Inspiration
The inspiration behind BrewMetric stemmed from our love for coffee and the desire to make informed choices about what we consume. As a frequent Starbucks customer, we often found ourselves wishing for a tool that could help us navigate the extensive menu while considering our dietary preferences. This sparked the idea of creating an app that would provide detailed nutritional analysis and personalized recommendations for Starbucks beverages.
What it does
This app already has Starbucks's entire beverage menu uploaded into a dataset. It consists of all drinks ever offered, from limited to seasonal. With this app, users can filter and analyze the nutritional values for calories, protein, sugars, and sodium. (These are the most important nutritional categories that the average consumer accounts for). The app will provide a histogram to show how diverse the drinks are in terms of the nutritional category selected (aka the amount per category). It also provides a regression model, for users to see the correlation in the drinks if they were to compare 2 selected categories. The app also allows the users to select/filter the amount of each nutritional category they want in their beverage and will provide the top 5 recommended drinks to choose from. This was through the implementation of a prediction model.
How we built it
BrewMetric was developed using R for backend data processing and analysis, along with R Shiny for creating the interactive web application. We utilized Starbucks' API to gather comprehensive data on their beverages, which was then processed and analyzed using R scripts.
With R Shiny, we designed and implemented a user-friendly interface that allowed users to filter drinks based on their dietary preferences, view detailed nutritional information, and receive personalized recommendations. The combination of R and R Shiny enabled us to create a powerful and intuitive tool for Starbucks customers to make informed decisions about their beverage choices.
Challenges we ran into
One of the primary challenges we encountered was ensuring the accuracy and reliability of the data obtained from the Starbucks API. As the data directly impacted the nutritional analysis and recommendations provided by BrewMetric, any discrepancies or inaccuracies had to be meticulously addressed to maintain the integrity of the application.
Designing an intuitive and user-friendly interface with R Shiny presented its own set of challenges. Balancing aesthetic appeal with functionality while ensuring a seamless user experience required iterative design iterations and feedback gathering. Overcoming design challenges such as layout optimization, interactive element placement, and responsiveness across various devices was crucial to enhancing user engagement and satisfaction
Processing and analyzing large datasets within the R environment posed significant performance challenges. We faced difficulties in optimizing the code to handle the vast amount of data efficiently, particularly when generating real-time recommendations based on user filters. Implementing optimization techniques and leveraging caching mechanisms helped mitigate performance bottlenecks and improve the overall responsiveness of the application.
Accomplishments that we're proud of
The project itself was very intense, but we managed to pull through. The formatting of the UI was the best accomplishment, as we managed to get all the graphs to look very neat and presentable. The code worked for all the sliders and buttons that we implemented. The best thing we were proud of was our prediction model. We developed prediction models to forecast the nutritional content of Starbucks beverages based on user-selected criteria. These prediction models utilized machine learning algorithms, such as decision trees or random forests, to analyze historical data and generate predictions for calorie, protein, sugar, and sodium levels in drinks. By leveraging predictive modeling techniques, BrewMetric could provide users with personalized recommendations tailored to their specific dietary preferences and restrictions.
What's next for BrewMetric
Throughout the development and deployment process, we have remained dedicated to continuously improving BrewMetric based on user feedback and evolving needs. Our commitment to ongoing refinement and updates ensures that BrewMetric remains a valuable resource for Starbucks customers. Maybe in the future present it to other coffee/ beverage companies and implement such apps for their customers!
Log in or sign up for Devpost to join the conversation.