MODELING THE ADAPTION RULE IN CONTEXTAWARE SYSTEMS

International Journal of Ad hoc, Sensor & Ubiquitous Computing (IJASUC) Vol.7, No.3/4, August 2016
DOI : 10.5121/ijasuc.2016.7401 1
MODELING THE ADAPTION RULE IN CONTEXT-
AWARE SYSTEMS
Mao Zheng1
, Qian Xu2
and Hao Fan3
1
Department of Computer Science, University of Wisconsin-LaCrosse, LaCrosse, USA,
2
Amazon, Seattle,USA and
3
School of Information Management, Wuhan University,Wuhan,China
ABSTRACT
Context awareness is increasingly gaining applicability in interactive ubiquitous mobile computing
systems. Each context-aware application has its own set of behaviors to react to context modifications. This
paper is concerned with the context modeling and the development methodology for context-aware systems.
We proposed a rule-based approach and use the adaption tree to model the adaption rule of context-aware
systems. We illustrate this idea in an arithmetic game application.
KEYWORDS
Ubiquitous Mobile Computing; Context; Context-awareness; Rule-based Approach; Adaption Tree
1.INTRODUCTION
Our world gets more connected everyday. These connections are driven in part by the changing
market of smart phones and tablets. Pervasive computing environments are fast becoming a
reality. The term “pervasive”, introduced first by Weiser [1], refers to the seamless integration of
devices into the user’s everyday life. One field in the wide range of pervasive computing is the
so-called context-aware system. Context-aware systems are able to adapt their operations to the
current context without an explicit user intervention and thus aim at increasing usability and
effectiveness by taking environmental context into account. Each context-aware application has
its own set of behaviors to react to context modifications. Hence, every software engineer needs
to clearly understand the goal of the development and to categorize the context in the application.
We incorporate context-based modifications into the appearance or the behavior of the interface,
either at the design time or at the run time. In this paper, we present application behavior adaption
to the context modification via a context-based user interface in a mobile application, arithmetic
game. The application’s mobile user interface (MUI) will be automatically adapted based on the
context information.
The user interface (UI) can include many features such as font color, sound level, data entry, etc.
Every feature has some variables. For example, data entry can be done using typing, voice and
tapping. From the designer’s perspective, the adaptability of these features is planned either at the
design time or during the runtime. Through the literature study, we proposed a rule-based
approach model, and used an adaption tree to present this model. The adaption tree is what we
named in our methodology. It is based on the extension of a decision table, the decision tree. We
use the adaption tree to represent the adaption of the mobile device user interface to various
context information. The context includes the user’s domain information and dynamic
environmental changes. Each path in the adaption tree, from the root to the leaf, presents an
adaption rule. To illustrate our methodology, we implemented a context-aware application in the
Android platform, the arithmetic game application.

