Our company advertises thousands of products through a user interface. We would like to work with students to create a new recommendation system that returns products based on what a user has previously viewed or rated. The recommendations should also consider products viewed by other users who are like the given user.

This will involve several different steps for the students, including:

• Analyzing our existing dataset of users, products, and reviews.
• Developing a recommendation engine software.
• Optimizing software runtime performance and assessing areas for improvement.
• Researching other variables that can improve the quality of product recommendations.
• Accounting for additional variables in the recommendation engine software.
• Testing the developed software and making improvements based on additional data.

By the end of the project, students should demonstrate:

• Understanding of our existing dataset of users, products, and reviews
• Understanding of variables that affect the quality of product recommendations
• Identification of areas for future improvement of the software

Bonus steps would include:

• Testing the developed software with users and accounting for additional parameters

Final deliverables should include

• All source code.
• A written report explaining the design process and outcomes.

Students will connect directly with the employer for mentorship and supervision throughout the project.

In order to complete this project, students can self-teach, but it is beneficial to be familiar with:

• Recommendation engine development and data analysis
• Machine learning languages like Python

Again, students can complete this project without any of the prior knowledge listed here. Students will be expected to research and learn more about the above as the project goes along.

We can provide students with:

• Our existing dataset of users, products, and ratings
• A walkthrough / tutorial of our existing recommendation engine software