Prediction of Lower Extremity Multi-Joint Angles during Overground Walking by Using a Single IMU with a Low Frequency Based on an LSTM Recurrent Neural Network

The joint angle during gait is an important indicator, such as injury risk index, rehabilitation status evaluation, etc. To analyze gait, inertial measurement unit (IMU) sensors have been used in studies and continuously developed; however, they are difficult to utilize in daily life because of the inconvenience of having to attach multiple sensors together and the difficulty of long-term use due to the battery consumption required for high data sampling rates. To overcome these problems, this study propose a multi-joint angle estimation method based on a long short-term memory (LSTM) recurrent neural network with a single low-frequency (23 Hz) IMU sensor. IMU sensor data attached to the lateral shank were measured during overground walking at a self-selected speed for 30 healthy young persons. The results show a comparatively good accuracy level, similar to previous studies using high-frequency IMU sensors. Compared to the reference results obtained from the motion capture system, the estimated angle coefficient of determination (R2) is greater than 0.74, and the root mean square error and normalized root mean square error (NRMSE) are less than 7° and 9.87%, respectively. The knee joint showed the best estimation performance in terms of the NRMSE and R2 among the hip, knee, and ankle joints.

Download Full-text

Prediksi Permintaan Kargo pada Cargo Service Center Tangerang City Menggunakan Metode Gated Recurrent Unit

Repositor ◽

10.22219/repositor.v2i8.972 ◽

2020 ◽

Vol 2 (8) ◽

Author(s):

Rifky Ahmad Saputra

Keyword(s):

Neural Network ◽

Root Mean Square Error ◽

Mean Square Error ◽

Root Mean Square ◽

Recurrent Neural Network ◽

Mean Square ◽

Service Center ◽

Gated Recurrent Unit

Pada saat ini persaingan bisnis dalam bidang layanan kargo khususnya di Indonesia semakin ketat. Terdapat beberapa perusahaan layanan kargo di Indonesia, salah satunya yaitu Cargo Service Center Tangerang City. Untuk mengantisipasi persaingan bisnis tersebut, Cargo Service Center Tangerang City harus dapat menentukan strategi manajemen usaha, baik dalam jangka menengah maupun jangka panjang. Salah satunya hal yang dapat dilakukan yaitu prediksi permintaan kargo. Pada Cargo Service Center Tangerang City terdapat data transaksi kargo mulai dari Januari 2016 hingga Septermber 2019, oleh karena itu dilakukanlah penelitian yaitu mengimplementasikan metode Gated Recurrent Unit untuk melakukan prediksi permintaan kargo. metode Gated Recurrent Unit merupakan model pengembangan dari Recurrent Neural Network yang biasa digunakan untuk melakukan prediksi pada data sekuens. Pengujian model prediksi dalam penelitian ini dilakukan dengan mencari nilai Root Mean Square Error terkecil dari beberapa percobaan. Hasil dari penelitian ini menunjukkan bahwa model cukup baik dalam melakukan prediksi permintaan kargo, namun terdapat beberapa hasil prediksi metode Gated Recurrent Unit yang masih belum maksimal mendekati nilai aktual misalnya pada nilai aktual yang berada di titik puncak.

Download Full-text

Forecasting Reference Evapotranspiration Using Time Lagged Recurrent Neural Network

WSEAS TRANSACTIONS ON ENVIRONMENT AND DEVELOPMENT ◽

10.37394/232015.2020.16.72 ◽

2020 ◽

Vol 16 ◽

Keyword(s):

Neural Network ◽

Root Mean Square Error ◽

Mean Square Error ◽

Root Mean Square ◽

Network Model ◽

Recurrent Neural Network ◽

Reference Evapotranspiration ◽

Irrigation Scheduling ◽

Mean Square ◽

Time Lagged

The aim of this study is to employ a Time Lagged Recurrent Neural Network (TLRNN) model for forecasting near future reference evapotranspiration (ETo) values by using climate data taken from meteorological station located in Velestino, a village near the city of Volos, in Thessaly, centre of Greece. TLRNN is Multilayer Perceptron Neural Network (MLP-NN) with locally recurrent connections and short-term memory structures that can learn temporal variations from the dataset. The network topology is using input layer, hidden layer and a single output with the ETo values. The network model was trained using the back propagation through time algorithm. Performance evaluations of the network model done by comparing the Mean Bias Error (MBE), Root Mean Square Error (RMSE), Coefficient of Determination (R2) and Index of Agreement (IA). The evaluation of the results showed that the developed TLRNN model works properly and the forecasting ETo values approximate the FAO-56 PM values. A good proximity of predictions with the experimental data was noticed, achieving coefficients of determination (R2) greater than 75% and root mean square error (RMSE) values less than 1.0 mm/day. The forecasts range up to three days ahead and can be helpful to farmers for irrigation scheduling.

