Zhang Wanyi, College of Computer Science and Technology, Jilin University
Nowadays computer and other electronic equipment have become highly intelligent with the development of computer science and technology. However, computer is still not smart enough to identify images automatically, and this problem has plagued the scientists for a long time. In recent years, scientists work on the Deep Learning and Neural Network Algorithm, and it helps to solve the automatically image annotation problem. However, since the operation of human brain is much more complicated than any machine learning models, the annotation results can not satisfy us sometimes, and may need the subsequent manual validation. In our work, we propose a crowdsourcing system to collect the manual annotation produced by users. We treat every single user as an independent image annotation processor, and crowdsourcing is equivalent to a Distributed Image Annotation System which helps us to accomplish the annotation of images efficiently and quickly. And most of all, the users’ language diversity leads to the multilingual annotation of the images, even the Deep Learning algorithm cannot get the multilingual results at one time. In order to improve the user engagement and behavior, we design a game system to help us to attract users’ interest and add some elements of gamification such as PBL and the mechanism of competition and cooperation. In our presentation we will introduce the detail of our game system design and discuss the role played by gamification from the perspective of computer science.
Author info
Zhang Wanyi is a master candidate in College of Computer Science and Technology of Jilin University. Zhang Wanyi received her B.Sc degree in computer science in Jilin University. Her current research interests include Data Mining in Open Science, and Gamification in Education. Zhang Wanyi have published a workshop paper in SIGCHI 2015.