The popularity of prediction games such as fantasy sports has been on the rise and the amount of data available for the players to make the prediction in such games is growing rapidly. Prediction games, an area of data driven games require the user to interpret archival data along with real-time data from a domain to make a prediction about a future event. This work is being done in the context of a prediction game where players select geographic locations as part the game. News articles can serve as a source of information and have the potential to improve engagement and learning. To make sense of the millions of news artifacts published online is not possible manually. Moreover, keyword-based search is very limited when it comes to exploring data rather than just searching for something particular. The proposed work will develop a visualization and user interface to represent the news articles in an activity-appropriate manner and allow the users to explore news data using approximate search along with keyword-based search. The first component of this system extracts news articles related to the game, then geotags and clusters them based on the geographic references in the articles. Displaying these clusters on a map takes advantage of the spatially referenced news data. This thesis will compare alternative visualizations of the geo-tagged and clustered news articles and their value to players of the game. A user study was conducted to evaluate the visualizations and their effect on the engagement of the players with a data driven game. The results show that the map visualization is very effective in engaging players with the game when compared to the regular list form of news representation. Moreover, the overall performance of the players who used the map visualization was better than the performance of the users who used the list visualization. Future work will explore more on fine-tuning data sources which provide the input to the map visualization as well as variations in the display and accessible features on the map interface to enable users to control data visualization according to their imagination and preference.
The popularity of prediction games such as fantasy sports has been on the rise and the amount of data available for the players to make the prediction in such games is growing rapidly. Prediction games, an area of data driven games require the user to interpret archival data along with real-time data from a domain to make a prediction about a future event. This work is being done in the context of a prediction game where players select geographic locations as part the game. News articles can serve as a source of information and have the potential to improve engagement and learning. To make sense of the millions of news artifacts published online is not possible manually. Moreover, keyword-based search is very limited when it comes to exploring data rather than just searching for something particular. The proposed work will develop a visualization and user interface to represent the news articles in an activity-appropriate manner and allow the users to explore news data using approximate search along with keyword-based search. The first component of this system extracts news articles related to the game, then geotags and clusters them based on the geographic references in the articles. Displaying these clusters on a map takes advantage of the spatially referenced news data. This thesis will compare alternative visualizations of the geo-tagged and clustered news articles and their value to players of the game.
A user study was conducted to evaluate the visualizations and their effect on the engagement of the players with a data driven game. The results show that the map visualization is very effective in engaging players with the game when compared to the regular list form of news representation. Moreover, the overall performance of the players who used the map visualization was better than the performance of the users who used the list visualization. Future work will explore more on fine-tuning data sources which provide the input to the map visualization as well as variations in the display and accessible features on the map interface to enable users to control data visualization according to their imagination and preference.
