Prediction Games
More datasets are becoming available on the web. This provides new opportunities for data-driven systems that can entertain and inform. We introduce prediction games, data-driven games modeled after fantasy sports. We hypothesize that prediction games can motivate people to explore and analyze online datasets in order to develop their own understanding of the data’s domain and to improve their data analysis skills. The mechanics of prediction games revolve around activities where players analyze historical data and information resources to make predictions about future events.
This project investigates the practices of players of existing prediction games, the development of a domain-independent prediction game engine, and the design of prediction games in new domains. We are particularly interested in the application of such games to data science education.
This project began by understanding the history of and developing a model of fantasy sports. This involves understanding the practices of fantasy sports players. Based on this analysis we have developed initial data-driven prediction games to motivate data analysis skills and rich domain knowledge.
Publications and Further Readings
Effect of Visualization of News Articles in Data Driven Games. Thesis. 2016
Design of a Prediction Game in the Domain of Computer Security. Capstone. (2020)
Prediction Games: Encouraging Engagement with Data. Dissertation. (2020)
Prediction Games & Demos
As of now, we have developed two games based on our prediction games engine in the climate domain using weather data: Fantasy Climate and Fantasy Precipitation.
Fantasy Climate (Demo coming soon) encourages players to look for and understand long-term change in weather data. Players are asked to make two selections from a given set of U.S. cities each scoring period: one for the location that will be warmest relative to its historic norm and the second for the location coolest relative to its historic norm. These selections are then scored on how much the observed temperature for the prediction date deviates from the historic norm.
Fantasy Precipitation (Demo coming soon) encourages the players to observe long term frequency in precipitation data, and employ such observations to do well during the game. Players are asked to predict at a specific date whether precipitation (snow or rain) will occur or not within a date range (e.g. 7 days) in every city of a given set. Scoring is based on the number of correct predictions.
Technical Documentation
Here is the technical documentation underlying the prediction games system. Although a little behind, it gets updated after every major changes to the system.