With the recent opening of the "Amazon Go" store in Seattle, Washington, our group was very interested in this technology. However, we realized this was only limited to Amazon and wished to create a generalized application that could be implemented on a larger scale.
What it does
Our app allows customers to scan products at a supermarket as they walk around, adding to their cart as they go. At the time of checkout, users can pay with their phone, eliminating the need for long and time-consuming lines. As items are added to the user's cart, our machine learning algorithm uses sklearn to recommend similar products to the consumer.
How we built it
After creating pencil and paper drawings, we used Sketch for design. Python was used to implement sklearn, the machine learning library we used to conduct the calculations needed for the recommendation system. Using x-code and swift, we then built the overall layout of the app and implemented the barcode scanner.
Challenges we ran into
Finding a common framework to use for our app was the main challenge. Our team ran into problems when attempting to implement python script into a Swift-based iOS app.
Accomplishments that we're proud of
We were especially proud of the successful implementation of our python script into our iOS-based application. Additionally, no team member had prior experience with scanning UPC barcodes, and being able to successfully extract information from each barcode is something we are very proud of.
What we learned
Through the difficulties we encountered throughout this process, we learned that not all barcodes are reliable for information extraction.
What's next for Express Check
In the future, we hope to implement a few new aspects into our app. First, we would like to implement GPS tracking so that customers may be charged as soon as they leave the store's boundaries. Second, we would like to expand methods of payment outside of Apple Pay, implementing other financial APIs. Last, we would like to develop methods to protect the integrity of this service, eliminating any possibility of theft or fraudulent behavior.