top of page

AI user experience

Experience Design 2 Course

Spring 2023

Under Professor Tad Hirsch

With Linda Zeng, Kiera Huang, and Victoria Chen

The goal of this project was to create an experience utilizing Artificial Intelligence, bringing the in-person shoe shopping experience, online. In-person shoe stores often lack the variety that is offered online, and many people don’t have access to specialty stores, such as those carrying shoes for running. While shopping online, people would often buy several pairs of shoes, try them on, and return the ones they don’t like or don’t fit.

Through the Trekka app, users are able to scan their feet to be matched to a short list of shoes. User feedback allows the app to learn and recommend better shoes over time.

With the rise of artificial intelligence and machine learning, there is an obvious question of ethics. Our main approach is to have data anonymized so that data cannot be shared with other users or third parties. We also emphasize that the goal is to provide shoes that fit and feel comfortable, rather than help with any existing medical conditions. 

User Profile

concept development
image 66.png

Initial Brainstorming

The project began by targeting an AI/ML application to drive our solution. After studying its affordances, limitations, and associated activities, the team ultimately decided to design a recommender system to improve the shoe shopping experience.

Concept Refinement

Preliminary research was conducted to evaluate the feasibility and desirability of the concept.
In consideration, the AI technology would require training data to define ideal footwear compatibilities, and how to process foot dimensions.

user flows
Inboarding copy.jpg
user map
Journey Map with WFs 1.png

The team created a journey map to identify the goals, actions and thoughts of the user to understand their needs. The observations of the experience was organized by phases: Trigger, Search, Evaluation, Selection, and Purchase.

Based on our analysis, we determined that the Evaluation and Selection stages were the most suitable for AI augmentation. The best stage for AI automation was the Evaluation stage. Therefore, we will focus on the Evaluation and Selection stages for our design project.

We focused on these stages to build a clear idea of how Trekka should function, and what values are involved.

 

discovery
PREBOARDING - 01_3x.png

The team created an app storefront mockup to show how Trekka would reach out to users.

scanning process
Scanning Process.jpg

Trekka’s scanning is performed with a smartphone. Advanced computer vision and machine learning algorithms are used to create powerful 3D foot models. The technology can provide precise measurements of arch height, length, and width, which can be used to recommend appropriate footwear and support health.

user feedback survey
User Survey.png

Users will be offered a short survey with an incentive reward. The goal of the survey is to better understand the accuracy of measurement, and the quality of the recommendation, and reflect the data in the system.

bottom of page