There isn't one of us that has not at some point gotten annoyed with themselves for spending hour after hour looking for that one item, comparing review against price in order to find the product with the best ratio. We appreciate the hard work people put into finding the best product and as such decided to make it easier for everyone.

What it does

Choose is a web application that can search a variety of sites for different categories of products such as TVs and Headphones, comparing the site's review score to its price. We provide a score and display the items in an easier to compare way, to take the hassle out of searching for that perfect product.

How we built it

A lightweight Flask server was used to host our site, so that we can interface with our Python based price/review identifier. Individual sites were inspected to extract the relevent data. We used some data analysis and looked into clustering the data within the user's price range to allow the user to select whether or not they want to aim to a more expensive product or a cheaper product but due to time constraints have not got past identifying clusters within the Python program.

Challenges we ran into

One of the challenges was collecting product information from a different website, such as Argus, John Lewis. We had to adapt to each website HTML layout to get correct data, and this meant only 3 sites were implemented in the time limit.

Accomplishments that we're proud of

We work in an efficient manner sharing all information and communicating with each other. We manage to brainstorm and create our idea into Minimum Valuable Product to test with real customers.

What we learned

Each of us learned new software and developer skills. For example, Lana enjoyed working together with people who she didn't know before. As well Nicole learned to use Flask server to display the data.

What's next for Choose

In the future, we are looking to enhance the website according to created design. As well providing easy access to customers with a mobile app. We aim to add more sites to the product, and provide a more reliable way of rating products, maybe based on number of reviews, similar item clustering based on preference and showing the user what percentage of other users bought items.

Share this project: