Game Piece Density Optimization AI : Location Based Music App
Preferred learners
- Canada
- Academic experience
Categories
Project scope
- What is the main goal for this project?
-
Overview
Hipster Bait is a location-based music discovery app that deploys game pieces (songs) on a map interface. Users follow the map to find unreleased rock songs in cities across North America, and soon the globe.
We need a data science team to help us analyze current game data, and develop a ML model for optimizing our the density of game pieces in our app with the goal of maximizing user retention and gameplay for each major city that we deploy songs to.
Background
The Hipster Bait app is free for iOS and Android. Try it for yourself:
Android:
https://play.google.com/store/apps/details?id=com.hipsterbait.hipsterbaitandroid
iPhone:
https://apps.apple.com/app/id1518649365
How it Currently Works
Each location in the Hipster Bait app is entered into the CMS for major cities across North America. Locations are curated (carefully researched) by our staff to ensure they're interesting and match the rock n' roll theme of the app.
Locations consist of a geocoordinate (latitude/longitude), a photo and a text description of the location. Together these elements form a location. The admin saves these locations in the CMS and when published these locations display as pins (bait) on the map inside the mobile app.
In the case of the Hipster Bait mobile app, more than 60,000 unique locations have been created, saved in the CMS and deployed to to the app as shown in the example below. We're constantly adding new locations.
The image above shows the CMS in the web browser as well as the map from the Hipster Bait mobile app. Additional product screenshots of the CMS have been attached.
Density Optimization A.I.
As mentioned above the locations are curated by admin staff and are not optimized for density in our active cities.
We want build an A.I. to provide recommendations for admins to ensure we set the perfect number of game pieces in a specific region or city.
We know from qualitative surveys and user interviews, that if we space the game pieces (songs) too far apart, users won't travel to find them. Conversely, if the songs are too close together, there is not enough scarcity and gameplay suffers.
We don't know the ideal density, nor do we know how the optimal density differs between cities. We don't know the density optimization in order to program intelligence into the CMS to guide the administrators and ensure each city has the perfect number of game pieces to maximize retention and gameplay.
Game mechanics, scarcity psychology and data science will be required for this component of the platform and we do not have this expertise in-house.
- What is the main goal for this project?
-
Overview
Hipster Bait is a location-based music discovery app that deploys game pieces (songs) on a map interface. Users follow the map to find unreleased rock songs in cities across North America, and soon the globe.
We need a data science team to help us analyze current game data, and develop a ML model for optimizing our the density of game pieces in our app with the goal of maximizing user retention and gameplay for each major city that we deploy songs to.
Background
The Hipster Bait app is free for iOS and Android. Try it for yourself:
Android:
https://play.google.com/store/apps/details?id=com.hipsterbait.hipsterbaitandroid
iPhone:
https://apps.apple.com/app/id1518649365
How it Currently Works
Each location in the Hipster Bait app is entered into the CMS for major cities across North America. Locations are curated (carefully researched) by our staff to ensure they're interesting and match the rock n' roll theme of the app.
Locations consist of a geocoordinate (latitude/longitude), a photo and a text description of the location. Together these elements form a location. The admin saves these locations in the CMS and when published these locations display as pins (bait) on the map inside the mobile app.
In the case of the Hipster Bait mobile app, more than 60,000 unique locations have been created, saved in the CMS and deployed to to the app as shown in the example below. We're constantly adding new locations.
The image above shows the CMS in the web browser as well as the map from the Hipster Bait mobile app. Additional product screenshots of the CMS have been attached.
Density Optimization A.I.
As mentioned above the locations are curated by admin staff and are not optimized for density in our active cities.
We want build an A.I. to provide recommendations for admins to ensure we set the perfect number of game pieces in a specific region or city.
We know from qualitative surveys and user interviews, that if we space the game pieces (songs) too far apart, users won't travel to find them. Conversely, if the songs are too close together, there is not enough scarcity and gameplay suffers.
We don't know the ideal density, nor do we know how the optimal density differs between cities. We don't know the density optimization in order to program intelligence into the CMS to guide the administrators and ensure each city has the perfect number of game pieces to maximize retention and gameplay.
Game mechanics, scarcity psychology and data science will be required for this component of the platform and we do not have this expertise in-house.
- What tasks will learners need to complete to achieve the project goal?
-
No deliverables exist for this project.
About the company
- https://hipsterbait.com/
- 2 - 10 employees
- Entertainment
We love rock music. But not watered-down, over-produced radio rock. We mean the good stuff. The authentic, raw and dangerous music that made parents everywhere fear its angst-inducing power.
Rock is up against the internet, globalization, cultural atomization and the evolution of the traditional music industry. The goal of our game is simple: bring back rock n’ roll and have fun doing it.