2
There are two major platforms in the mobile device community: iOS and Android. This project
chose Android development mainly for the reason of its openness. In addition, all the tools in the
Android development are free and no special hardware is required.
The rest of the paper is organized as follows: in Section 2 we compare how our views are similar
to other researchers and how they are different. Section 3, we briefly describe the arithmetic game
application. Section 4 presents the rule-based approach and the fundamental concepts of the
adaption tree. Section 5 discusses the development of the arithmetic game based on the adaption
tree. Section 6 concludes the paper and outlines the directions of our on-going research.
2. RELATED WORK
Some researchers define context as the user’s physical, social, emotional or informational state, or
as the subset of physical and conceptual states of interest to a particular entity [2]. The authors in
[2] have presented the definition or interpretation of the term by various researchers, including
Schilit and Theimer [3], Brown et al. [4], Ryan et al. [5], Dey [6], Franklin & Flaschbart [7],
Ward et al. [8], Roddenet al. [9], Hull et al. [10], and Pascoe [11]. In Dey and Abowd [2], the
authors are interested in context-aware systems, and so they focused on characterizing the term
itself. In Pascoe [11], the author’s interest is wearable computers, so his view of context is based
on environmental parameters as perceived by the senses. Our work depends on the internal
sensors of a mobile device, and the adaption of the mobile user interface features for both
entering and accessing data. Our model is based on separating how context is acquired from how
it is used, by adapting the mobile user interface features to the user’s context.
Most of the research in this area has been based on analyzing context-aware computing that uses
sensing and situational information to automate services, such as location, time, identity and
action. More detailed adaption has been generally ignored. For example, input data based on
context. In our research, we attempted to build the user’s characteristics from both domain
experience and mobile technology experience, and to collect all the context values corresponding
to the user’s task and then to automatically adapt the mobile user interfaces to the context
information.
The process of developing context-based user interface has been explored in a number of other
projects. Clerckset al. [12], for example, discuss various tools to support the model-based
approach. Many studies have been conducted on adaption using a decision table. In [13], an
approach is proposed for modeling adaptive 3D navigation in a virtual environment. In order to
adapt to different types of users, they designed a system of four templates corresponding to four
different types of users. Our work differs in that our adaption technique is based on composite
context information that extracts values from sensors in smartphones and relates with the user’s
domain and mobile technology experiences. Then we develop a set of rules for the mobile user
interface adaption. We used the adaption tree to model the context information and represent
adaption rules. It also serves as the model for the design and implementation.
3.THE ARITHMETIC GAME APPLICATION
The arithmetic application is developed for users of different ages, with different arithmetic skills.
The mobile user interface will adapt to the user’s profile, actual performance and current time and
local weather information. The device’s orientation will also be discussed as one of the context
information.
Users are required to register and obtain an account to login. During the user’s registration
process, the user’s age is stored as one of the logic context information that will be used in the
beginning of the app to assign the appropriate question level to the user.

3
There are two modes in the application: standard mode and review mode. The standard mode is to
let the user practice or test their arithmetic skills through questions in different levels and units.
The review mode is to let the user redo the questions he/she made mistakes on before.
In the standard mode, there are three levels corresponding to three age domains. Each level is
divided into 10 units and each unit contains 10 problems. When a user answered all questions
correctly in the last unit, or answered more than or equal to 90 questions correctly in the last
level, the user can choose to level up, otherwise the user will stay at the same level.
When the user logs in for the first time, he/she is automatically assigned to a level based on the
age information. All the problems are generated randomly based on the rules shown in the Table
1 below. Generating questions randomly instead of retrieving questions from a database, can
avoid the users remembering the order of answers when they play the game again and again.
Table 1: Question Characteristics of Different Levels
Age Level Operands for + Operands for - Operands for * Operands for /
[0, 5] 1 Both [0, 10] Both [0, 10]
minuend> subtrahend
Not available Not available
[6, 12] 2 Both [0, 50] Both [0, 50]
minuend> subtrahend
Both[0, 10] Both [0, 10]
Quotient is integer
>= 13 3 Both [-100, 100] Both [-100, 100] Both[-10, 10] Both [-100, 100]
Quotient is integer
When the users answer questions, there are 10 seconds for each question. A graphic countdown
timer should work as a reminder.
The accuracy of the last unit is divided into three groups: [0%, 60%], (60%, 90%), [90%, 100%].
The arithmetic game presents three themes for three different accuracy groups respectively. For
the first group, whose last unit accuracy is between 0% and 60% inclusively, the application
simply presents a default theme. For the second group, whose last unit accuracy is between 60%
and 90% exclusively, the user is able to design their own theme with their preferred color. In the
application’s setting, the user can choose preferred colors for different UI widgets. For the third
group, whose last unit accuracy is between 90% and 100% inclusively, the user can design their
own theme with their preferred color, local time and local weather. The weather icon follows the
local weather information. The background image follows the local time period by default.
During 6:00 – 17:00, the picture of a daytime scene is displayed as the background image; during
17:01 – 19:00, the picture of a sunset is displayed as the background image; during 19:01 – 5:59,
the picture of a nighttime scene is displayed as the background image. The user can also choose
to close the “time-based background image” option in the setting, thus the background will be
presented in color style.
4.RULE-BASED APPROACH
Our work depends on the internal sensors of a mobile device, the user profile and the user’s task.
The key point of the approach is to capture and represent the knowledge required for the mobile
user interface to automatically adapt to dynamics at run time, or to implement the adaptions at
design time. The rule-based approach representation is what we are proposing. Figure 1 below
shows our proposed approach.

