1.3
Research Questions
7
Computer games, including MMOGs, have been considered as the killer applica-
tions of human-level AI, Laird and Kent [2001]. We share this view, but stress
that human-level AI in MMOGs is neither the goal nor the result of this thesis.
1.3
Research Questions
The corresponding main research question that this thesis attempts to answer is:
MRQ: How can we provide scalable and flexible user representation and predic-
tion methods for increased automatic customization of cyberspace services?
This question is important because if answered it can lead to what users con-
sider as better cyberspace services. And if users become more satisfied with the
cyberspace services they are likely to use them more hence providing increased
revenue for the cyberspace service providers. The reason the services become
better is because the user representation and prediction methods can provide
more targeted, interesting and relevant information for each individual user and
the performance can potentially be improved by the prediction methods
The set of more specific research questions are as follows:
RQ1: How can mobile commerce customers be supported by software agents?
1. How to handle resource constraints imposed by mobile devices?
2. How to represent the customer's profile and interests?
3. Which types of customer services can be handled?
4. Can the approach be generalized to handle electronic commerce?
5. How can customers get product or service recommendations?
RQ2: How to test the proposed solution to RQ1, and extending it toward sup-
porting Massively Multiplayer Online Games?
1. Why are Massively Multiplayer Online Games (MMOGs) a relevant
test case?
2. How can data for customer personalization be gathered in the MMOG
case?
3. How do customer data gathering in MMOGs compare to the existing
standard for Web Usage Logging?
4. How can customer data from MMOG services be used for customer
personalization?