scholarly journals Novel Spatiotemporal Feature Extraction Parallel Deep Neural Network for Forecasting Confirmed Cases of Coronavirus Disease 2019

2020 ◽  
Author(s):  
Chiou-Jye Huang ◽  
Yamin Shen ◽  
Ping-Huan Kuo ◽  
Yung-Hsiang Chen
Keyword(s):  
Neural Network ◽  
Deep Neural Network ◽  
Absolute Error ◽  
Mean Absolute Percentage Error ◽  
The Other ◽  
Percentage Error ◽  
Model Assessment ◽  
Absolute Percentage Error ◽  
Spatiotemporal Feature ◽  
The Mean

AbstractThe coronavirus disease 2019 pandemic continues as of March 26 and spread to Europe on approximately February 24. A report from April 29 revealed 1.26 million confirmed cases and 125 928 deaths in Europe. This study proposed a novel deep neural network framework, COVID-19Net, which parallelly combines a convolutional neural network (CNN) and bidirectional gated recurrent units (GRUs). Three European countries with severe outbreaks were studied—Germany, Italy, and Spain—to extract spatiotemporal feature and predict the number of confirmed cases. The prediction results acquired from COVID-19Net were compared to those obtained using a CNN, GRU, and CNN-GRU. The mean absolute error, mean absolute percentage error, and root mean square error, which are commonly used model assessment indices, were used to compare the accuracy of the models. The results verified that COVID-19Net was notably more accurate than the other models. The mean absolute percentage error generated by COVID-19Net was 1.447 for Germany, 1.801 for Italy, and 2.828 for Spain, which were considerably lower than those of the other models. This indicated that the proposed framework can accurately predict the accumulated number of confirmed cases in the three countries and serve as a crucial reference for devising public health strategies.

A method for predicting the capacity of a lithium-ion battery based on deep neural network

2021 ◽  
Author(s):  
JamesChan
Keyword(s):  
Neural Network ◽  
Deep Neural Network ◽  
Lithium Ion ◽  
Chemical Conversion ◽  
Absolute Error ◽  
Production Costs ◽  
Percentage Error ◽  
Average Percentage ◽  
True Value ◽  
Battery Capacity

This paper proposes a solution to predict the capacity of the lithium-ion battery's capacity division process using deep learning methods. This solution extracts the physical observation records of part of the process steps from the chemical conversion and volumetric processes as features, and trains a Deep Neural Network (DNN) to achieve accurate prediction of battery capacity. According to the test, the average percentage absolute error (Mean Absolute Percentage Error, MAPE) of the battery capacity predicted by this model is only 0.78% compared with the true value. Combining this model with the production line can greatly reduce production time and energy consumption, and reduce battery production costs.

Prediksi Kedatangan Turis Menggunakan Algoritma Weighted Exponential Moving Average

2020 ◽  
Vol 12 (2) ◽  
pp. 129-132
Author(s):  
Sherly Florencia ◽  
Alethea Suryadibrata
Keyword(s):  
Moving Average ◽  
Mean Absolute Percentage Error ◽  
Percentage Error ◽  
Error Measurement ◽  
Absolute Percentage Error ◽  
Natural Beauty ◽  
Expenditure Data ◽  
The Mean ◽  
Tourism Expenditure ◽  
Using Data

Tourism is an important factor for the development of a country. Tourism can be used as a promotion to introduce natural beauty and cultural uniqueness. Government needs to predict how many tourists will come every year to do a planning. Therefore, an application is needed to help to predict the arrival of tourists in each country. In this paper, we use Weighted Exponential Moving Average (WEMA) method to predict the arrival of tourist, tourism expenditure in the country, and departure using data from 2008 to 2018. Error measurement is calculated using the Mean Absolute Percentage Error (MAPE). The result shows that the lowest average MAPE on arrival data with span 2 is at 3.28. The lowest average MAPE on tourism expenditure data with span 2 is at 3.99%. The result shows that the lowest average MAPE on departure data with span 2 is at 3.63%.