International Journal of Ad hoc, Sensor & Ubiquitous Computing (IJASUC) Vol.
We use the adaption tree in this paper
adaptation. The inspiration comes from mHealth [14
represent the adaption rules of Mobile User Interfaces. However, in the process of adapting the
decision table to our applications, we met some challenges; 1) the sequence is not clearly shown
in the decision table, 2) the decision table method is useful for those applications that include
several independent relationships among the input parameters, but it does not consider the
relationships among the conditions, such as overlapping or redundancy, 3) a decision table does
not scale up very well – when there are
be evaluated as true, false, or not applicable. It i
conditions when n increases. Because of those limitations within the decision table
our UI adaption rule using an adaption tree instead of a decision table.
The concept of adaption tree comes f
decision-making situation. Compared to a tabular decision table, it takes up more room, but it
shows the order of evaluating the conditions.
A decision tree is a flowchart-like structure in whi
attribute (e.g. whether a coin flip comes up heads or tails), each branch represents the outcome of
the test and each leaf node represents a class label (decision taken after computing all attributes).
The paths from root to leaf represent classification rules [15
4.1.Adaption Tree Used in Our Research
We call the decision tree used in our project an “adaption tree,” and our approach of adaption is
to change the UI based on context. Below are some basic co
UI feature: UI features are the smallest atomic unit for describing UI content on a mobile device.
Table 2 shows some UI features and their actions (also known as values) below:
International Journal of Ad hoc, Sensor & Ubiquitous Computing (IJASUC) Vol.7, No.3/4
Figure1.Rule-based Approach
in this paper to show our rule-based approach in the context
nspiration comes from mHealth [14] where the decision table was selected to
ions, we met some challenges; 1) the sequence is not clearly shown
when there are n rules, there are 2^n
rules. Since each condition needs to
be evaluated as true, false, or not applicable. It is not easy to make a full description for those
increases. Because of those limitations within the decision table
our UI adaption rule using an adaption tree instead of a decision table.
The concept of adaption tree comes from the decision tree. It is a graphical representation of a
making situation. Compared to a tabular decision table, it takes up more room, but it
shows the order of evaluating the conditions.
like structure in which each internal node represents a "test" on an
epresent classification rules [15].
Adaption Tree Used in Our Research
to change the UI based on context. Below are some basic concepts used in our research.
UI features are the smallest atomic unit for describing UI content on a mobile device.
Table 2 shows some UI features and their actions (also known as values) below:
/4, August 2016
4
sed approach in the context-based UI
] where the decision table was selected to
ions, we met some challenges; 1) the sequence is not clearly shown
rules. Since each condition needs to
make a full description for those
increases. Because of those limitations within the decision table, we presented
rom the decision tree. It is a graphical representation of a
making situation. Compared to a tabular decision table, it takes up more room, but it
ch each internal node represents a "test" on an
ncepts used in our research.
UI features are the smallest atomic unit for describing UI content on a mobile device.

5
Table 2:UI features and their actions
UI features Action(values)
Font size Small, medium, large
Font color RGB color, black, white
Background color Auto adjust, change manually
Data entry Typing, voice, tapping...
Display information Text, sound
Message delivery Text, voice, alert…
Brightness level Increase/decrease
Ring volume Low, medium, high
Sound Mute, regular, loud
Video On, off
.... ....
UI feature set: A UI feature set is a non-empty set of UI features.
Disjoint UI feature sets: Two UI feature sets are said to be disjoint if they have no UI feature in
common. It also means that the UI features do not interfere with each other in the process of
adaption. For example: {video}, {media sound} and {brightness level} are three disjoint UI
feature sets, but {video, media sound} and {video, brightness level} are not disjoint UI feature
sets because their intersection is the set {video}.
Action set: Anaction set is a set of actions (also known as values) applied to UI. For example:
{Font size is small, Font color is black} is an action set.
Context category:We categorized the context information into two categories as shown in Table
3. Physical context information is collected by the mobile device’s sensors. The logical
information is gathered through the user’s registration process, the user’s performance, and the
user’s selections in the setting menu.
Table 3: Context Information Categorization
Physical
Context
local time, local weather (local here also implies the context location
considered), device orientation
Logical
Context
user’s profile (age, first time using the app or not, performance)
user’s preference (color preference, image preference)
Context condition:
A context condition is the predicate of the context value. For example: “whether the battery level
is low” is a context condition.
Context set:
A context set is a non-empty set of context. For example: {local time, local weather, device
orientation} is a context set.
Adaption function:
Let F be a UI feature set, let C be a context set and let A (C, F) be a function defined over inputs
F and C, the output is action set, that is C applied to F.

