能力相若配對：提昇相得遊戲產出品質

Despite its awesome computation power, so far computer cannot satisfyingly solve many tasks that are extremely easy for human, such as image recognition or common sense reasoning. A partial solution is to delegate algorithmically difficult computa-tion task to human, called human computation. The Game with a Purpose (GWAP), in which computational task is transformed into a game, is perhaps the most popular form of human computation. An adverse selection problem for output-agreement/simultaneous- verification GWAP was built, using the ESP Game as example. The experiment results favored an adverse selection model over an moral hazard model. I was particularly interested in output quality of a GWAP affected by how players are matched with each other, and proposed capability-aligned matching (CAM) versus commonly-used ran- dom matching. The analysis showed that when compared with random matching, the CAM improved output quality. The experiment confirmed conclusions drew from the analysis, and further pointed out that task-human matching scheme was as important as human-human matching scheme studied in this thesis. The main contribution of this thesis is the analysis and empirical evaluation of human-human matching scheme, showing that capability-aligned matching can improve quality of GWAP.

並列關鍵字

ESP Game ； Game with a Purpose ； Human Computation ； Adverse Selection ； Mechanism Design

參考文獻

[5] C.-J. Ho, T.-H. Chang, J.-C. Lee, J. Y.-J. Hsu, and K.-T. Chen. Kisskissban: A competi-

tive human computation game for image annotation. In Proceedings of the First Human

[6] C.-J. Ho and K.-T. Chen. On formal models for social verification. In Proceedings of the

First Human Computation Workshop (HCOMP 2009), June 2009.

[7] S. Jain and D. C. Parkes. A game theoretic analysis of games with a purpose. In In Proc.

國際替代計量

能力相若配對：提昇相得遊戲產出品質

全文下載

主題瀏覽