scholarly journals Estimating Water Reflectance at Near-Infrared Wavelengths for Turbid Water Atmospheric Correction: A Preliminary Study for GOCI-II

2020 ◽  
Vol 12 (22) ◽  
pp. 3791
Author(s):  
Jae-Hyun Ahn ◽  
Young-Je Park
Keyword(s):  
Near Infrared ◽  
Accurate Result ◽  
Atmospheric Correction ◽  
Mean Absolute Percentage Error ◽  
Percentage Error ◽  
Absolute Percentage Error ◽  
Turbid Waters ◽  
The Mean ◽  
Relationship Of ◽  
Spectral Relationship

Atmospheric correction is a fundamental process to remove the atmospheric effect from the top-of-atmosphere level. The atmospheric correction algorithm developed by the Korea Institute of Ocean Science and Technology employs a near-infrared (NIR) water reflectance model to deal with non-negligible NIR water reflectance over turbid waters. This paper describes the NIR water reflectance models using visible bands of the Second Geostationary Ocean Color Imager (GOCI-II). Whereas the previous GOCI uses the 660 nm band to estimate NIR water reflectance (SR660), GOCI-II uses additional 620 and 709 nm bands, which improves estimation of NIR water reflectance. We developed two reflectance models with the additional bands based on a spectral relationship of water reflectance (SR709) and a spectral relationship of inherent optical properties (SRIOP) from red to NIR wavelengths. A preliminary validation of these two reflectance models was performed using both simulations and an in situ dataset. The validation result showed that the mean absolute percentage error of the SR709 model compared with SR660 was reduced by approximately 6% and 10% at 745 and 865 nm, respectively. Moreover, the mean absolute percentage error of the SRIOP model compared with SR660 was reduced by approximately 12% and 16% at 745 and 865 nm, respectively. Note that SR709 produces the most accurate result when there is only one sediment type, and SRIOP shows the most accurate result when various sediment types exist. Users will be able to optionally select the appropriate NIR water reflectance models in the GOCI-II atmospheric correction process to enhance the accuracy of aerosol reflectance correction over turbid waters.

scholarly journals Fuzzy Inference System Mamdani dan the Mean Absolute Percentage Error (MAPE) untuk Prediksi Permintaan Dompet Pulsa pada XL Axiata Depok

2017 ◽  
Vol 2 (2) ◽  
pp. 97
Author(s):  
Mochammad Bagoes Satria Junianto
Keyword(s):  
Fuzzy Inference System ◽  
Fuzzy Inference ◽  
Mean Absolute Percentage Error ◽  
Percentage Error ◽  
Inference System ◽  
Absolute Percentage Error ◽  
Sample Data ◽  
The Mean

Kemajuan perkembangan teknologi informasi pada era globalisasi sekarang ini sangat pesat; hal ini menuntut setiap perusahaan untuk dapat saling bersaing dalam dunia bisnis yang dinamis dan penuh persaingan. Pada proses manjaemen permintaan dompet pulsa di XL Axiata cabang Depok memerlukan peramalan yang cukup matang agar dompet pulsa yang diminta kepada pusat tidak berlebihan atau tidak terlalu sedikit untuk menjaga kestabilan antara penjualan; persediaan dan jumlah permintaan. Untuk dapat melakukan peramalan yang lebih akurat; maka diperlukan suatu metode yang dapat menghitung ketidakpastian yang terjadi; dalam hal ini metode yang digunakan adalah dengan menggunakan Fuzzy inference system metode Mamdani untuk meramalkan jumlah permintaan dompet pulsa berdasarkan jumlah penjualan dan persediaan. Dengan 12 sample data untuk masing-masing sistem satuam yang digunakan hasil yang didapatkan yaitu dengan menggunakan Fuzzy inference system metode mamdani MAPE yang didapat sebesar 18;56% untuk Dompul XL 5k; 5;38% untuk Dompul XL 10k dan 14;2% untuk Dompul XL Rupiah.

