Optimization Approaches to Adaptive Menus

Loading...
Thumbnail Image

URL

Journal Title

Journal ISSN

Volume Title

Sähkötekniikan korkeakoulu | Master's thesis

Date

2017-05-22

Department

Major/Subject

Communications Engineering

Mcode

ELEC3029

Degree programme

CCIS - Master’s Programme in Computer, Communication and Information Sciences (TS2013)

Language

en

Pages

82 + 7

Series

Abstract

Graphical menus perform as vital components and offer essential controls in today’s graphical interface. However, few studies have been conducted to modelling the performance of a menu. Furthermore, menu optimization methods previously proposed have been largely concentrating on reshaping layout of the whole menu system. In order to model menu performance, this thesis extends the Search-Decision-Pointing model by introducing two additional factors, i.e. the cost function and semantic function. The cost function is a penalty function which decreases the user expertise regarding a menu layout according to the degree of modification done to the menu. The semantic function is a reward function which encourages items with strong relations be positioned close to each other. Centered on this menu performance model, several optimization methods have been implemented. Each method focuses on improving menu performance by applying distinctive strategies, such as increasing item size or reducing item pointing distance. Three test cases have been exercised to evaluate the optimization methods in a simulated software which displays graphical user interfaces and emulates the menu utilization of real users. The results of test cases reveal that the menu performance has been successfully improved in all test cases by the fundamental heuristic search algorithm. Moreover, other optimization methods have been able to further increase menu performance ranging from 3% to 8% depending on test cases. In addition, it is identified that increasing the size of an item offers surprisingly little benefit. Conversely, reducing item pointing distance has greatly improved menu performance. Moreover, positioning items by their semantic relations may also enhance group saliency. On the other hand, optimization methods may not always succeed in providing usable menus due to design constraints. Hence, menu performance optimization shall be carefully exercised by considering the entire graphical user interface.

Description

Supervisor

Oulasvirta, Antti

Thesis advisor

Oulasvirta, Antti

Keywords

user interface, graphical menu, model-based design and optimization, heuristic algorithm, multi-objective function

Other note

Citation