Adaption function distributive rule:
In A (C, F), let F be divided to disjoint UI feature sets: F = f1
distributive rule:
A (C, F) = A (C, f1 ∪f2 ∪…
Below is an example of different ways of distributing an adaption function:
Problem: make adaptions on UI features: video, media sound,
battery is low.
Solution:
1. UI feature set F = {video, media sound, brightness level}
2. Context set C = {battery is low},
3. Function: A (C, F) = ({battery is low}, {video, media sound, brightness level})
4. According to our distributive rule:
a. If F is divided to disjoint UI feature sets: {video}, {media sound, brightness level}
Then A ({battery is low}, {video, media sound, brightness level}) = A({battery is
low}, {video}) ∪A({battery is low}, {media sound, brightness level})
b. If F is divided to disjoint UI feature sets: {video, media sound}, {brightness level}
low}, {video, media sound, brightness level})= A(({battery is low}, {video, media
sound}) ∪A({battery is low}, {brightness level})
c. If F is divided to disjoint UI feature sets: {video}, {media sound, brightness level}
Then : A ({battery is low}, {video, media sound, brightness level}) = A({battery is
low}, { video, media sound, brightness level}) = A(({battery is low}, { video})
A({battery is low}, {media sound })
Adaption tree:
We define decision tree over function A (C, F), and we named the decision tree used in our
approach as an adaption tree. An adaption tree consists of two types of nodes: condition node and
conclusion node. Table 4 shows the nodes and their descript
Name Shape Description
Condition
Node
non
conditions.
Conclusion
Node
leaf node, denoted as a rectangle. It represents a UI
context
Drawn from top to down, each path from root node to leaf node represents an adaption rule. Each
condition node (also recognized as non
context it checked, and it has several branches coming out of it
recognized as leaf node) is denoted as rectangle, and labeled with UI action.
Adaption function distributive rule:
In A (C, F), let F be divided to disjoint UI feature sets: F = f1 ∪ f2 ∪... ∪ fn, then we have
…∪fn,) ≡A (C, f1) ∪A (C, f2) ∪....∪A (C, fn)
Below is an example of different ways of distributing an adaption function:
: make adaptions on UI features: video, media sound, brightness level, based on context:
UI feature set F = {video, media sound, brightness level}
Context set C = {battery is low},
Function: A (C, F) = ({battery is low}, {video, media sound, brightness level})
istributive rule:
If F is divided to disjoint UI feature sets: {video}, {media sound, brightness level}
A({battery is low}, {media sound, brightness level})
If F is divided to disjoint UI feature sets: {video, media sound}, {brightness level}
A({battery is low}, {brightness level})
low}, { video, media sound, brightness level}) = A(({battery is low}, { video})
A({battery is low}, {media sound }) ∪A(({battery is low}, {brightness level })
conclusion node. Table 4 shows the nodes and their descriptions in an adaption tree.
Table 4: Node in the Adaption Tree
Description
non-leaf node, denoted as a diamond. It checks the context
conditions.
leaf node, denoted as a rectangle. It represents a UI action after
context-based adaption.
condition node (also recognized as non-leaf node) is represented as a diamond, labeled with the
context it checked, and it has several branches coming out of it. Each conclusion node (also
recognized as leaf node) is denoted as rectangle, and labeled with UI action.
/4, August 2016
6
fn, then we have
brightness level, based on context:
Function: A (C, F) = ({battery is low}, {video, media sound, brightness level})
If F is divided to disjoint UI feature sets: {video, media sound}, {brightness level}
low}, { video, media sound, brightness level}) = A(({battery is low}, { video}) ∪
A(({battery is low}, {brightness level })
ions in an adaption tree.
leaf node, denoted as a diamond. It checks the context
action after
leaf node) is represented as a diamond, labeled with the
. Each conclusion node (also