scholarly journals PERBANDINGAN PERFORMA PREDIKSI TINGKAT KEMISKINAN ANTARA BACKPROPAGATION NEURAL NETWORK DAN GENERALIZED REGRESSION NEURAL NETWORK

2018 ◽  
Vol 6 (2) ◽  
pp. 89
Author(s):  
Rina Mamase ◽  
Ruli S. Sinukun
Keyword(s):  
Neural Network ◽  
Mean Absolute Percentage Error ◽  
Percentage Error ◽  
Generalized Regression Neural Network ◽  
Backpropagation Neural Network ◽  
Absolute Percentage Error ◽  
Generalized Regression

Menurunkan tingkat kemiskinan penduduk merupakan suatu program kerja Pemerintah Indonesia yang hingga saat ini masih berlangsung.  Pemberian bantuan secara merata, tepat dan cepat merupakan salah satu upaya pemerintah dalam menangani masalah kemiskinan. Upaya tersebut dapat diwujudkan dengan penyajian data kemiskinan secara cepat dan akurat melalui prediksi tingkat kemiskinan menggunakan suatu metode yang efektif. Kemiskinan adalah masalah multi dimensional, sehingga diperlukan kesepakatan pendekatan/metode  yang dipakai apabila ingin memprediksi tingkat kemiskinan. Masalah kemiskinan tidak hanya berasal dari ketidakmampuan dalam memenuhi kebutuhan dasar saja, melainkan ada juga faktor atau indikator lain yang dapat mempengaruhi tingkat kemiskinan penduduk disuatu daerah/wilayah, seperti indikator pertanian, perdagangan dan industri.  Selain penggunaan indikator kebutuhan dasar  seperti kependudukan, tenaga kerja, pendidikan, dan kesehatan, penelitian ini juga mencoba menambahkan indikator pertanian, industri, dan perdagangan dalam prediksi tingkat kemiskinan. Metode prediksi yang digunakan dalam penelitian ini adalah Backpropagation Neural Network (BPNN) dan Generalized Regression Neural Network (GRNN). Pengujian dilakukan dengan menggunakan data tingkat kemiskinan di Provinsi Gorontalo pada tahun 2016 dan 2017. Mean  Absolute Percentage Error (MAPE) digunakan sebagai kriteria evaluasi model prediksi. Hasil dari prediksi tingkat kemiskinan diperoleh bahwa metode GRNN memiliki performa 14-16% lebih baik jika dibandingkan dengan metode BPNN.

FRACTIONAL RESIDUAL PLOT FOR MODEL VALIDATION

2016 ◽  
Vol 79 (1) ◽  
Author(s):  
Nur Arina Bazilah Kamisan ◽  
Muhammad Hisyam Lee ◽  
Suhartono Suhartono ◽  
Abdul Ghapor Hussin ◽  
Yong Zulina Zubairi
Keyword(s):  
Model Validation ◽  
Goodness Of Fit ◽  
Mean Absolute Error ◽  
Pairwise Comparison ◽  
Absolute Error ◽  
Mean Absolute Percentage Error ◽  
Percentage Error ◽  
Absolute Percentage Error ◽  
Residual Plot

A pairwise comparison is important to measure the goodness-of-fit of models. Error measurements are used for this purpose but it only limit to the value, thus a graph is used to help show the precision of the models. These two should show a tally result in order to defense the hypothesis correctly. In this study, a fractional residual plot is proposed to help showing the precision of forecasts. This plot improvises the scale of the graph by changing the scale into decimal ranging from -1 to 1. The closer the point to 0 will indicate that forecast is robust and value closer to -1 or 1 will indicate that the forecast is poor. Two error measurements which are mean absolute error (MAE) and mean absolute percentage error (MAPE) and residual plot are used to justify the results and make comparison with the proposed fractional residual plot. Three difference data are used for this purpose and the results have shown that the fractional residual plot could give as much information as the residual plot but in an easier and meaningful way. In conclusion, the error plot is important in visualize the accurateness of the forecast.  