Download Full-text

IMPLEMENTASI METODE NEURAL NETWORK BACKPROPAGATION DALAM PREDIKSI INDEKS HARGA SAHAM GABUNGAN (IHSG)

SEMINASTIKA ◽

10.47002/seminastika.v3i1.252 ◽

2021 ◽

Vol 3 (1) ◽

pp. 79-85

Author(s):

Okky Barus ◽

Christopher Wijaya

Keyword(s):

Neural Network ◽

Root Mean Square Error ◽

Mean Square Error ◽

Root Mean Square ◽

Composite Index ◽

Mean Square

Pada era saat ini, Investasi saham di pasar modal merupakan aset yang sangat penting bagi beberapa golongan masyarakat dan juga bagi perusahaan. Dengan adanya investasi, secara langsung maupun tidak langsung dapat memberikan dampak bagi perusahaan maupun bagi masyarakat. Penelitian ini bertujuan untuk memprediksi Indeks Harga Saham Gabungan (IHSG) dengan indeks saham: Jakarta Composite Index (JKSE). Metode yang digunakan pada penelitian ini adalah Neural Network Backpropagation. Pengumpulan dataset melalui website finance.yahoo.com dengan periode 8 Mei 2018 sampai dengan 7 Mei 2021 sebanyak 757 data. Setelah melakukan proses pengolahan data, data yang tersisa adalah 724 data. Kemudian data akan dibagi menjadi 70% data training dan 30% data testing yang akan digunakan pada proses pengolahan data. Hasil pengujian menggunakan metode Neural Newtwork Backpropagation mendapatkan hasil terbaik menggunakan Kondisi ke-10 dengan nilai Root Mean Square Error (RMSE) senilai 0.010. Kemudian akan didapatkan hasil perbandingan antara harga Close aktual dengan harga Close prediksi dengan akurasi sebesar 63.06% yang dapat membantu dalam pengambilan keputusan para investor.

Download Full-text

Prediction of quality attributes and maturity classification of pear fruit using laser imaging and Artificial Neural Network

Food Research ◽

10.26656/fr.2017.5(s1).022 ◽

2021 ◽

Vol 5 (S1) ◽

pp. 144-151

Author(s):

S.E. Adebayo ◽

N. Hashim

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Optical Properties ◽

Root Mean Square Error ◽

Mean Square Error ◽

Root Mean Square ◽

Mean Square ◽

Classification Models ◽

Pear Fruit ◽

Laser Imaging

In this study, the application of laser imaging technique was utilized to predict the quality attributes (firmness and soluble solids content) of pear fruit and to classify the maturity stages of the fruit harvested at different days after full bloom (dafb). Laser imaging system emitting at visible and near infra-red region (532, 660, 785, 830 and 1060 nm) was deployed to capture the images of the fruit. Optical properties (absorption ma and reduced scattering ms ʹ coefficients) at individual and combined wavelengths of the laser images of the fruit were used in the prediction and classifications of the maturity stages. Artificial neural network (ANN) was employed in the building of both prediction and classification models. Root mean square error of calibration (RMSEC), root mean square error of crossvalidation (RMSECV), correlation coefficient (r) and bias were used to test the performance of the prediction models while sensitivity and specificity were used to evaluate the classification models. The results showed that there was a very strong correlation between the ma and ms ʹ with pear development. This study had shown that optical properties of pears with ANN as prediction and classification models can be employed to both predict quality parameters of pear and classify pear into different (dafb) non-destructively.

Download Full-text

ANALISA KOMPARASI NEURAL NETWORK BACKPROPAGATION DAN MULTIPLE LINEAR REGRESSION UNTUK PERAMALAN TINGKAT INFLASI

10.31227/osf.io/wp58d ◽

2019 ◽

Author(s):

Amrin Amrin

Keyword(s):

Neural Network ◽

Linear Regression ◽

Root Mean Square Error ◽

Multiple Linear Regression ◽

Mean Square Error ◽

Root Mean Square ◽

Mean Absolute Deviation ◽

Mean Square ◽

Absolute Deviation

Tingkat inflasi tidak dapat dianggap remeh dalam sistem perekonomian suatu negara dan pelaku bisnis pada umumnya. Jika inflasi dapat diramalkan dengan akurasi yang tinggi, tentunya dapat dijadikan dasar pengambilan kebijakan pemerintah dalam mengantisipasi aktivitas ekonomi di masa depan. Pada penelitian ini akan digunakan metode prediksi neural network backpropagation dan multiple linear regression untuk memprediksi tingkat inflasi bulanan di indonesia, selanjutnya membandingkan manakah yang terbaik dari kedua metode tersebut. Data inflasi yang digunakan bersumber dari Badan Pusat Statistik dari tahun 2006-2015, dimana 80% sebagai data training dan 20% sebagai data testing. Dari hasil analisis data yang dilakukan disimpulkan bahwa Performa model multiple linear regression lebih baik dibandingkan dengan metode neural network backpropagation dengan nilai mean absolute deviation (MAD) sebesar 0.0380, mean square error (MSE) sebesar 0.0023, dan nilai Root Mean Square Error (RMSE) sebesar 0.0481

