Towards a person-centric interface for information re-finding and sharing tasks

Holmes, Paul (2018) Towards a person-centric interface for information re-finding and sharing tasks. PhD thesis, University of Nottingham.

[img] PDF (Thesis - as examined) - Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (13MB)

Abstract

After the identification of the role that the connections between people and information can play in supporting personal information tasks, some means of exploiting these connections to support information re-finding and sharing were considered. Some past research has examined the use of people in relation to information to perform information tasks, primarily applied to information re-finding and sharing. This small body of work, however, has not explored in depth a basis for how best to design interfaces focused on people, to support users in performing personal information tasks in this manner.

Two further studies were therefore conducted to explore how to design interfaces that support a greater focus on people -- interfaces that are 'person-centric' in nature. The first of these studies provided a basis for how to design interfaces focused on the use of people for personal information tasks, and the second evaluated a 'person-centric' design based on the findings of the prior study.

Designing interfaces that provide a means of accessing or sharing information through interaction with personal contacts requires a means of organising and representing those contacts. Just as the diary study revealed some of the prominent dimensions that information is recalled and ordered by, the design of 'person-centric' interfaces requires a similar understanding of what dimensions influence how to structure and order contacts. In order to design an interface based on supporting information tasks using people as a central component of the interface, how people think about relationships between their contacts and how people would choose to represent their contacts was examined.

This was approached in two ways. First, through use of a modified repertory grid method, whereby participants made comparisons between many combinations of their own personal contacts to elicit constructs which described the different forms of relationships they perceived between their contacts. Second, through participants making free-hand drawings that illustrated how they would choose to represent all of their contacts. From analysing this data a categorisation of the different forms of contact groups people perceived was generated. Applying this to the data revealed which forms of expressing relationships were most prevalent, which at the highest level were those related to organisations, locations, relationship types and events. Different forms of representations of contacts, which each structured the contacts in different ways, were also identified from the data. These representations each emphasised different aspects such as overlaps between groups of contacts, the relative importance of contacts, the location of contacts, and the links between contacts.

A prototype 'person-centric' system was then developed to evaluate a range of 'person-centric' design concepts for supporting people in performing information sharing and re-finding tasks. This incorporated design ideas based on examination of the re-finding behaviours from the diary study and the examination of contact relationships and representations. The prototype system extracted contacts with whom the user had reciprocated contact from their email and Facebook accounts, as well as related messages, files and links. The user was then able to use the system to create groups of contacts, which could then later be used to aid performing information sharing and re-finding tasks through a series of different 'person-centric' interface presentations. The structure of these presentations could be changed, re-ordered, and filtered based on the results of the repertory grid and free-hand diagramming, which provided a basis for understanding different ways people may wish to order or filter contacts, and the different ways to structure contacts in an interface.

The prototype was evaluated in two parts. In the first part the process of contact grouping was studied, and in the second part the design ideas and the use of people as a primary unit of interaction was evaluated. In both parts of the evaluation participants performed think-aloud protocol while interacting with the interfaces to perform the tasks, which was recorded using audio and video capture.

In the evaluation of the contact grouping two grouping interface designs were used. The participant's task with the interfaces was to group all of their contacts in each interface condition. The aim of this was to explore what constituted a meaningful contact group and to understand what role interface design played in this. In the re-finding and information sharing interface evaluation participants were questioned about their personal data in order to generate tasks to re-find information, or items to share from their personal data using the prototype system. Using the identified information items, tasks were set using four different representations that supported information sharing and re-finding. These tasks aimed to evaluate the utility of the different 'person-centric' design concepts that had been identified. Following both parts of the evaluation interviews were conducted to understand the experiences and preferences of the participants in using the different interfaces.

From the two part evaluation, the decision-making processes of constructing groups and the factors that influence that process were outlined, as well as preferences and behaviour regarding different features of the interfaces that supported performing the information tasks. Evaluation of design ideas from the previous two studies through the prototype confirmed the value of utilising connections made between people and information when re-finding information, and allowed factors that have implications for 'person-centric' interface design to be identified in relation to how contacts and groups should be represented.

Through its examination of the use of connections that are made between people and information when performing personal information tasks and the representation of people in interface design, this thesis provides a greater degree of understanding of typical information re-finding behaviours, the representation and relationships of people in interfaces for performing information tasks, the process of constructing contact groups and what constitutes a meaningful contact grouping. It contributes guidelines that inform how to design 'person-centric' interfaces in relation to the structure of representations, ordering and filtering of contacts, and the linking of information to contacts. An improved understanding of the processes involved in contact group creation and factors that influence it with implications for supporting for manual, semi-automated, and automated group creation and identification. As well as a more complete picture of what information re-finding entails through analytical frameworks that describe the how, what, and why of re-finding tasks

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Sharples, Sarah C.
Goulding, James
Keywords: Interfaces; information re-finding behaviours
Subjects: T Technology > TA Engineering (General). Civil engineering (General)
Faculties/Schools: UK Campuses > Faculty of Engineering
Item ID: 55490
Depositing User: Holmes, Paul
Date Deposited: 23 Jan 2019 09:54
Last Modified: 07 Feb 2019 18:31
URI: http://eprints.nottingham.ac.uk/id/eprint/55490

Actions (Archive Staff Only)

Edit View Edit View