Grant: Alliance Grants
Toronto Metropolitan University
$25,000.00
Fortifying Online Recommendation Systems Against Complex Multilingual Obfuscated Attacks
EF 2024-25
Résumé
Effective recommendation systems in commerce hold significant promise for enhancing cross-selling opportunities, increasing average order values, boosting sales revenue, and improving customer satisfaction. By optimizing resource consumption, these systems contribute to overall business efficiency. However, the widespread adoption of user-adaptive recommendation systems (RSs) in open platforms and information access domains like e-commerce has also attracted malicious actors. These individuals employ "shilling attacks," injecting fake profiles into RSs to manipulate recommendation rankings for economic gain. Ensuring the integrity and fairness of these recommendations is paramount, necessitating the development of robust systems capable of maintaining stability even in the face of such attacks. Current detection methods, while focused on enhancing accuracy, often fall short in identifying malicious users and targeted items during obfuscated attacks. Attackers frequently add noise, shift targets, modify user ratings, or use varied natural language to conceal their activities and evade detection. Notably, there has been an uptick in spamming attacks conducted in Arabic, aiming to disrupt the Canadian e-commerce market. This international collaboration with British University in Dubai is essential to address and develop effective strategies for understanding and combating multi-lingual shilling attacks, particularly those conducted in Arabic. The innovative models and methods emerging from this collaboration are designed to: 1) Protect both businesses and consumers, 2) Enhance the customer shopping experience, and 3) Increase sales and revenue. The primary goal of this initiative is to create improved marketplace opportunities that empower businesses to strengthen their e-commerce presence and digitize their operations. This will enable them to perform more effectively in the Canadian retail e-commerce market through comprehensive training of highly qualified personnel (HQP) in cutting-edge techniques
Programme
Alliance Grants
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
597647-2024
Name nom
Kashef, Rasha RK
Department département
Electrical, Computer and Biomedical Engineering
Organization id
43
Institution établissement
Toronto Metropolitan University
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
25000
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
2801
Research subject group en
Artificial intelligence
Research subject group fr
Intelligence artificielle
Research subject en
Natural language and speech understanding
Research subject fr
Langage naturel et reconnaissance de la parole
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
Fortifying Online Recommendation Systems Against Complex Multilingual Obfuscated Attacks
Keywords
large language models; mathematical modeling; multi-lingual; natural language processing; obfuscated attacks; recommendation systems
Application summary
Effective recommendation systems in commerce hold significant promise for enhancing cross-selling opportunities, increasing average order values, boosting sales revenue, and improving customer satisfaction. By optimizing resource consumption, these systems contribute to overall business efficiency. However, the widespread adoption of user-adaptive recommendation systems (RSs) in open platforms and information access domains like e-commerce has also attracted malicious actors. These individuals employ "shilling attacks," injecting fake profiles into RSs to manipulate recommendation rankings for economic gain. Ensuring the integrity and fairness of these recommendations is paramount, necessitating the development of robust systems capable of maintaining stability even in the face of such attacks. Current detection methods, while focused on enhancing accuracy, often fall short in identifying malicious users and targeted items during obfuscated attacks. Attackers frequently add noise, shift targets, modify user ratings, or use varied natural language to conceal their activities and evade detection. Notably, there has been an uptick in spamming attacks conducted in Arabic, aiming to disrupt the Canadian e-commerce market. This international collaboration with British University in Dubai is essential to address and develop effective strategies for understanding and combating multi-lingual shilling attacks, particularly those conducted in Arabic. The innovative models and methods emerging from this collaboration are designed to: 1) Protect both businesses and consumers, 2) Enhance the customer shopping experience, and 3) Increase sales and revenue. The primary goal of this initiative is to create improved marketplace opportunities that empower businesses to strengthen their e-commerce presence and digitize their operations. This will enable them to perform more effectively in the Canadian retail e-commerce market through comprehensive training of highly qualified personnel (HQP) in cutting-edge techniques