Download Full-text

Artificial Neural Network Approach for Modelling of Mercury Ions Removal from Water Using Functionalized CNTs with Deep Eutectic Solvent

International Journal of Molecular Sciences ◽

10.3390/ijms20174206 ◽

2019 ◽

Vol 20 (17) ◽

pp. 4206 ◽

Cited By ~ 1

Author(s):

Seef Saadi Fiyadh ◽

Mohamed Khalid AlOmar ◽

Wan Zurina Binti Jaafar ◽

Mohammed Abdulhakim AlSaadi ◽

Sabah Saadi Fayaed ◽

...

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Root Mean Square Error ◽

Adsorption Capacity ◽

Mean Square Error ◽

Root Mean Square ◽

Deep Eutectic Solvent ◽

Mean Square ◽

Mercury Ions ◽

Narx Model

Multi-walled carbon nanotubes (CNTs) functionalized with a deep eutectic solvent (DES) were utilized to remove mercury ions from water. An artificial neural network (ANN) technique was used for modelling the functionalized CNTs adsorption capacity. The amount of adsorbent dosage, contact time, mercury ions concentration and pH were varied, and the effect of parameters on the functionalized CNT adsorption capacity is observed. The (NARX) network, (FFNN) network and layer recurrent (LR) neural network were used. The model performance was compared using different indicators, including the root mean square error (RMSE), relative root mean square error (RRMSE), mean absolute percentage error (MAPE), mean square error (MSE), correlation coefficient (R2) and relative error (RE). Three kinetic models were applied to the experimental and predicted data; the pseudo second-order model was the best at describing the data. The maximum RE, R2 and MSE were 9.79%, 0.9701 and 1.15 × 10−3, respectively, for the NARX model; 15.02%, 0.9304 and 2.2 × 10−3 for the LR model; and 16.4%, 0.9313 and 2.27 × 10−3 for the FFNN model. The NARX model accurately predicted the adsorption capacity with better performance than the FFNN and LR models.

Download Full-text

Prediction of Significant Wave Height Using Neural Network in the Java Sea (North of Surabaya)

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.862.72 ◽

2017 ◽

Vol 862 ◽

pp. 72-77

Author(s):

Wimala L. Dhanistha ◽

R.A. Atmoko ◽

P. Juniarko ◽

Ridho Akbar

Keyword(s):

Neural Network ◽

Neural Networks ◽

Root Mean Square Error ◽

Mean Square Error ◽

Root Mean Square ◽

Wave Height ◽

Mean Square ◽

The North ◽

Java Sea ◽

The City

Indonesia is an archipelago, Surabaya is the second crowded city in Indonesia. So the shipping lane and the city is comparable. Neural network is models inspired by biological neural networks and used to estimate or approximate functions that can depend on a large number of inputs and are generally unknown. Neural network is used to predict the wave height in Java Sea (The North of Surabaya). The Root Mean Square Error average for the next one hour is 0.03 and the Root Mean Square Error average for the next six hours is 0.09. That’s mean the longest the prediction, the biggest Root Mean Square error.

Download Full-text

O OPTIMALISASI NEURAL NETWORK BERBASIS PARTICLE SWARM OPTIMIZATION UNTUK MEMPREDIKSI LAMA PENYINARAN MATAHARI DALAM MEMENUHI KEBUTUHAN ENERGI

Jurnal Teknologi Informasi Indonesia (JTII) ◽

10.30869/jtii.v4i2.391 ◽

2019 ◽

Vol 4 (2) ◽

pp. 52-62

Author(s):

Wahyudin Hasyim ◽

Alter Lasarudin

Keyword(s):

