Conclusions and Results achieved:
- Videogames and recommender systems represent a cross-section of areas in which women are underrepresented. While women make up nearly half of all gamers, they compose less than five percent of videogame programmers and about twenty-five percent of general programmers. By identifying areas in which gender disparity exists and discovering ways to view users as a personality rather than a demographic, this project helped to advance gender equality in the fields of both computer science and videogames. It also addressed the relatively overlooked issue of demographic stereotyping in recommender systems in general.
- Moving to a new home is a difficult process especially getting information about the neighborhood and its safety. With the use of data integrated from multiple domains, a variety of information can be collected about a house, its surroundings, amenities, and the value of the area. Our survey results show that home buyers consider crime rates and home price range as the top factors influencing their home buying decision.
- This project resulted in the design of 2 ontologies: first one combines information about video games from GamesDB.net and user metadata from Steam gaming platform; second one combines real estate data and crime data from a public data provider called Quandl.
- Future work includes creation of a video game recommender system and home buying advisor application using the ontologies and semantic data generated from this project.
- One of the students on this project has decided to pursue higher studies as a result of this project and might continue this work towards a Master’s thesis.
Reports:
| rebeccalittle_creu_report.pdf |
| kristelbasra_creu_report.pdf |
Publications:
- R. Little. "Using Semantic Technology to Address Gender Stereotyping in Videogame Recommender Systems" in General Poster Session at Grace Hopper Celebration, 2015.
|
|