7
5.THE ADAPTION TREE IN THE ARITHMETIC GAME APPLICATION
We are presenting the adaption rules by constructing an adaption tree in the arithmetic game
application. There are various contexts that we can gather. We only collected contexts that could
result in at least one action over our application UI.
1. UI feature set: F= {font color, background color, button font color, button background
color, weather icon, background image, orientation mode}
If we divided F to disjoint UI feature sets: F = F1∪F2 = {font color, background color,
button font color, button background color, weather icon, background image}∪
{orientation mode} and name F1 as Theme which is a UI feature set, then we have F
= Theme ∪{orientation mode}
2. Context set: C = {first time, last unit accuracy, device orientation, local time, local
weather} ∪ User’s Preference, the User’s Preference = {font color preference,
background color preference, button font color preference, button background color
preference, background image preference}
3. Function: A (C, F) = A ({first time, last unit accuracy, device orientation, local time,
local weather, User’s Preference}, Theme ∪{orientation mode})
4. According to our distributive rule:
A ({first time, last unit accuracy, device orientation, local time, local weather,
User’s Preference}, Theme∪{orientation mode })
≡A ({first time, last unit accuracy, device orientation, local time, local weather,
User’s Preference}, Theme)
∪A ({first time, last unit accuracy, device orientation, local time, local weather,
User’s Preference }, {orientation mode })
Then define the output:
1. A ({first time, last unit accuracy, device orientation, local time, local weather, User’s
Preference}, Theme) has three discrete output values: Default Theme, Preferred Color
Theme, Weather & time based Theme.
2.
a. Default Theme = {font color is black, background color is white, button font
color is black, button background color is white, weather icon is null, background
image is null}
b. Preferred Color Theme = {font color is A (User’s Preference, {font color}),
background color is A (User’s Preference, {background color}), button font color
is A (User’s Preference, {button font color}), button background color is A
(User’s Preference, {button background color}), weather icon is null, background
image is null}
c. Weather & time based Theme = {font color is A (User’s Preference, {font
color}), background color is A (User’s Preference, {background color}), button
font color is A (User’s Preference, {button font color}), button background color
is A (User’s Preference, {button background color}), weather icon is A ({local
weather}∪User’s Preference, {weather icon}), background image is A ({local
time}∪User’s Preference, {background image})
3. A ({first time, last unit accuracy, device orientation, local time, local weather}∪ User’s
Preference, {orientation mode}) has two discrete output values: {portrait mode} and
{landscape mode}

Figure 2 is an adaption tree over A ({first time, last unit accuracy, device orientation, local time,
local weather, User’s Preference}, Theme
Adaption Tree for Theme
Figure 2 shows four adaption rules for theme:
1. If a user is using this app for the first time, then the theme action is the default color theme.
2. If a user is not using this app for the first time, and the last unit accuracy is between 0%
and 60% both inclusive, then the theme action is the default theme.
and 90% both exclusive, then the theme action is the preferred color theme.
and 100% both inclusive, then the theme action is the weather & time based theme.
If the fourth adaption rule for theme is met (the theme action is the weather
then we can construct an adaption tree for font color, background color, button font color, button
background, weather icon, background image to show our rule for those UI features, respectively.
Among those UI features, we selected t
Figure 4 shows our adaption tree for background image that is based on the
adaption tree for theme.
her, User’s Preference}, Theme
Figure 2:Adaption Tree for Theme
Figure 2 shows four adaption rules for theme:
If a user is using this app for the first time, then the theme action is the default color theme.
If a user is not using this app for the first time, and the last unit accuracy is between 0%
sive, then the theme action is the default theme.
and 90% both exclusive, then the theme action is the preferred color theme.
If the fourth adaption rule for theme is met (the theme action is the weather & time based theme.),
Among those UI features, we selected the background image to present our approach.
igure 4 shows our adaption tree for background image that is based on the
/4, August 2016
8
If a user is using this app for the first time, then the theme action is the default color theme.
& time based theme.),
he background image to present our approach.

