-
Embodiment perception of a smart home assistant
Authors:
Mariya Kilina,
Tommaso Elia,
Syed Yusha Kareem,
Alessandro Carfi,
Fulvio Mastrogiovanni
Abstract:
Demographic growth and rise in the average age of the population is increasing the demand for the elderly assistance. Health care oriented ambient intelligence technologies are fundamental to support elderly peoples' autonomy. In this paper, we present a smart home system that is able to recognize human activities and is integrated with a proactive vocal assistant. We chose one of possible user sc…
▽ More
Demographic growth and rise in the average age of the population is increasing the demand for the elderly assistance. Health care oriented ambient intelligence technologies are fundamental to support elderly peoples' autonomy. In this paper, we present a smart home system that is able to recognize human activities and is integrated with a proactive vocal assistant. We chose one of possible user scenarios to show the performance of this smart home system and to perform a preliminary comparison between users' experience while watching videos of a volunteer interacting with an embodied versus a not-embodied assistant. The scenario is recorded from the user's point of view, while the user interacts with a robot assistant or a simple vocal assistant. The results of the User Experience Questionnaire show that participants found the robot assistant considerably more attractive, innovative and stimulating in comparison to the vocal assistant.
△ Less
Submitted 28 April, 2023;
originally announced April 2023.
-
OWLOOP: A Modular API to Describe OWL Axioms in OOP Objects Hierarchies
Authors:
Luca Buoncompagni,
Syed Yusha Kareem,
Fulvio Mastrogiovanni
Abstract:
OWLOOP is an Application Programming Interface (API) for using the Ontology Web Language (OWL) by the means of Object-Oriented Programming (OOP). It is common to design software architectures using the OOP paradigm for increasing their modularity. If the components of an architecture also exploit OWL ontologies for knowledge representation and reasoning, they would require to be interfaced with OW…
▽ More
OWLOOP is an Application Programming Interface (API) for using the Ontology Web Language (OWL) by the means of Object-Oriented Programming (OOP). It is common to design software architectures using the OOP paradigm for increasing their modularity. If the components of an architecture also exploit OWL ontologies for knowledge representation and reasoning, they would require to be interfaced with OWL axioms. Since OWL does not adhere to the OOP paradigm, such an interface often leads to boilerplate code affecting modularity, and OWLOOP is designed to address this issue as well as the associated computational aspects. We present an extension of the OWL-API to provide a general-purpose interface between OWL axioms subject to reasoning and modular OOP objects hierarchies.
△ Less
Submitted 31 December, 2021;
originally announced December 2021.
-
Human Activity Recognition Models in Ontology Networks
Authors:
Luca Buoncompagni,
Syed Yusha Kareem,
Fulvio Mastrogiovanni
Abstract:
We present Arianna+, a framework to design networks of ontologies for representing knowledge enabling smart homes to perform human activity recognition online. In the network, nodes are ontologies allowing for various data contextualisation, while edges are general-purpose computational procedures elaborating data. Arianna+ provides a flexible interface between the inputs and outputs of procedures…
▽ More
We present Arianna+, a framework to design networks of ontologies for representing knowledge enabling smart homes to perform human activity recognition online. In the network, nodes are ontologies allowing for various data contextualisation, while edges are general-purpose computational procedures elaborating data. Arianna+ provides a flexible interface between the inputs and outputs of procedures and statements, which are atomic representations of ontological knowledge. Arianna+ schedules procedures on the basis of events by employing logic-based reasoning, i.e., by checking the classification of certain statements in the ontologies. Each procedure involves input and output statements that are differently contextualised in the ontologies based on specific prior knowledge. Arianna+ allows to design networks that encode data within multiple contexts and, as a reference scenario, we present a modular network based on a spatial context shared among all activities and a temporal context specialised for each activity to be recognised. In the paper, we argue that a network of small ontologies is more intelligible and has a reduced computational load than a single ontology encoding the same knowledge. Arianna+ integrates in the same architecture heterogeneous data processing techniques, which may be better suited to different contexts. Thus, we do not propose a new algorithmic approach to activity recognition, instead, we focus on the architectural aspects for accommodating logic-based and data-driven activity models in a context-oriented way. Also, we discuss how to leverage data contextualisation and reasoning for activity recognition, and to support an iterative development process driven by domain experts.
△ Less
Submitted 5 May, 2021;
originally announced May 2021.
-
Arianna+: Scalable Human Activity Recognition by Reasoning with a Network of Ontologies
Authors:
Syed Yusha Kareem,
Luca Buoncompagni,
Fulvio Mastrogiovanni
Abstract:
Aging population ratios are rising significantly. Meanwhile, smart home based health monitoring services are evolving rapidly to become a viable alternative to traditional healthcare solutions. Such services can augment qualitative analyses done by gerontologists with quantitative data. Hence, the recognition of Activities of Daily Living (ADL) has become an active domain of research in recent tim…
▽ More
Aging population ratios are rising significantly. Meanwhile, smart home based health monitoring services are evolving rapidly to become a viable alternative to traditional healthcare solutions. Such services can augment qualitative analyses done by gerontologists with quantitative data. Hence, the recognition of Activities of Daily Living (ADL) has become an active domain of research in recent times. For a system to perform human activity recognition in a real-world environment, multiple requirements exist, such as scalability, robustness, ability to deal with uncertainty (e.g., missing sensor data), to operate with multi-occupants and to take into account their privacy and security. This paper attempts to address the requirements of scalability and robustness, by describing a reasoning mechanism based on modular spatial and/or temporal context models as a network of ontologies. The reasoning mechanism has been implemented in a smart home system referred to as Arianna+. The paper presents and discusses a use case, and experiments are performed on a simulated dataset, to showcase Arianna+'s modularity feature, internal working, and computational performance. Results indicate scalability and robustness for human activity recognition processes.
△ Less
Submitted 21 September, 2018;
originally announced September 2018.
