Determing client engagement through machine learning - QC-105
Preferred Disciplines: Mathematics; Masters and above
Project length: 4 to 12 months
Desired start date: As soon as possible
Location: Montreal, Quebec
Preferences: No preferences for language
Start-up fintech company revolutionizing digital customer engagement
Contrary to banners, intercept pages and pop-ups, the company’s solution neither forces a user to go through undesired content or presents offers relentlessly, instead choosing the right time to present the most useful offer to the individual user. The focus of this project is finding the “right time”.
The first steps to reach that goal are to be able to accurately predict :
- When will the user login again ?
- At login, will this be a long session ?
- At login time, is the user going to perform a specific activity during the session ?
- Is the user going to do something else in the site before leaving ?
Data analysis will be performed on more than 3 million active data files.
Background and required skills
Several tasks need to be accomplished and challenges overcome by the team of which the researcher will be a part including :
- Build and evaluate different machine learning models to answer the different questions
- Build clustering techniques to group users by behavior
- Optimize the machine learning algorithms so that the predictive models can be built and be updated in reasonable time
- Build different models/scenarios to test and maximize sensitivity or specificity depending on the needs of the offer
- Find the best technologies to suit those goals
- The amount of variables involved in the prediction can also pose a challenge as some variables will not be available at all times and some others are more difficult to capture
- To be determined
Expertise and Skills Needed:
Mathematics, probability & statistics
For more info or to apply to this applied research position, please
- Check your eligibility and find more information about open projects.
- Complete this webform. You will be asked to upload your CV. Remember to indicate the title of the project(s) you are interested in and obtain your professor’s approval to proceed!
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 or directly to Jean-Philippe Valois at, firstname.lastname@example.org