Determining when to interact: The Interaction Algorithm

The Atrium, University of Guelph Institutional Repository

Determining when to interact: The Interaction Algorithm

Show simple item record

dc.contributor.advisor Stacey, Deb
dc.contributor.author Sykes, Edward
dc.date 2012-07-06
dc.date.accessioned 2012-09-07T13:23:54Z
dc.date.available 2012-09-07T13:23:54Z
dc.date.issued 2012-09-07
dc.identifier.uri http://hdl.handle.net/10214/3942
dc.description.abstract Current trends in society and technology make interruption a central human computer interaction problem. Many intelligent computer systems exist, but one that determines when best to interact with a user at appropriate times as s/he performs computer-based tasks does not. In this work, an Interaction Algorithm was designed, developed and evaluated that draws from a user model and real-time observations of the user’s actions as s/he works on computer-based tasks to determine ideal times to interact with the user. This research addresses the complex problem of determining the precise time to interrupt a user and how to best support him/her during and after the interruption task. Many sub-problems have been taken into account such as determining the task difficulty, the intent of the user as s/he is performing the task and how to incorporate personal user characteristics. This research is quite timely as the number of interruptions people experience on a daily basis has grown considerably over the last decade and this growth has not shown any signs of subsiding. Furthermore, with the exponential growth of mobile computing, interruptions are permeating the user experience. Thus, systems must be developed to manage interruptions by reasoning about ideal timings of interactions and determining appropriate notification formats. This research shed light on this problem as described below: 1. The algorithm developed uses a user model in its’ reasoning computations. Most of the research in this area has focused on task-based contextual information when designing systems that reason about interruptions. Researchers support additional work should be done in this area by including subjective preferences. 2. The algorithm’s performance is quite promising at 96% accuracy in several models created. 3. The algorithm was implemented using an advanced machine learning technology—an Adaptive Neural-Fuzzy Inference System—which is a novel contribution. 4. The algorithm developed does not rely on any user involvement. In other systems, users laboriously review video sessions after working with the system and record interruption annotations so that the system can learn. 5. This research shed light on reasoning about ideal interruption points for free-form tasks. Currently, this is an unsolved problem. en_US
dc.language.iso en en_US
dc.subject human computer interaction en_US
dc.subject artificial intelligence en_US
dc.subject context-aware computing en_US
dc.subject interaction algorithm design, development and evaluation en_US
dc.title Determining when to interact: The Interaction Algorithm en_US
dc.type Thesis en_US
dc.degree.programme Computer Science en_US
dc.degree.name Doctor of Philosophy en_US
dc.degree.department Department of Computing and Information Science en_US


Files in this item

Files Size Format View Description
sykes_dissertation.pdf 33.06Mb PDF View/Open Ed Sykes Thesis

This item appears in the following Collection(s)

Show simple item record

Search the Atrium


Advanced Search

Browse

My Account