Context Inference Engine (CiE): Classifying Activity of Context using Minkowski Distance and Standard Deviation-Based Ranks

Context Inference Engine (CiE): Classifying Activity of Context using Minkowski Distance and Standard Deviation-Based Ranks

Umar Mahmud (National University of Sciences and Technology (NUST), Pakistan) and Muhammad Younus Javed (National University of Sciences and Technology (NUST), Pakistan)
DOI: 10.4018/978-1-4666-6098-4.ch003
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Context Awareness is the ability of systems and applications to sense the environment and infer the activity going on in the environment. Context encompasses all knowledge bounded within an environment and includes attributes of both machines and users. A context-aware system is composed of context gathering and context inference modules. This chapter proposes a Context Inference Engine (CiE) that classifies the current context as one of several known context activities. This engine follows a Minkowski distance-based classification approach with standard deviation-based ranks to identify likeliness of classified activity of the current context. Empirical results on different data sets show that the proposed algorithm performs closer to Support Vector Machines (SVM) while it is better than probabilistic reasoning methods where the performance is quantified as success in classification.
Chapter Preview
Top

2. Evidences Of Context Classification Approaches In Literature

The early context aware systems are designed as applications that consider spatial and temporal factors as the context. These applications follow RBA for classification of the activity of contextual data (Schilit, Hilbert, & Trevor, 2002), (Want, Hopper, Falcao, & Gibbons, 1992), (Cheverst, Mitchell, & Davies, 1999), (Abowd, Atkenson, Hong, Long, Kooper, & Pinkerton, 1997), (Román, Hess, Cerqueira, Campbell, & Nahrstedt, 2002), (Hofer, Pichler, Leonhartsberger, Altmann, & Werner, 2002), (Wang Y.-K., 2004), (de Deugd, Carroll, Kelly, Millett, & Ricker, 2006), (Dey & Abowd, 1999), (Riaz, Kiani, Lee, Han, & Lee, 2005), (Chen, Finin, & Joshi), (Mahmud, Iltaf, Rehman, & Kamran, 2007), (Gu, Pung, & Zhang, 2004), (Fahy & Clarke, 2004), (Guo, Gao, Ma, Li, & Huang, 2008). Use of context matching is an improvement on RBA (Samulowitz, Michahelles, & Linnhoff-Popien, 2001), (Xue, Pung, & Sen, 2013). RBA is an efficient technique but lacks flexibility in terms of learning new axioms and situations. The effort put in by the developer to introduce new axioms or modify them in different software engineering phases is high. Context aware systems are thus restricted to the quantity as well as the quality of the axioms developed.

Complete Chapter List

Search this Book:
Reset