Figure 3:
Figure 3 should be only being considered after Figure 2. The contexts in Figure 2 have higher
priority than the contexts in Figure 3.
Figure 2 and Figure 3 together present lots of rules. For example,if the fourth adaption rule for
theme is met, the background image preference
time is between 17:00 to 19:00, then the background image is a sunset background image.
The implementation of the arithmetic game strictly followed the adaption trees in Figures 2 and
Below are the screen shots for the application with different themes and with local weath
time information as well.
Figure 4 Black and White Theme Figure 5 User’s Preferred Colors
Figure 3:Adaption Tree for Background Image
priority than the contexts in Figure 3.
ground image preference is a time-based background image, and the local
The implementation of the arithmetic game strictly followed the adaption trees in Figures 2 and
Below are the screen shots for the application with different themes and with local weath
/4, August 2016
9
based background image, and the local
The implementation of the arithmetic game strictly followed the adaption trees in Figures 2 and 3.
Below are the screen shots for the application with different themes and with local weather and

Figure 6 Snowy Day
5.CONCLUSIONS
With ubiquitous computing, users access their applications in a wide variety of environments.
To cope with various and dynamic execution environments, the adaptive mobile user interface
is desired to enhance human
address this issue. We used the rule
the adaption rule for the mobile user interface based on the various context information. Our
implementations strictly followed our proposed appr
It is important to point out we are separating how context is acquired from how it is used, by
adapting mobile user interface features to various context information. The user, as a
composite entity, is part of the context.
Each context-aware application has its own set of behaviors to react to context modifications.
Hence, every software engineer needs to clearly understand the goal of the development and
categorize the context in the application. We have proven this idea in two differen
aware applications [16].
The contributions of this research work lie in 1) considering both the user’s domain and mobile
technology experience in context, 2) detailed modeling inclusion on both input and output
data, 3) using the rule to present acquired
a mobile user interface can enhance the accessibility in the e
additional benefits are a) increased usability. For example, if the mobile user interface only
supports one interaction model, such as typing or voice input/sound output, the usability of the
service would be drastically decreased. b) increased awareness of social ethics, e.g. in a quiet
room after midnight, the sound could be turned off automatically. c) improved workfl
productivity because the mobile user interface is automatically adapted to the dynamic
environments.
Figure 6 Snowy Day Figure 7 Rainy Night
ith ubiquitous computing, users access their applications in a wide variety of environments.
is desired to enhance human-computer interactions. This research work is our attempt to
address this issue. We used the rule-based approach, represented as adaption tree to describe
implementations strictly followed our proposed approach.
composite entity, is part of the context.
lication has its own set of behaviors to react to context modifications.
categorize the context in the application. We have proven this idea in two differen
data, 3) using the rule to present acquired knowledge in the application. The adaption built into
a mobile user interface can enhance the accessibility in the e-commerce domain. The
on model, such as typing or voice input/sound output, the usability of the
room after midnight, the sound could be turned off automatically. c) improved workfl
/4, August 2016
10
ith ubiquitous computing, users access their applications in a wide variety of environments.
s our attempt to
based approach, represented as adaption tree to describe
lication has its own set of behaviors to react to context modifications.
categorize the context in the application. We have proven this idea in two different context-
knowledge in the application. The adaption built into
commerce domain. The
on model, such as typing or voice input/sound output, the usability of the
room after midnight, the sound could be turned off automatically. c) improved workflow

