Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Master's Award: Canada Graduate Research Scholarships - Master's program

McGill University

$27,000.00

Statistical Network Inference and Analysis to Characterize Co-Purchasing Patterns in Quebec from Large-Scale Grocery Transaction Data

FY 2025-26

Summary
Dietary pattern is associated with numerous chronic diseases. It is well understood that diets, rather than individual food items, are more important risk factors for health outcomes. Past studies summarized dietary patterns in data using subjective summaries such as the Healthy Eating Index, resulting in a loss of information. In order to avoid this, some researchers have begun to estimate and visualize dietary patterns as interconnected networks of food groups. These provide more interpretable summaries of the dietary patterns, as the structure of the network describes how pairs of food groups influence each other. For example, if chips and soda are connected in this network, it means that purchasing chips makes you more likely to have purchased soda. This would allow us to see which food groups are associated with unhealthy diets. However, these networks are typically built using large-scale food-frequency surveys. We will extend these methods to allow us to instead infer dietary pattern networks from grocery transaction data. We will use data from a major Quebec grocery retailer, with approximately 30 thousand food items contained in 300 million transactions from 1.5 million shoppers over a 3-year period. By grouping food items into 53 categories, and combining transactions into weekly purchases for each customer, we will construct a network for each week of the 3-year period. This will allow us to not only study dietary patterns for any given week, but to also study how these dietary patterns change over time. Overall, this project will allow researchers to conduct large scale nutritional assessments when food frequency questionnaires are too costly to conduct.
Program
Master's Award: Canada Graduate Research Scholarships - Master's program
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database

As published

Id
572785
Name
McCormick James
Region
Québec
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Statistical Network Inference and Analysis to Characterize Co-Purchasing Patterns in Quebec from Large-Scale Grocery Transaction Data
Programtype2
Studentship Programs
Orgnameinp2
McGill University
Cihramount2
$27,000
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Research Scholarships - Master's program
Keyworddelim
Dietary Pattern Analysis; Graphical Models; Longitudinal Data; Network Inference
Abstract
Dietary pattern is associated with numerous chronic diseases. It is well understood that diets, rather than individual food items, are more important risk factors for health outcomes. Past studies summarized dietary patterns in data using subjective summaries such as the Healthy Eating Index, resulting in a loss of information. In order to avoid this, some researchers have begun to estimate and visualize dietary patterns as interconnected networks of food groups. These provide more interpretable summaries of the dietary patterns, as the structure of the network describes how pairs of food groups influence each other. For example, if chips and soda are connected in this network, it means that purchasing chips makes you more likely to have purchased soda. This would allow us to see which food groups are associated with unhealthy diets. However, these networks are typically built using large-scale food-frequency surveys. We will extend these methods to allow us to instead infer dietary pattern networks from grocery transaction data. We will use data from a major Quebec grocery retailer, with approximately 30 thousand food items contained in 300 million transactions from 1.5 million shoppers over a 3-year period. By grouping food items into 53 categories, and combining transactions into weekly purchases for each customer, we will construct a network for each week of the 3-year period. This will allow us to not only study dietary patterns for any given week, but to also study how these dietary patterns change over time. Overall, this project will allow researchers to conduct large scale nutritional assessments when food frequency questionnaires are too costly to conduct.
Competitiondate
202512
Approvedterm2
1 yr 0 mth
Pinamesdelim
McCormick, James
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $27,000~ $0~