Bayesian approach to sensor-based context awareness

Visually demanding interfaces on a mobile phone can diminish the user experience by monopolizing the user's attention when they are focusing on another task and impede accessibility for visually impaired users. Because mobile devices are often located in pockets when users are mobile, explicit foot movements can be defined as eyes-and-handsfree input gestures for interacting with the device. In this work, we study the human capability associated with performing foot-based interactions which involve lifting and rotation of the foot when pivoting on the toe and heel. Building upon these results, we then developed a system to learn and recognize foot gestures using a single commodity mobile phone placed in the user's pocket or in a holster on their hip. Our system uses acceleration data recorded by a built-in accelerometer on the mobile device and a machine learning approach to recognizing gestures. Through a lab study, we demonstrate that our system can classify ten different foot gestures at approximately 86% accuracy.

show abstract

“…have previously discussed the feasibility of using Naïve Bayes in a mobile system to recognize user contexts [16].…”

Section: Classification Algorithmmentioning

confidence: 99%

Sensing foot gestures from the pocket

Scott

Dearman

Yatani

et al. 2010

Proceedings of the 23nd Annual ACM Symposium on User Interface Software and Technology

View full text Add to dashboard Cite

show abstract

“…Fuzzy context information has been derived in case of uncertainty [11]. Naive Bayesian networks have also been used for context classification [12]. Kramer et al [13] proposed an engine for context acquisition, composition and broadcasting to be shared by Android context-aware mobile applications.…”

Section: Related Workmentioning

confidence: 99%

Context Management for Supporting Context-aware Android Applications Development

Elazhary

Althubyani

Ahmed

et al. 2017

Int. J. Interact. Mob. Technol.

View full text Add to dashboard Cite

Abstract-Building context-aware mobile applications is one of the most ambitious areas of research. Such applications can change their behavior according to context or perform specific tasks in specific contexts. Regardless of the application, all context-aware mobile applications share the need to retrieve and process context information. This paper presents a Context Management tool for the Android platform (ACM). ACM allows easy access to internal onboard mobile sensors and hardware features extracting corresponding raw data. Raw context is processed into higher-level more human-readable context that is provided seamlessly to the mobile applications. Different methods are used for this purpose including fuzzy classifiers. Since different mobiles have different sensors and hardware features, ACM can adapt to the mobile device by deactivating access to unavailable ones. Information regarding the available sensors and hardware features and their specifications can also be queried. Additionally, applications can request notifications regarding context change or specific context values. In addition to providing developers with supporting classes and methods, ACM is accompanied by an application that allows developers to examine its functionality and capabilities before using it. The application can be also used to examine the readings of the different sensors in different situations and thus calibrate them as needed. Additionally, it can be used to modify and personalize default interpretations of raw context values to high-level ones. ACM has been tested empirically and the results show extreme interest of context-aware mobile application developers in its promising capabilities and that it is conducive to facilitating, speeding up and triggering development of many more of such applications.

show abstract

“…Korpipää et al [14] employ a naive Bayesian classifier and an extensive set of audio features derived partly from the algorithms of the MPEG-7 standard. The classification is based mainly on audio features measured in a home scenario.…”

Section: B Acoustic Context Recognitionmentioning

confidence: 99%

Adaptive Audio-Based Context Recognition

Dargie

2009

IEEE Trans. Syst., Man, Cybern. A

View full text Add to dashboard Cite

Abstract-Context recognition is an essential aspect of intelligent systems and environments. In most cases, the recognition of a context of interest cannot be achieved in a single step. Between measuring a physical phenomenon and the estimation or recognition of what this phenomenon represents, there are several intermediate stages which require a significant computation. Understanding the resource requirements of these steps is vital to determine the feasibility of context recognition on a given device. In this paper, we propose an adaptive context-recognition architecture that accommodates uncertain knowledge to deal with sensed data. The architecture consists of an adaptation component that monitors the capability and workload of a device and dynamically adapts recognition accuracy and processing time. The architecture is implemented for an audio-based context recognition. A detail account of the tradeoff between recognition time and recognition accuracy is provided.

show abstract

Bayesian approach to sensor-based context awareness

Cited by 92 publications

References 13 publications

Sensing foot gestures from the pocket

Sensing foot gestures from the pocket

Context Management for Supporting Context-aware Android Applications Development

Adaptive Audio-Based Context Recognition

Contact Info

Product

Resources

About