Using situation lattices in sensor analysis

Ye, Juan; Coyle, Lorcan; Dobson, Simon; Nixon, Paddy

doi:10.1109/percom.2009.4912762

Cited by 16 publications

(20 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As more events are injected, the chance of them being separated in different groups becomes higher. The recalls on the TVK B and PlaceLab datasets are worse than the other two datasets, which is partly due to the inherent noise in these datasets [20,21,32]; e.g., we detect abnormal events that are not injected but already exist in the data.…”

Section: Rade Resultsmentioning

confidence: 91%

Detecting abnormal events on binary sensors in smart home environments

Stevenson

Dobson

2016

Pervasive and Mobile Computing

Self Cite

View full text Add to dashboard Cite

With a rising ageing population, smart home technologies have been demonstrated as a promising paradigm to enable technology-driven healthcare delivery. Smart home technologies, composed of advanced sensing, computing, and communication technologies, offer an unprecedented opportunity to keep track of behaviours and activities of the elderly and provide context-aware services that enable the elderly to remain active and independent in their own homes. However, experiments in developed prototypes demonstrate that abnormal sensor events hamper the correct identification of critical (and potentially life-threatening) situations, and that existing learning, estimation, and time-based approaches to situation recognition are inaccurate and inflexible when applied to multiple people sharing a living space. We propose a novel technique, called CLEAN, that integrates the semantics of sensor readings with statistical outlier detection. We evaluate the technique against four real-world datasets across different environments including the datasets with multiple residents. The results have shown that CLEAN can successfully detect sensor anomaly and improve activity recognition accuracies.

show abstract

Section: Rade Resultsmentioning

confidence: 91%

Detecting abnormal events on binary sensors in smart home environments

Stevenson

Dobson

2016

Pervasive and Mobile Computing

Self Cite

View full text Add to dashboard Cite

show abstract

“…Pieces of context information are combined and interpreted to deduce the situation that users or devices are currently involved in. This classification of the information is similar to that followed by [13], and allows us to accommodate learning-based techniques for the recognition of user activities like [11] within a specification-based approach.…”

Section: A Model For Pervasive Situation-awarenessmentioning

confidence: 93%

“…These require the annotation of the examples needed for the learning to take place. In recent years, there has been some work to try and minimize the number of examples, like [1,13], by exploiting additional structure and knowledge to simplify the learning task. However, these approaches just trade the effort required for the annotation of the examples, with work by domain experts for the specification of the necessary structure and knowledge.…”

Section: Related Work On Situation-awarenessmentioning

confidence: 99%

See 1 more Smart Citation

A Middleware for Pervasive Situation-Awareness

Thomson

Terzis

2012

Distributed Applications and Interoperable Systems

View full text Add to dashboard Cite

Abstract. Situation-awareness is the ability of applications to adapt to the current situation of their users. For situation-awareness to be truly pervasive it should support the individual needs of every user, everywhere. We present a middleware for pervasive situation-awareness based on the idea of separating the features of a situation from the specification of how it should be recognised. The features of a situation can be seen as an interface that can be easily customised to satisfy individual user needs, while alternative specifications can be used to recognise a situation in different environments. The middleware views situations as collections of roles that individuals and devices play. Its implementation follows an agent-based architecture where collaborating agents acquire and reason over context data. We also show that the middleware can recognise a variety of highly customised situations using alternative specifications with performance that is acceptable for interactive situation-aware applications in realistic deployment sizes.

show abstract

“…It is challenging, however. For instance, there are diverse kinds of noise existed in the captured contexts [13] and significant errors in recovering actual situations based on given contexts even though current state-of-the-art techniques have been used [14]. Still, programmers are expected to develop reliable solutions for the actual situations.…”

Section: Introductionmentioning

confidence: 99%

Weaving Context Sensitivity into Test Suite Construction

Wang

Chan

2009

2009 IEEE/ACM International Conference on Automated Software Engineering

View full text Add to dashboard Cite

Abstract-Context-aware applications capture environmental changes as contexts and self-adapt their behaviors dynamically. Existing testing researches have not explored context evolutions or their patterns inherent to individual test cases when constructing test suites. We propose the notation of context diversity as a metric to measure how many changes in contextual values of individual test cases. In this paper, we discuss how this notion can be incorporated in a test case generation process by pairing it with coverage-based test data selection criteria.

show abstract

Using situation lattices in sensor analysis

Cited by 16 publications

References 20 publications

Detecting abnormal events on binary sensors in smart home environments

Detecting abnormal events on binary sensors in smart home environments

A Middleware for Pervasive Situation-Awareness

Weaving Context Sensitivity into Test Suite Construction

Contact Info

Product

Resources

About