Paper reading #5:Tell Me More? The Effects of Mental Model Soundness on Personalizing an Intelligent Agent

Intro:
Title:Tell Me More?The Effects of Mental Model Soundnesson Personalizing an Intelligent Agent
Reference Information: In London
Author Bios:Todd Kulesza1, Simone Stumpf2, Margaret Burnett1, Irwin Kwan1
Students and Staff at Oregon State University and City University London. They interest in artificial intelligence.

Summary:
People are using intelligent agents more frequently, such as Facebook, amazon, and Pandora. In Facebook, they use facial detection to guess who is in a particular photo. But this kind of intelligent agents need to learn from the users. And this paper is wonder how to improve the intelligent agents by users. The result is people who knows how to debug the intelligent agents will have better agents. This is due to their knowledge how to fully operate the agents.

Related Works:

1. Intelligent agent supported personalization for virtual learning environments
2. Intelligent agent-based systems for personalized recommendations in Internet commerce
3. User models to personalize an intelligent agent
4. Intelligent techniques for web personalization
5. User–interface agent interaction: personalization issues
6. Personalizing information appliances: a multi-agent framework for TV programme recommendations
7. “ARCHIMIDES”: an intelligent agent for adaptive-personalized navigation within a WEB server
8. Programming with an intelligent agent
9. Learning for automatic personalization in a semantic taxonomy-based meta-search agent

10. Agent technology for personalized information filtering: the PIA-system

I think this topic is about novel, due to existing and recently trends of usage of intelligent agents. The work I found has many research on this topic, however, I still think the work I read about is novel in this field. Many research has the familiar result as the paper I read of.

Evaluation:
There are 62 participants in this study and separated into two groups. One group with only necessary information with the AuPair (Radio station that researchers developed to research on) and one group with deep explanation how the radio station will operate depends on their actions. The researchers use quantitative and unbiased measures. The method is systemic.

Discussion:
I think the contribution of this paper is medium that is due to there are many familiar researches out there. This paper did not mention how it will really applied their research into the real world.