Human Activity Recognition using PCA and BiLSTM Recurrent Neural Networks

Aljarrah, Amir A.; Ali, Ali H.

doi:10.1109/iiceta47481.2019.9012979

Cited by 33 publications

(7 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because of its structure, Bi-LSTM can always access previous and next information. It generally outperforms one-way LSTM in data with a heavy dependence on two-way information [ 72 ].…”

Section: Methodsmentioning

confidence: 99%

Artificial Intelligence Based Approach for Classification of Human Activities Using MEMS Sensors Data

Khan

Imaduddin

Singh

et al. 2023

Sensors

View full text Add to dashboard Cite

The integration of Micro Electronic Mechanical Systems (MEMS) sensor technology in smartphones has greatly improved the capability for Human Activity Recognition (HAR). By utilizing Machine Learning (ML) techniques and data from these sensors, various human motion activities can be classified. This study performed experiments and compiled a large dataset of nine daily activities, including Laying Down, Stationary, Walking, Brisk Walking, Running, Stairs-Up, Stairs-Down, Squatting, and Cycling. Several ML models, such as Decision Tree Classifier, Random Forest Classifier, K Neighbors Classifier, Multinomial Logistic Regression, Gaussian Naive Bayes, and Support Vector Machine, were trained on sensor data collected from accelerometer, gyroscope, and magnetometer embedded in smartphones and wearable devices. The highest test accuracy of 95% was achieved using the random forest algorithm. Additionally, a custom-built Bidirectional Long-Short-Term Memory (Bi-LSTM) model, a type of Recurrent Neural Network (RNN), was proposed and yielded an improved test accuracy of 98.1%. This approach differs from traditional algorithmic-based human activity detection used in current wearable technologies, resulting in improved accuracy.

show abstract

“…Because of its structure, Bi-LSTM can always access previous and next information. It generally outperforms one-way LSTM in data with a heavy dependence on two-way information [ 72 ].…”

Section: Methodsmentioning

confidence: 99%

Artificial Intelligence Based Approach for Classification of Human Activities Using MEMS Sensors Data

Khan

Imaduddin

Singh

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…aljarrah et all. [44] obtained 97.64% recognition efficiency using a PCA-BLSTM model in recognising human activities with wearable sensors. sukor et al [45] used accelerometer sensors to recognise human activities such as sitting, lying down, walking and walking up and down stairs, the classifier achieved a recognition rate of 96.11% for the frequency domain features after dimensionality reduction of PCA.…”

Section: Traditional Machine Learning Algorithmsmentioning

confidence: 99%

A Comprehensive Survey on Human Activity Recognition Using Sensing Technology

Wang¹

2022

HSET

View full text Add to dashboard Cite

With the rapid development of machine learning technology and the popularity of various electronic devices such as smartphones and smart bracelets, sensor technology has been applied to all aspects of human society. On this basis, HAR already plays an important role in various fields including sports and the military. In the medical field, particularly, HAR is gaining increasing recognitions by researchers to be one of the main technologies for future behavioral detection in the elderly. The application of HAR technology allows users to identify human behavior through applications equipped with specific sensors, reducing human costs. This survey sorted out the published articles related to HAR research in recent years, provided an analysis of the background, applications, and advantages and disadvantages of sensing technologies, and made a brief exposition of the HAR process.

show abstract

“…The Ref. [ 14 ] achieved accuracy of about 97.64% using principle component analysis (PCA)-bidirectional long short term memory(LSTM) approach so that LSTM recursive neural networks can be trained for prediction of the identified activities performed in the datasets and by using PCA results in the reduction in the datasets number of dimensions of 12 activities.…”

Section: Related Workmentioning

confidence: 99%

“…The dataset used here for the method proposed in this is the [ 14 ] collection of data from various volunteers of 19–48 age group having a smartphone. The phone having both accelerometer and gyroscope can be used to collect the signals’ acceleration and angular velocity at a sampling rate of 50 Hz.…”

Section: Public Datasetsmentioning

confidence: 99%

An Efficient and Lightweight Deep Learning Model for Human Activity Recognition Using Smartphones

Ankita

Rani

Babbar

et al. 2021

Sensors

View full text Add to dashboard Cite

Traditional pattern recognition approaches have gained a lot of popularity. However, these are largely dependent upon manual feature extraction, which makes the generalized model obscure. The sequences of accelerometer data recorded can be classified by specialized smartphones into well known movements that can be done with human activity recognition. With the high success and wide adaptation of deep learning approaches for the recognition of human activities, these techniques are widely used in wearable devices and smartphones to recognize the human activities. In this paper, convolutional layers are combined with long short-term memory (LSTM), along with the deep learning neural network for human activities recognition (HAR). The proposed model extracts the features in an automated way and categorizes them with some model attributes. In general, LSTM is alternative form of recurrent neural network (RNN) which is famous for temporal sequences’ processing. In the proposed architecture, a dataset of UCI-HAR for Samsung Galaxy S2 is used for various human activities. The CNN classifier, which should be taken single, and LSTM models should be taken in series and take the feed data. For each input, the CNN model is applied, and each input image’s output is transferred to the LSTM classifier as a time step. The number of filter maps for mapping of the various portions of image is the most important hyperparameter used. Transformation on the basis of observations takes place by using Gaussian standardization. CNN-LSTM, a proposed model, is an efficient and lightweight model that has shown high robustness and better activity detection capability than traditional algorithms by providing the accuracy of 97.89%.

show abstract

Human Activity Recognition using PCA and BiLSTM Recurrent Neural Networks

Cited by 33 publications

References 15 publications

Artificial Intelligence Based Approach for Classification of Human Activities Using MEMS Sensors Data

Artificial Intelligence Based Approach for Classification of Human Activities Using MEMS Sensors Data

A Comprehensive Survey on Human Activity Recognition Using Sensing Technology

An Efficient and Lightweight Deep Learning Model for Human Activity Recognition Using Smartphones

Contact Info

Product

Resources

About