.  Meet expectations so management &
2 Objectives 1.To identify and distinguish among government agencies and their role in food safety 2.To analyze and apply the laws set by government agencies.
Food Science: An Old but New Subject
FOOD FOR GROUPS: Food Safety Awareness Campaign Ellie Chan and Karina O’Loughlan.
Machine Learning Supervised Learning Classification and Regression
Human Services Delivery Systems and Organizations
Human Services Delivery Systems and Organizations
Food Safety and Government Relations
Deep Learning Amin Sobhani.
FDA Inspections Stephen Joseph Joy’s Quality Management Systems
The American Public Health Association’s 2007 Annual Meeting & Exposition November 7, 2007.
Food Safety Regulations and Standards
Human Services Delivery Systems and Organizations
Self organizing networks
Human Services Delivery Systems and Organizations
Machine Learning Today: Reading: Maria Florina Balcan
Instructor Notes There is no DVD associated with this topic.
اهمیت تجزیه و تحلیل خطر در ایمنی مواد غذایی
From Lab to Label: Innovations That Feed The World
Supervised vs. unsupervised Learning
Response Teams – Planning and Preparation
Structure of a typical back-propagated multilayered perceptron used in this study. Structure of a typical back-propagated multilayered perceptron used.
COSC 4335: Part2: Other Classification Techniques
What Would You Do? Ethics in Restaurant Management.
What Would You Do? Ethics in Culinary Arts.
Machine Learning Perceptron: Linearly Separable Supervised Learning
Deep Learning Authors: Yann LeCun, Yoshua Bengio, Geoffrey Hinton
Food Safety Management Systems
Unsupervised Machine Learning: Clustering Assignment
Mrs. Senick Perspectives in Art
Presentation transcript:

Using Supervised Machine Learning to Classify Customer Input Adrianna Steers-Smith: Statistician at Food Safety Inspection Service (FSIS) Background The Food Safety and Inspection Service (FSIS) is the public health agency in the U.S. Department of Agriculture responsible for ensuring that the nation's commercial supply of meat, poultry, and egg products is safe, wholesome, and correctly labeled and packaged. FSIS receives more than 50,000 policy related food safety questions from consumers, the food production industry, and Agency’s inspection staff through askFSIS. Methods Model: Multilayer Perceptron askFSIS AskFSIS is a resource available to the public that provides access to FSIS policy knowledge through Frequently Asked Questions and a portal that allows direct e-mail communication with a FSIS policy subject matter expert. Every e-mail to FSIS, and response from FSIS is saved within the application. The consumer routes their question to the subject expert (SME) by selecting a topic. Topics are assigned to each policy staff. Parameters: Bag of Words Tokenizer Gradient Decent Optimizer Sigmoid Function Activation Discussion/Results While there are other models to use for this type of classification problem using Bag of Words to Tokenize, sigmoid function for activation and gradient descent (a=0.01) for optimization with one hidden layer (20 nodes) produced a model that can with 82% accuracy, not much less than our intern, classify our complex data. For future studies we are exploring unsupervised clustering for training sets and expanding our classification problem to include all askFSIS data. ANN was used in this analysis, but we are exploring the value of LSTM for type of data. Purpose of Work Classifying and Analyzing trends in askFSIS data will refine our existing trend analysis process by reducing subjectivity and manual effort. This will allow FSIS to more rapidly clarify and improve policy guidance. Rapid responses will improve customer service and potentially reduce foodborne illness.