Abstract by Jacob Carter
Automated Video Game Categorization
With such a large quantity of video games on the market it can be challenging for individuals to decide where to invest their time and money to have an enjoyable video gaming experience. In order to minimize the time, efforts, and expenses of gamers invested in games they will enjoy, we propose to develop a recommendation algorithm that can automatically classify video games into the various gaming genres and recommend games to the gamers based on their gaming genre preferences. We experimented using a Support Vector Machine (SVM) approach, the LIBSVM implementation to be exact, in order to automate the game classification process. SVMs generally categorize effectively thanks to their ability to map the data to higher planes. Our results showed an average 85% prediction accuracy. While more research is needed, the SVM appears to be a good choice for automated video game categorization.