Human activities recognition in smart living environment
Today, in the mobile internet era, sensors are now widely used around the world. The development and application of sensors benefits humans in many fields. In the recent years, sensor-based activities recognition has made great progress. Among them, activities recognition research based on wearable...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
2019
|
Subjects: | |
Online Access: | http://hdl.handle.net/10356/78432 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-78432 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-784322023-07-07T16:18:50Z Human activities recognition in smart living environment Ye, Yuchen Soh Yeng Chai School of Electrical and Electronic Engineering DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Control engineering Today, in the mobile internet era, sensors are now widely used around the world. The development and application of sensors benefits humans in many fields. In the recent years, sensor-based activities recognition has made great progress. Among them, activities recognition research based on wearable sensors and smartphones’ sensors have occupied a major position and provide a lot of support of application in human’s daily life. Smartphone is easy to carry, due to this advantage, a large number of researchers use smartphone to collect sensor data and research. In this project, MATLAB was used to process accelerometer sensor data from smartphone, then examines the best accuracy that can be achieved by using different machine learning algorithms including K-Near Neighbor (K-NN), Support Vector Machines (SVM) and Ensemble Learner. MATLAB is a convenient software to do machine learning. MATLAB’s Classification Learner app which provides users several classifiers and visual interface of sensor data will be used in this experiment. The result of experiment shows that all the machine learning algorithms can reach 85 or higher. The highest accuracy that can be achieved is 94.74% by using Cubic SVM. Bachelor of Engineering (Electrical and Electronic Engineering) 2019-06-20T02:45:28Z 2019-06-20T02:45:28Z 2019 Final Year Project (FYP) http://hdl.handle.net/10356/78432 en Nanyang Technological University 63 p. application/pdf |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Control engineering |
spellingShingle |
DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Control engineering Ye, Yuchen Human activities recognition in smart living environment |
description |
Today, in the mobile internet era, sensors are now widely used around the world. The development and application of sensors benefits humans in many fields. In the recent years, sensor-based activities recognition has made great progress. Among them, activities recognition research based on wearable sensors and smartphones’ sensors have occupied a major position and provide a lot of support of application in human’s daily life. Smartphone is easy to carry, due to this advantage, a large number of researchers use smartphone to collect sensor data and research. In this project, MATLAB was used to process accelerometer sensor data from smartphone, then examines the best accuracy that can be achieved by using different machine learning algorithms including K-Near Neighbor (K-NN), Support Vector Machines (SVM) and Ensemble Learner. MATLAB is a convenient software to do machine learning. MATLAB’s Classification Learner app which provides users several classifiers and visual interface of sensor data will be used in this experiment. The result of experiment shows that all the machine learning algorithms can reach 85 or higher. The highest accuracy that can be achieved is 94.74% by using Cubic SVM. |
author2 |
Soh Yeng Chai |
author_facet |
Soh Yeng Chai Ye, Yuchen |
format |
Final Year Project |
author |
Ye, Yuchen |
author_sort |
Ye, Yuchen |
title |
Human activities recognition in smart living environment |
title_short |
Human activities recognition in smart living environment |
title_full |
Human activities recognition in smart living environment |
title_fullStr |
Human activities recognition in smart living environment |
title_full_unstemmed |
Human activities recognition in smart living environment |
title_sort |
human activities recognition in smart living environment |
publishDate |
2019 |
url |
http://hdl.handle.net/10356/78432 |
_version_ |
1772826795885723648 |