Modeling of Fixed Joints Stiffness Based on Modified Least-Square

2014 ◽  
Vol 536-537 ◽  
pp. 1365-1368
Author(s):  
Ming De Duan ◽  
Hao Liang Feng ◽  
Kang Hua Liu ◽  
Jun Yong Lu
Keyword(s):  
Experimental Data ◽  
Regression Analysis ◽  
Relative Error ◽  
Machine Tools ◽  
Mean Absolute Percentage Error ◽  
Least Square ◽  
Percentage Error ◽  
Regression Equations ◽  
Absolute Percentage Error ◽  
The Mean

According to experimental data, the model of fixed Joints stiffness in machine tools was built by least square of relative error. The new regression equations were obtained by regression analysis. Compared to the original equations with Gaussian least-square, the relative error of new regression equations is within 3.5%, which reduces by 12.5% and the mean absolute percentage error (MAPE) decreases by 18.0%, 12.4%and 19.0% respectively.

scholarly journals Peramalan Nilai Tukar Petani Subsektor Peternakan Menggunakan Fuzzy Time Series Lee

2021 ◽  
Vol 3 (1) ◽  
pp. 1-15
Author(s):  
Mahadi Muhammad ◽  
Sri Wahyuningsih ◽  
Meiliyani Siringoringo
Keyword(s):  
Time Series ◽  
Fuzzy Sets ◽  
Historical Data ◽  
Mean Absolute Percentage Error ◽  
Fuzzy Time Series ◽  
Percentage Error ◽  
Logical Relationship ◽  
Absolute Percentage Error ◽  
The Mean ◽  
Fuzzy Logical

ABSTRAKFuzzy time series (FTS) Lee adalah suatu metode peramalan yang digunakan ketika jumlah data historis yang tersedia sedikit, serta tidak mensyaratkan asumsi-asumsi tertentu yang harus terpenuhi. Metode ini menggunakan data historis berupa himpunan fuzzy yang berasal dari bilangan real atas himpunan semesta pada data aktual. FTS Lee adalah perkembangan dari FTS Song dan Chissom, FTS Cheng, serta FTS Chen. Pada penelitian ini dibahas penerapan FTS Lee pada data Nilai Tukar Petani Subsektor Peternakan (NTPT) di Kalimantan Timur. Tujuan penelitian ini adalah memperoleh hasil peramalan NTPT di Kalimantan Timur pada bulan Januari 2020 dengan menggunakan FTS Lee. Langkah awal dalam penelitian ini yaitu menentukan himpunan semesta pembicaraan, langkah kedua menentukan banyaknya himpunan fuzzy, langkah ketiga mendefinisikan derajat keanggotaan himpunan fuzzy terhadap  dan melakukan fuzzyfikasi pada data aktual, langkah keempat membuat fuzzy logical relationship, langkah kelima membuat fuzzy logical relationship group, langkah keenam melakukan defuzzyfikasi sehingga diperoleh hasil peramalan, serta dilanjutkan dengan menghitung nilai mean absolute percentage error. Hasil penelitian menunjukkan bahwa peramalan menggunakan FTS Lee pada bulan Januari 2020 adalah 110,25. Nilai mean absolute percentage error pada  hasil peramalan dengan menggunakan FTS Lee adalah sangat baik.  ABSTRACTLee’s Fuzzy time series (FTS) is a forecasting method that is used when the number of historical data that available was small and does not require certain assumptions to be fulfilled. This method uses historical data in the form of fuzzy sets derived from real numbers over the set of universes in the actual data. FTS Lee is a development of FTS Song and Chissom, FTS Cheng, and FTS Chen. This research discusses the application of FTS Lee to the Exchange Rate of Farmers Subsectors Farm (ERFSF) in Kalimantan Timur. The purpose of this study was to obtain the results of ERFSF forecasting in Kalimantan Timur in January 2020 using FTS Lee. The first step during research is to determine the set of speech universes, the second step is to determine the number of fuzzy sets, the third step is to define the degree of fuzzy association membership and fuzzification on the actual data, the fourth step is to create a fuzzy logical relationship, the fifth step is to create a fuzzy logical relationship group, the sixth step is to perform defuzzification in order to obtain forecasting results, and continue by calculating the mean absolute percentage error value. The results showed that forecasting using FTS Lee in January 2020 was 110,25. The mean absolute percentage error value in forecasting results using FTS Lee is very good.

