Feature selection on Human Activity Recognition dataset using Minimum Redundancy Maximum Relevance

Doewes, Afrizal; Swasono, Sri Edi; Harjito, Bambang

doi:10.1109/icce-china.2017.7991050

Cited by 14 publications

(6 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Doewes et al used the minimum redundancy and maximum correlation feature selection algorithms to analyze the number of selected features and the classification accuracy under different proportions of training sets and test sets, and considered the operation time of the classification process. However, only two classifiers, the support vector machine (SVM) and multilayer perception (MLP), were analyzed, and the MLP training easily dropped into a local optimum, which resulted in a training failure and an affected accuracy evaluation [11]. Li et al proposed a feature set selection algorithm based on adaptive character activity and improved genetic algorithm, which could dynamically guide the process of feature selection and obtain small-scale feature sets on the basis of a higher classification accuracy and a faster running time.…”

Section: Introductionmentioning

confidence: 99%

A Feature Selection and Classification Method for Activity Recognition Based on an Inertial Sensing Unit

Fan

Jia

2019

Information

View full text Add to dashboard Cite

The purpose of activity recognition is to identify activities through a series of observations of the experimenter’s behavior and the environmental conditions. In this study, through feature selection algorithms, we researched the effects of a large number of features on human activity recognition (HAR) assisted by an inertial measurement unit (IMU), and applied them to smartphones of the future. In the research process, we considered 585 features (calculated from tri-axial accelerometer and tri-axial gyroscope data). We comprehensively analyzed the features of signals and classification methods. Three feature selection algorithms were considered, and the combination effect between the features was used to select a feature set with a significant effect on the classification of the activity, which reduced the complexity of the classifier and improved the classification accuracy. We used five classification methods (support vector machine [SVM], decision tree, linear regression, Gaussian process, and threshold selection) to verify the classification accuracy. The activity recognition method we proposed could recognize six basic activities (BAs) (standing, going upstairs, going downstairs, walking, lying, and sitting) and postural transitions (PTs) (stand-to-sit, sit-to-stand, stand-to-lie, lie-to-stand, sit-to-lie, and lie-to-sit), with an average accuracy of 96.4%.

show abstract

Section: Introductionmentioning

confidence: 99%

A Feature Selection and Classification Method for Activity Recognition Based on an Inertial Sensing Unit

Fan

Jia

2019

Information

View full text Add to dashboard Cite

show abstract

“…[41,42]. Finally, different FS methods were used to reduce the number of variables without any transformation, such as Minimum Redundancy Maximum Relevance [43], recursive feature elimination [34], Information Gain [25], or evolutionary algorithms [44].…”

Section: Related Workmentioning

confidence: 99%

Comparison of Different Sets of Features for Human Activity Recognition by Wearable Sensors

Rosati

Balestra

Knaflitz

2018

Sensors

View full text Add to dashboard Cite

Human Activity Recognition (HAR) refers to an emerging area of interest for medical, military, and security applications. However, the identification of the features to be used for activity classification and recognition is still an open point. The aim of this study was to compare two different feature sets for HAR. Particularly, we compared a set including time, frequency, and time-frequency domain features widely used in literature (FeatSet_A) with a set of time-domain features derived by considering the physical meaning of the acquired signals (FeatSet_B). The comparison of the two sets were based on the performances obtained using four machine learning classifiers. Sixty-one healthy subjects were asked to perform seven different daily activities wearing a MIMU-based device. Each signal was segmented using a 5-s window and for each window, 222 and 221 variables were extracted for the FeatSet_A and FeatSet_B respectively. Each set was reduced using a Genetic Algorithm (GA) simultaneously performing feature selection and classifier optimization. Our results showed that Support Vector Machine achieved the highest performances using both sets (97.1% and 96.7% for FeatSet_A and FeatSet_B respectively). However, FeatSet_B allows to better understand alterations of the biomechanical behavior in more complex situations, such as when applied to pathological subjects.

show abstract

“…Then we load our input video (or frame) in gray scale mode. After that we find the midpoint in each frame by using equations (1) and (2). Fig.…”

Section: A Experimental Set Upmentioning

confidence: 99%

“…We can say someone questionable on basis of some specific activities: repetitive touching of face, excessive head-turning, avoids eye contact, powerful grip of a bag or hand inside the bag, excessive clock watching etc. Although a lot of work has been done for basic human activity detection (walking [2][3][4][5], [8], running [10][11][12]18], sitting still [14][15][16]20] etc. [22,23,[26][27][28]); to the best of our knowledge none of these activities mentioned earlier for questionable observer detection have been detected.…”

Section: Introductionmentioning

confidence: 99%

Efficient Framework Using Morphological Modeling for Frequent Iris Movement Investigation towards Questionable Observer Detection

Anisuzzaman¹,

Saif²

2018

IJIGSP

View full text Add to dashboard Cite

This research presents a framework to detect a questionable observer depending on a specific activity named "frequent iris movement". We have focused on some activities and behaviors upon which we can classify one as questionable. So this research area is not only an important part of computer vision and artificial intelligence, but also a major part of human activity recognition (HAR). We have used Haar Cascade Classifier to detect irises of both left and right eyes. Then running some morphological operation we have detected the midpoint between left and right irises; and based on some characteristics of midpoint movement we have detected a specific activityfrequent iris movement. Depending on this activity we are declaring someone as questionable observer. To validate this research we have created our own dataset with 86 videos, where 15 individuals have volunteered. We have achieved an accuracy of 90% for the first 100 frames or 3.33 seconds of each of our videos and an accuracy of 93% for the first 150 frames or 5.00 seconds of each of our videos. No work has been done yet on basis of this specific activity to detect someone as questionable and furthermore our work outperforms most of the existing work on questionable observe detection and suspicious activity recognition.

show abstract

Feature selection on Human Activity Recognition dataset using Minimum Redundancy Maximum Relevance

Cited by 14 publications

References 6 publications

A Feature Selection and Classification Method for Activity Recognition Based on an Inertial Sensing Unit

A Feature Selection and Classification Method for Activity Recognition Based on an Inertial Sensing Unit

Comparison of Different Sets of Features for Human Activity Recognition by Wearable Sensors

Efficient Framework Using Morphological Modeling for Frequent Iris Movement Investigation towards Questionable Observer Detection

Contact Info

Product

Resources

About