11
REFERENCES
[1] Weiser, M., “The computer for the 21st century”, Scientific American, 1991, pp. 94-104.
[2] Dey AK, Abowd GD, “Towards a better understanding of context and context awareness”, New
York: ACM Press, 2000.
[3] Schilit B, Theimer M., “Disseminating active map information to mobile hosts”, IEEE Network
1994, 8(5):22-32.
[4] Brown PJ, Bovey JD, Chen X, “Context-aware applications: from laboratory to the marketplace”,
IEEE Personal Communications 1997, 4(5):58-64.
[5] Ryan N, Pascoe J, Morse D, “Enhanced reality fieldwork: the context-aware archaeological
assistant”, In Dingwall, L., Exon, S., Gaffney, V., Laflin, S., & van Leusen, M., Eds.,Archaeology
in the Age of the Internet. CAA97. Computer Applications and Quantitative Methods in
Archaeology. Proceedings of the 25th Anniversary Con- ference, University of Birmingham, April
1997 (BAR International Series 750).Archaeopress, Oxford, 269-274. Leu- sen, Exxon (Eds.),
Computer Applications in Archaeology. 
[6] Dey AK, “Context aware computing: the cyberdesk project”, Technical Report SS-98-02, 1998, 51-
54.
[7] Franklin D, Flaschbart J., “All gadget and no representation makes jack a dull environment”,
Technical Report SS-98-02. 1998, 155-160.
[8] Ward A, Jones A, Hopper A, “A new location technique for the active office”, IEEE Personal
Commun. 1997, 4(5):42-47.
[9] Rodden T, Cheverst K, Davies K, Dix A, “Exploiting context in HCI design for mobile systems”,
Workshop on Human Computer Interaction with Mobile Devices 1998.
[10] Hull R, Neaves P, Bedford-Roberts J, “Towards situated computing”, 1st International Symposium
on Wearable Computers, 1997, 146-153.
[11] Pascoe J, “Adding generic contextual capabilities to wearable computers”, 2nd International
Symposium on Wearable Computers, 1998, 92-99.
[12] Clerckx, T., Winters, F. and Coninx, K. (2005) Tool Support for Designing Context-Sensitive User
Interface Using a Model-Based Approach. TAMODIA’05 Proceedings of the 4th International
Workshop on Task Models and Diagrams, Gdansk, 26-27 September 2005, 11-18.
[13] Cheng, S.W. and Sun, S.Q. (2009) Adaptive 3D Navigation User Interface Design Based on Rough
Sets. IEEE 10th International Conference on Computer-Aided Industrial Design & Conceptual
Design, Wenzhou, 26-29 November 2009, 1935-1940. 
[14] ReemAlnaniha, b, Olga Ormandjievaa, T. Radhakrishnana, Context-based and Rule-based
Adaptation of Mobile User Interfaces in mHealth, The 3rd International Conference on Current
and Future Trends of Information and Communication Technologies in Healthcare (ICTH),
Procedia Computer Science 21(2013) pp. 390-397.
[15] https://en.wikipedia.org/wiki/Decision_tree
[16] Mao Zheng, Sihan Cheng, QianXu, Context-based Mobile User Interface, Journal of Computer and
Communications, Vol. 4, 2016, pp: 1-9.10.4236/jcc.2016.49001

AUTHORS
Dr. Mao Zheng is a tenured faculty member and associated professor in the
Department of Computer Science at the
of research include Software Engineering, Software Testing and Formal Methods. The
courses she has taught include Software Engineering, Software Design, Object
Oriented Development, Software Testing and Object
in Java. Dr.Zheng received her Ph.D. in Computer Science at Concordia University in
Montreal Canada in 2002. Dr.Zheng has been actively involved with IEEE conferences and served as
reviewers and co-organizers for some IEEE conf
for an international journal.
QianXu is currently working in Amazon in Seattle USA. She obtained the Mater of
Software Engineering degree from the University of Wisconsin
May 2016.
Dr.Hao Fan is a professor at the School of Information Manageme
University in China. His research areas include Data Mining, Data Representation and
Management, and Software Engineering.
Dr. Mao Zheng is a tenured faculty member and associated professor in the
Department of Computer Science at the University of Wisconsin-La Crosse. Her areas
of research include Software Engineering, Software Testing and Formal Methods. The
courses she has taught include Software Engineering, Software Design, Object-
Oriented Development, Software Testing and Object-Oriented Software Development
in Java. Dr.Zheng received her Ph.D. in Computer Science at Concordia University in
organizers for some IEEE conferences. She also has served as an editorial board member
QianXu is currently working in Amazon in Seattle USA. She obtained the Mater of
Software Engineering degree from the University of Wisconsin – LaCrosse in USA in
Dr.Hao Fan is a professor at the School of Information Management in Wuhan
China. His research areas include Data Mining, Data Representation and
Management, and Software Engineering.
/4, August 2016
12
erences. She also has served as an editorial board member

MODELING THE ADAPTION RULE IN CONTEXTAWARE SYSTEMS

More Related Content

What's hot (18)

Similar to MODELING THE ADAPTION RULE IN CONTEXTAWARE SYSTEMS (20)

Recently uploaded (20)

MODELING THE ADAPTION RULE IN CONTEXTAWARE SYSTEMS