Use this URL to cite or link to this record in EThOS:
Title: Trend analysis for human activities recognition
Author: Elbayoudi, Abubaker
ISNI:       0000 0004 7233 9556
Awarding Body: Nottingham Trent University
Current Institution: Nottingham Trent University
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
Smart environments equipped with appropriate sensory devices are used to measure people's activities. These activities represent Activities of Daily Living (ADL) or Activities of Daily Working (ADW). Measuring progressive changes in activities is a subject of research interest. A number of medical conditions and their treatments are associated with progressive changes such as reduced movement over time. The aim of this research is to determine means of inspecting trends in the ADL/ADW to identify progressive changes and predict behavioural abnormalities. The ADL/ADW pattern will change over time and this is a consequence of the individual's condition. Identifying evolving behavioural patterns will help to predict the trend in the ADL/ADW behavioural pattern before any abnormalities are identifed. The data provided for this investigation are from real environments home and office). Additionally, a simulator is developed to generate simulated data for ADLs. To answer the research question identifed in this research, the initial investigation was conducted and a novel Human Behaviour Momentum Indicator (HBMI) is proposed. The HBMI is introduced to identify changes based on activities recorded from a single sensor. To show the effectiveness of the proposed approach, results are compared with Relative Strength Index (RSI). The results show that trends in ADL or ADW can be detected and the direction of the activity's trend is predicted. To represent a holistic report based on a multiple sensors/activities representing progressive changes in the participant's behaviour, a novel Human Behaviour Indicator (HBI) is also proposed. The proposed HBI indicator is constructed as a composite indicator, which will compute progressive changes in behaviour based on the events that are performed during the entire day. The percentage of changes between events is used to compare events and measure the progressive changes. The proposed technique identifies the user's daily behaviour and distinguishes between normal and abnormal behavioural patterns of the ADLs or ADWs. Analysis of the data indicates that the HBI could clearly differentiate between the normal and the abnormal behaviour and give a warning status with a confidence level. Identifying trends in ADLs or ADWs using trend analysis techniques are investigated to interpret the behavioural changes in a suitable format to be understood by the carers or supervisors.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available