Lorcan Coyle's Home on Tinterweb

Keywords:

Abstract:

Activity recognition has mainly focused to date on identifying repetitious and/or clearly delineated events. Our experience, drawing on many years' research in smart and sensorised systems, leads us to observe that many (if not most) interesting activities fall into a different category: sporadically occurring and poorly differentiated from other concurrent activities. This implies that decision-making remains uncertain across the entire system, and suggests that progress would be greatly supported by standard evaluation methodologies and data sets.

• Tagged
• XML
• BibTex

Keywords:

Abstract:

The ability to identify the behavior of people in a home is at the core of Smart Home functionality. Such environments are equipped with sensors that unobtrusively capture information about the occupants. Reasoning mechanisms transform the technical, frequently noisy data of sensors into meaningful interpretations of occupant activities. Time is a natural human way to reason about activities. Peoples' activities in the home often have an identifiable routine; activities take place at distinct times throughout the day and last for predicable lengths of time. However, the inclusion of temporal information is still limited in the domain of activity recognition. Evidence theory is gaining increasing interest in the field of activity recognition, and is suited to the incorporation of time related domain knowledge into the reasoning process. In this paper, an evidential reasoning framework that incorporates temporal knowledge is presented. We evaluate the effectiveness of the framework using a third party published smart home dataset. An improvement in activity recognition of 70% is achieved when time patterns and activity durations are included in activity recognition. We also compare our approach with Naïve Bayes classifier and J48 Decision Tree, with temporal evidence theory achieving higher accuracies than both classifiers.

• Tagged
• XML
• BibTex

Keywords:

Abstract:

Pervasive computing requires the ability to detect user activity in order to provide situation-specific services. Case-based reasoning can be used for activity recognition by using sensor data obtained from the environment. Pervasive computing systems can grow to be very large, containing many users, sensors, objects and situations, thus raising the issue of scalability. This paper presents a case-based reasoning approach to activity recognition in a smart home setting. An analysis is performed on scalability with respect to case storage, and an ontology-based approach is proposed for case base maintenance. We succeeded in reducing the casebase size by a factor of one thousand, while increasing the accuracy in recognising some activities.

• Tagged
• XML
• BibTex

Keywords:

Abstract:

Through advances in sensing technology, a huge amount of data is available to context-aware applications. A major challenge is extracting features of this data that correlate to high-level human activities. Time, while being semantically rich and an essentially free source of information, has not received sufficient attention for this task. In this paper, we examine the potential for taking temporal features—inherent in human activities—into account when classifying them. Preliminary experiments using the PlaceLab dataset show that absolute time and temporal relationships between activities can improve the accuracy of activity classifiers.

• Tagged
• XML
• BibTex

Keywords:

Abstract:

Much research on context quality in context-aware systems divides into two strands: (1) the qualitative identification of quality measures and (2) the use of uncertain reasoning techniques. In this paper, we combine these two strands, exploring the problem of how to identify and propagate quality through the different context layers in order to support the context reasoning process. We present a generalised, structured context quality model that supports aggregation of quality from sensor up to situation level. Our model supports reasoning processes that explicitly aggregate context quality, by enabling the identification and quantification of appropriate quality parameters. We demonstrate the efficacy of our model using an experimental sensor data set, gaining a significant improvement in situation recognition for our voting based reasoning algorithm.

• Tagged
• XML
• BibTex

Keywords:

Abstract:

The area of activity identification is maturing well in the HCI and ubiquitous computing fields. However, although algorithm development is proceedings well, without publicly available datasets on which to compare results it is difficult to consolidate the disparate work being done. This problem exists because realistic datasets describing human activity are difficult and expensive to gather and because there are significant barriers to releasing the data once gathered. We review positive recent development with the release of two high-quality datasets. From our experiences using these datasets we list some recommendations for the gathering and release of future datasets. Finally, we propose a strategy of our own for gathering a new dataset from these recommendations.

• Tagged
• XML
• BibTex

Keywords:

Abstract:

Applications that rely on mobile devices for user interaction must be mindful of the user's limited attention, which will typically be split between several competing tasks. Content delivery in such systems must be adapted closely to users' evolving situations and shifting priorities, in a way that cannot be accomplished using static filtering determined a priori. We propose a more dynamic context-driven approach to content delivery. We demonstrate our approach using Scatterbox, a pervasive computing application we have developed which performs sensor fusion to derive a user's current situation. Based on the user's level of interruptibility, Scatterbox prioritises and forwards relevant messages to their mobile phone. We draw conclusions from a preliminary evaluation of the system.

• Tagged
• XML
• BibTex

URL:

Keywords:

• Tagged
• XML
• BibTex

ISBN:

Keywords:

Abstract:

Context-aware systems typically use data sensed from the environment to drive adaptive behaviour. Sensed data is inherently imprecise and uncertain; in addition, new uncertainties are introduced when sensed data is fused with other data to infer context at the more abstract level of situations. We present an uncertainty model which aggregates context uncertainty and provides mechanisms to capture uncertainty at situation level. We demonstrate the application of the model in a sample scenario using an experimental data set.

• Tagged
• XML
• BibTex

URL:

Keywords:

Abstract:

With the general increase of life span that our advances in health care have afforded us, more people are suffering from short term memory loss than ever before. Short term memory sufferers often forget what they were doing in the middle of a task and can find themselves in dangerous situations, such as leaving the stove on and leaving the house. They could benefit from an RFID based reminder system that would determine what they were doing based on what objects they touch. To use the system, the user wears an RFID glove which has a reader in the palm. The RFID glove reads the tags on the nearby objects. Along with the RFID glove we are developing an application that enables the user to interact with a reminder application. The application alerts the user of important activities they may have forgotten they started and when an activity is interrupted. It also keeps a record of the list of activities they have performed and objects they have touched through out the day.

• Tagged
• XML
• BibTex