Intelligent tutoring systems provide personalized feedback to students and improve learning at effect sizes approaching that of human tutors. However, the design and development of these systems is expensive and requires the collaboration of experts across multiple fields. An additional benefit of these educational systems is that they allow collection of detailed records of student actions. However, the complex nature of this interaction rich data makes it difficult to analyze. In previous work, researchers were able to use a past corpus of student data to generate one of the basic features of intelligent tutoring systems, a personalized hint about the next step in a problem.We have used a variety of methods to discover the effectiveness of these automatically generated hints in terms of tutor performance, student motivation, and student problem solving strategies.
We have created a complex network model called an interaction network that represents an empirical sample of student problem solving behavior.We have mined this structure to derive high level student approaches to solving problems. Using these ApproachMaps we are able to quantitatively evaluate between-group differences in how student approach problems. Interaction networks are designed handle environments where problems have more than one correct answer and multiple solution paths. To further our understanding of student problem-solving within these open-ended well-structured problems we create interaction networks from several different tutor domains and explore the common properties between them.We find evidence that Interaction Networks are scale-free, indicating that even when the problem state-space seems infinite, students visit a relatively small subset of the total state-space. Finally, we explore methods to estimate the size of the unobserved parts of the Interaction Networks.We use these estimates in order to understand how much data is needed before we can use the networks as samples of student problem-solving.
|School:||North Carolina State University|
|School Location:||United States -- North Carolina|
|Source:||DAI-B 77/10(E), Dissertation Abstracts International|
|Subjects:||Educational psychology, Cognitive psychology, Computer science|
|Keywords:||Complex network, Data driven, Data science, Educational data mining, Intelligent tutoring systems, Visualization|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be