Neural Network ◽

Particle Swarm Optimization ◽

Root Mean Square Error ◽

Mean Square Error ◽

Root Mean Square ◽

Particle Swarm ◽

Mean Square ◽

Swarm Optimization

Tingginya beban listrik yang mencapai 325 MegaWatt, hal ini merupakan perhatian penting bagi pemerintah Provinsi Gorontalo dalam kebutuhan energi listrik, maka perlu memprediksi lama penyinaran matarahari pada suatu daerah, Energi sel surya salah satunya bergantung pada lamanya penyinaran cahaya matahari. Diantaranya dengan melakukan perancangan model prediksi. Metode prediksi yang mimiliki nilai error terkecil adalah Neural Network, akan tetapi masih adanya kelemahan pada waktu pelatihan untuk mencapai konvergen dan overfitting. Maka perlu dilakukan optimalisasi pada bobot jaringan dengan menggunakan Particle Swarm Optimazition, yang merupakan salah satu metode terbaik dalam optimasi. Dengan penggunaan optimasi yang diukur melalui hasil peroleha Root Mean Square Error (RMSE). Hasil pengujian terhadap algoritma menunjukkan bahwa nilai RMSE mengunakan Neural Network 0,131, sedangkan dengan penerapan optimasi dengan particle swarm optimization hasil RMSE 0,127. Dengan penerapan metode optimasi terserbut dapat mengurangi nilai error

Download Full-text

Weather-Based Neural Network, Stepwise Linear and Sparse Regression Approach for Rabi Sorghum Yield Forecasting of Karnataka, India

Agronomy ◽

10.3390/agronomy10111645 ◽

2020 ◽

Vol 10 (11) ◽

pp. 1645

Author(s):

Shankarappa Sridhara ◽

Nandini Ramesh ◽

Pradeep Gopakkali ◽

Bappa Das ◽

Soumya Venkatappa ◽

...

Keyword(s):

Neural Network ◽

Root Mean Square Error ◽

Mean Square Error ◽

Root Mean Square ◽

Ridge Regression ◽

Principal Component ◽

Growth Period ◽

Weather Conditions ◽

Mean Square ◽

Yield Forecasting

Sorghum is an important dual-purpose crop of India grown for food and fodder. Prevailing weather conditions during the crop growth period determine the yield of sorghum. Hence, the crop yield forecasting models based on weather parameters will be an appropriate option for policymakers and researchers to develop sustainable cropping strategies. In the present study, six multivariate weather-based models viz., least absolute shrinkage and selection operator (LASSO), elastic net (ENET), principal component analysis (PCA) in combination with stepwise multiple linear regression (SMLR), artificial neural network (ANN) alone and in combination with PCA and ridge regression model are examined by fixing 90% of the data for calibration and remaining dataset for validation to forecast rabi sorghum yield for different districts of Karnataka. The R2 and root mean square error (RMSE) during calibration ranged between 0.42 to 0.98 and 30.48 to 304.17 kg ha−1, respectively, without actual evapotranspiration (AET) whereas, these evaluation parameters varied from 0.38 to 0.99 and 19.84 to 308.79 kg ha−1, respectively with AET inclusion. During validation, the RMSE and nRMSE (normalized root mean square error) varied between 88.99 to 1265.03 kg ha−1 and 4.49 to 96.84%, respectively without AET and including AET as one of the weather variable RMSE and nRMSE were 63.48 to 1172.01 kg ha−1 and 4.16 to 92.56%, respectively. The performance of six multivariate models revealed that LASSO was the best model followed by ENET compared to PCA_SMLR, ANN, PCA_ANN and ridge regression models because of reduced overfitting through penalisation of regression coefficient. Thus, it can be concluded that LASSO and ENET weather-based models can be effectively utilized for the district level forecast of sorghum yield.

Download Full-text

The Analysis of Atmospheric Water Vapor Based on Ground-Based Microwave Radiometer

Key Engineering Materials ◽

10.4028/www.scientific.net/kem.500.335 ◽

2012 ◽

Vol 500 ◽

pp. 335-340

Author(s):

Jie Ying He ◽

Feng Lin Sun ◽

Sheng Wei Zhang ◽

Yu Zhang

Keyword(s):

Neural Network ◽

Water Vapor ◽

Root Mean Square Error ◽

Mean Square Error ◽

Root Mean Square ◽

Microwave Radiometer ◽

Mean Square ◽

Vapor Density ◽

China Meteorological Administration ◽

Density Profiles

The paper introduces a widely used atmospheric absorption models: MPM by Liebe in 1989. Using this absorption model, the paper simulates the temperature and humidity weighting functions and brightness temperature according to the different frequencies and bandwidth of the multi-channel ground-based microwave radiometer. The results show that simulated brightness temperatures are very well agreement with the observation values with an acceptable root mean square error. This paper uses widely used retrieval method of artificial neural network to obtain the water vapor density profiles and calculates the root mean square error of each dataset. Also, to improve the accuracy of retrievals, this paper adopts multi-layers neural network which has two hidden layers. The results show that the retrievals of water vapor density profiles based on ground-based microwave radiometer are agreement with the water vapor density profile which is observed by radiosonde. Grant Nos. GYHY200906035 China Meteorological Administration nonprofit sector (meteorology) special research

Download Full-text