Validation of AI techniques for market research - ON-194

Preferred Disciplines: Computer Science/Technology, Computer Programming, Computer Engineering (Masters, PhD or Post-Doc)
Company: Delvinia
Project Length: 8-12 months (2 units)
Desired start date: As soon as possible
Location: Toronto, ON
No. of Positions: 3
Preferences: Colleges and Universities

About the Company: 

Delvinia is transforming the way organizations collect and use data, enabling them to make better and more informed decisions. Founded in 1998, the innovation company includes a successful portfolio of digital businesses, each with a focus on data collection. Delvinia Custom Solutions uses new and emerging technologies to help clients collect, visualize and enable data; AskingCanadians and AskingAmericans offer a range of data collection services to market researchers throughout North America, including access to an online research community of more than one million Canadians; and Methodify provides the ability to gain customer insights through an innovative online platform in as little as 12 hours. 

Project Description:

Artificial Intelligence Insight Applications (Ai-iA). The primary goal is to evaluate the use of Ai in market research. 

Problem 1: Virtual Research Concierge

Can a virtual concierge increase the quality and usage of market research?  The concierge would leverage matching algorithms and machine learning techniques to assemble the optimal set of research methodologies, data collection techniques, audience and reporting to best accomplish a researchers or marketers research goals.

Problem 2: Virtual Persona (Respondent)

Evaluate how feedback from a virtual persona compares to the feedback from a panel of human respondents. The virtual persona uses structured and unstructured data to create an interest graph that represents customer segments and can be used to test headlines, ad copy, value propositions, prices and other text-based content.

Problem 3: Virtual Analyst
Using data mining and machine learning techniques the Virtual Analyst will uncover and highlight insights and unexpected anomalies in structured and unstructured market research data including surveys, social media posts, media consumption, passive mobile metering, shopping data and more.

Research Objectives:

  • To determine if a virtual concierge can increase the quality and usage of market research.
  • To compare how feedback from a virtual persona compares to the feedback from a panel of human respondents.  To determine what market research methods could be use virtual personas.
  • To determine if a Virtual Analyst can provide the same level of insights as a market research analyst.


  • To be discussed.

Expertise and Skills Needed:

  • Data science
  • Understanding of market research methodologies
  • Experience creating and executing AI strategies

For more info or to apply to this applied research position, please

  1. Check your eligibility and find more information about open projects
  2. Interested students need to get the approval from their supervisor and send their CV along with a link to their supervisor’s university webpage by applying through the webform.