Peramalan terhadap Forex dengan Metode ARIMA Studi Kasus GBP/USD

2019 ◽  
Vol 11 (1) ◽  
pp. 6-10
Author(s):  
Michael Saputra Suryono ◽  
Raymond Oetama
Keyword(s):  
Knowledge Discovery ◽  
Foreign Exchange ◽  
Arima Model ◽  
Mean Absolute Percentage Error ◽  
Percentage Error ◽  
R Software ◽  
Absolute Percentage Error ◽  
Knowledge Discovery In Database ◽  
The Mean

Forex or Foreign Exchange is trading a country's currency with another country's currency. The purpose of this study is basically to test the accuracy of ARIMA on the GBP/USD currency pair. In addition, this research is expected to provide the benefits of knowledge about forecasting using ARIMA. This study resulted in forecasting the GBP/USD currency pair within 1 month, per 6 months from January 2018 to June 2018 using the ARIMA method and R software. Data to be used are data taken from January 2013 to June 2018. For the the process will follow the process of the KDD (Knowledge Discovery in Database). The results obtained by the ARIMA model (3,2,1) as the best model to be applied for 1 month per 6 months on the GBP/USD currency pair because it has the lowest AIC value and the mean absolute percentage error is 3.16%.

scholarly journals Evaluation of Several Error Measures Applied to the Sales Forecast System of Chemicals Supply Enterprises

2020 ◽  
Vol 11 (4) ◽  
pp. 39
Author(s):  
Ma. del Rocío Castillo Estrada ◽  
Marco Edgar Gómez Camarillo ◽  
María Eva Sánchez Parraguirre ◽  
Marco Edgar Gómez Castillo ◽  
Efraín Meneses Juárez ◽  
...  
Keyword(s):  
Forecast Model ◽  
Absolute Error ◽  
Mean Absolute Percentage Error ◽  
Percentage Error ◽  
Error Evaluation ◽  
Sales Forecast ◽  
Absolute Percentage Error ◽  
Sales Prediction ◽  
Prediction Techniques ◽  
Evaluation Techniques

The objective of the industry in general, and of the chemical industry in particular, is to satisfy consumer demand for products and the best way to satisfy it is to forecast future sales and plan its operations.Considering that the choice of the best sales forecast model will largely depend on the accuracy of the selected indicator (Tofallis, 2015), in this work, seven techniques are compared, in order to select the most appropriate, for quantifying the error presented by the sales forecast models. These error evaluation techniques are: Mean Percentage Error (MPE), Mean Square Error (MSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Mean Absolute Scaled Error (MASE), Symmetric Mean Absolute Percentage Error (SMAPE) and Mean Absolute Arctangent Percentage Error (MAAPE). Forecasts for chemical product sales, to which error evaluation techniques are applied, are those obtained and reported by Castillo, et. al. (2016 & 2020).The error measuring techniques whose calculation yields adequate and convenient results, for the six prediction techniques handled in this article, as long as its interpretation is intuitive, are SMAPE and MAAPE. In this case, the most adequate technique to measure the error presented by the sales prediction system turned out to be SMAPE.

Sign in / Sign up

Export Citation Format

Share Document