Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Alliance Grants

University of Toronto

$132,195.00

Novel Statistical and Machine Learning Algorithms for Analysis of Canadian Political Donations and Lobbying Data

EF 2024-25

Résumé
Our proposed project would bring together experts on data validation (Rohan Alexander), entity resolution (Laks V.S. Lakshmanan), statistics (Jesse Gronsbell), and open-source software (Shurui Zhou) to develop new methods, using Large Language Models (LLMs), for working with messy, real-world public interest datasets. These scholars will collaborate with the Investigative Journalism Foundation (IJF), which maintains large datasets of political donations and lobbying in Canada. Specifically, using IJF data, we will develop novel methods for solving two common problems faced by anyone working with messy real-world data. First, led by Dr. Lakshmanan, we will develop novel methods for entity resolution: linking together entities (e.g., companies, people) in datasets which go by different pseudonyms, but are actually the same real entity. Methods employed for dealing with malicious pseudonyms need to be robust against deliberate use of alternate representations compared to benign pseudonyms, e.g., those resulting from typos and differences in conventions. Unlike previous methods for entity resolution, ours will be scalable to many possible links, explainable in terms of intuitive parameters, and produce estimates of uncertainty for each link. Second, led by Dr. Alexander, we will develop a framework for automated data validation using LLMs to automatically generate expectations and validation tests for data. Unlike standard data validation workflows, ours will enable more rapid and comprehensive data validation through LLM assistance. These methods will produce immediate value by revealing hidden patterns in the IJF's public interest datasets, but will also produce long-term value as we will package our methods as open-source software that others can use on any data. Both projects will empower the public, policymakers, and academic researchers to more deeply understand the nature of political influence and government decision-making, ultimately strengthening Canadian democracy.
Programme
Alliance Grants
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
599949-2024
Name nom
Alexander, Rohan RP
Department département
Information, Faculty of
Organization id
31
Institution établissement
University of Toronto
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
132195
Program id
ALLRP
Program name en
Alliance Grants
Program name fr
Subventions Alliance
Group en
RESEARCH PARTNERSHIPS
Group fr
PARTENARIATS DE RECHERCHE
Committee code
96
Committee name en
RPP Internal Decision Cttee
Committee name fr
Cté de décision interne - PPR
Area of application code
802
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Information systems and technology
Area of application fr
Systèmes d'information et technologie de l'information
Research subject code
2711
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Database management
Research subject fr
Gestion de bases de données
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Novel Statistical and Machine Learning Algorithms for Analysis of Canadian Political Donations and Lobbying Data
Keywords
data cleaning; data validation; data validation testing; entity resolution; large language models; natural language processing; probabilistic record linkage; record linkage
Application summary
Our proposed project would bring together experts on data validation (Rohan Alexander), entity resolution (Laks V.S. Lakshmanan), statistics (Jesse Gronsbell), and open-source software (Shurui Zhou) to develop new methods, using Large Language Models (LLMs), for working with messy, real-world public interest datasets. These scholars will collaborate with the Investigative Journalism Foundation (IJF), which maintains large datasets of political donations and lobbying in Canada. Specifically, using IJF data, we will develop novel methods for solving two common problems faced by anyone working with messy real-world data. First, led by Dr. Lakshmanan, we will develop novel methods for entity resolution: linking together entities (e.g., companies, people) in datasets which go by different pseudonyms, but are actually the same real entity. Methods employed for dealing with malicious pseudonyms need to be robust against deliberate use of alternate representations compared to benign pseudonyms, e.g., those resulting from typos and differences in conventions. Unlike previous methods for entity resolution, ours will be scalable to many possible links, explainable in terms of intuitive parameters, and produce estimates of uncertainty for each link. Second, led by Dr. Alexander, we will develop a framework for automated data validation using LLMs to automatically generate expectations and validation tests for data. Unlike standard data validation workflows, ours will enable more rapid and comprehensive data validation through LLM assistance. These methods will produce immediate value by revealing hidden patterns in the IJF's public interest datasets, but will also produce long-term value as we will package our methods as open-source software that others can use on any data. Both projects will empower the public, policymakers, and academic researchers to more deeply understand the nature of political influence and government decision-making, ultimately strengthening Canadian democracy.