0000000000421748
AUTHOR
M. Ortolani
Fuzzy Information Granules in Time Series Data
Often, it is desirable to represent a set of time series through typical shapes in order to detect common patterns. The algorithm presented here compares pieces of a different time series in order to find such similar shapes. The use of a fuzzy clustering technique based on fuzzy c-means allows us to detect shapes that belong to a certain group of typical shapes with a degree of membership. Modifications to the original algorithm also allow this matching to be invariant with respect to a scaling of the time series. The algorithm is demonstrated on a widely known set of data taken from the electrocardiogram (ECG) rhythm analysis experiments performed at the Massachusetts Institute of Technol…
User detection through multi-sensor fusion in an AmI scenario
Recent advances in technology, with regard to sensing and transmission devices, have made it possible to obtain continuous and precise monitoring of a wide range of qualitatively diverse environments. This has boosted the research on the novel field of Ambient Intelligence, which aims at exploiting the information about the environment state in order to adapt it to the user’s preference. In this paper, we analyze the issue of detecting the user’s presence in a given region of the monitored area, which is crucial in order to trigger subsequent actions. In particular, we present a comprehensive framework that turns data perceived by sensors of different nature, and with possible imprecision, …
A wireless sensor network for vineyard management in Sicily (Italy)
Wine quality depends on many factors, such as the choice of variety, stock, training system, pruning as well as environmental parameters and cultivation techniques performed in the vineyard. Monitoring the micro-climate of grapevine allows to conveniently perform the most important cultivation techniques (soil management, pesticide treatments, green pruning, harvest) thus reducing the operating costs of the vineyard, and increasing the overall quality of the grapes. The aim of the present study is to monitor the micro-climate of grapevine in order to control spring period hazards, to reduce the operating costs of the vineyard and to increase the quality of grapes. For this purpose a Wireles…
An Intelligent System for Energy Efficiency in a Complex of Buildings
Energy efficiency has nowadays become one of the most challenging task for both academic and commercial organizations, and this has boosted research on novel fields, such as Ambient Intelligence. In this paper we address the issue of timely and ubiquitous monitoring of building complexes in order to optimize their energy consumption, and present an intelligent system addressed to the typical end user, i.e. the administrator, or responsible operator, of the complex. A three-level architecture has been designed for detecting the presence of the building inhabitants user and understanding their preferences with respect to the environmental conditions in order to optimize the overall energy eff…
Visual and Hearing Impairment Are Associated With Delirium in Hospitalized Patients: Results of a Multisite Prevalence Study
Objective: Sensory deficits are important risk factors for delirium but have been investigated in single-center studies and single clinical settings. This multicenter study aims to evaluate the association between hearing and visual impairment or bi-sensory impairment (visual and hearing impairment) and delirium. Design: Cross-sectional study nested in the 2017 "Delirium Day" project. Setting and participants: Patients 65 years and older admitted to acute hospital medical wards, emergency departments, rehabilitation wards, nursing homes, and hospices in Italy. Methods: Delirium was assessed with the 4AT (a short tool for delirium assessment) and sensory deficits with a clinical evaluation. …
An Autonomic System for Estimating Human Presence through Bayesian Networks
In the Ambient Intelligence (AmI) context, a relevant research topic is represented by the methods for determining users' presence in order to design context-aware systems capable of monitoring the environment in which they operate, and of timely reacting to changes. This work describes an autonomic software agent comprising a double-level reasoning. At the lower level, a Bayesian network merges the available sensory information related to the users' presence, whereas the upper level performs a meta-reasoning on the system performance and configuration in order to enable the system self-assessment. Experimental results show the validity of the proposed method on a sample scenario.