CONTEXT AWARENESS 1 CONTEXT

The word context can be lexically defined as a set of activities that are a result of a situation or event. Even though there is not much consensus about the idea or definition of context still most of the research work on context- awareness is based on the following definitions.

“Any information that can be used to characterize the situation of entities that are considered relevant to the interaction between a user and an application, including the user and the application themselves”Dey, (2001).

Henricksen et al. (2002) divides the idea of context into two major types, namely Static and Dynamic context. Static Context: is one in which the context never changes during the context period as in the case of the personal-calendar, contacts in a phonebook, user profile, user preference and hardware profile. But with the dynamic context, information is highly variable. Some of the dynamic context variablesinclude location, time, surrounding resources and system status Henricksen, et al. (2002). The researchers working on the idea of context awareness are two schools of thoughts: The positivist theory and Phenomenological theory. However, the fundamental work on context awareness heavily depends on the definition by Dey (2001), Mostéfaoui (2004), Gu (2005) and Baldauf (2007).

Two prominent views exist: Positivist theory which defines context as,

“Context can be described independently of the actions done” while Phenomenological theory defines context as, “Context emerges from the activity and cannot be described independently” Dourish (2004). Some of the most important definitions of context and context-aware computing are as follows:

“Something is context because of the way it is used in interpretation”Winograd (2001).

25

“A pattern of behaviour or relations among variables that are outside of the subjects of design manipulation and potentially affect user behaviour and system performance” Sato (2003).

The purpose of defining context is widely understood as the presentation of information and services to a user. These services can be executed or context could be tagged with information for later retrieval. Three main classes of context items of information are People (location and identity), Environment (time) and Activities.

2.2.2 CONTEXT-AWARENESS

Applications based on context-awareness have certain design principles.

Various models have been proposed for representing contextual information.

Conditions such as local or remote sensors, the possible number of users, and involvement of mobile devices or personal computers are all considered while designing context-aware systems. Based on the data structures used, Linnhoff-Popien and Strang (2004) classified context awareness which is summarized and tabulated in Table 2.1 McCarthy and Buvac (1997); Hofer, et al. (2002); Sheng and Benatalla, (2005)

When a system is context-aware, there would be the enrichment of communication and useful services can be unearthed. To determine Context- awareness it follows subprocesses namely context acquisition, processing, and usage. Data is first acquired from sensors like temperature, pressure, etc.

then in the next step, four actions are mandatory. At context processing stage, removal of Noise in the data, calibration of data, interpretation and prediction of context are processed Gite, et al. (2016).

Table 2.1 Context-awareness models

Model Features

Key-Value Models

Simple data structure. Mainly used in service frameworks. Describes the capabilities of a service as key-value pairs.

Markup Models

Hierarchical data structure using markup scheme model. Profiles are created using markup tags, attributes, and content.

Graphical Models Context data is modelled using Unified Modelling Language (UML).

Object-Oriented Models

Context data represented using this technique offers encapsulation, inheritance and reusability.

Context attributes like Location, Temperature, etc. are defined as objects.

Logic-based Models

The context model is defined using rules, facts, and expressions. Context reasoning, new facts, are derived using existing rules.

Ontology-based Models

Context data is represented as relationships with their concepts. Ontology reasoning techniques and formal expressions are used in modelling the contextual data.

User-Context perception Model

Context data derived as sensory information should be constantly relayed back to the user to avoid a mismatch between user’s sensory perception and sensory input.

27

2.2.3 CONTEXT AWARE COMPUTING

Objects or entities involved in a situation would indicate the context either explicitly or implicitly. Therefore, the pressing need of the context-aware application developer is to reduce the human effort to understand the context. Mark Weiser introduced the term context-aware computing which resulted from the research work on ubiquitous computing Weiser (1999).

Weiser defined is as “the most profound technologies are those that disappear”. The idea aims in integrating devices, mainly sensors, and actuators, seamlessly and offer anywhere and anytime communication and computation with anything allowing users to stay decoupled from devices Schilit and Theimer, (1994); Dey (2001); Hill, et al. (2004); Hong, et al.

(2009). “An application’s ability to adapt to changing circumstances and responds according to the context of use” Dey (2001).

It envisions multisensory perception and interaction with the intelligence and sense to interpret the information acquired from the sensors connected to an environment of seamless networking. Pervasive computing is being deployed in healthcare and this service is widely known as Pervasive healthcare Sriram, et al. (2015).

Any context-aware system should have the following context-oriented functionalities. It should be able to accommodate a variety of sensor-based devices. Should be able to understand context as it emanates from a distributed heterogeneous sources. Contextual data should be abstracted, while the interpretation of applications should be transparent. Context storage should be maintained well and data flow should be efficiently controlled Chitra and Adane, (2015).

Linnhoff-Popien and Strang (2004) states that context-aware systems have demanded the following terms.

1. Distributed Composition: absence of a central system for coordinating data maintenance and service provisioning.

2. Partial Validation: since it is recommended to be a distributed system, it is well known that complete contextual information will not be available in any particular node; however, it is still desirable to validate the information to make it error-free.

3. Richness and Quality of information: context-aware systems rely heavily on sensors which are hardware components and in due times are prone to give deteriorated quality of information. Also, sensors from different vendors may give the different richness of information.

4. Incompleteness and ambiguity: Interpolation of data might be required since data gathered from sensor networks might be incomplete or ambiguous.

5. The level of formality: Presenting contextual facts with their relationships is quite a challenge. In other words, interpretation should be common among all the participating objects of context.

6. Applicability of existing environments: context models should be within the available infrastructure.