Short-Term Building Load Forecast Based on Patch Learning with Long Short-Term Memory Network and Support Vector Regression

2021 ◽  
pp. 742-756
Author(s):  
Zhaohui Dan ◽  
Bo Wang ◽  
Huijin Fan ◽  
Lei Liu
Keyword(s):  
Support Vector Regression ◽  
Short Term Memory ◽  
Support Vector ◽  
Short Term ◽  
Term Memory ◽  
Load Forecast ◽  
Memory Network ◽  
Building Load ◽  
Long Short Term Memory

scholarly journals Comparative Study of Forecasting Models for Forex Predictions with the Impact of Different Currencies

2022 ◽  
Vol 12 (1) ◽  
pp. 11-22
Author(s):  
Satria Wiro Agung ◽  
◽  
Kelvin Supranata Wangkasa Rianto ◽  
Antoni Wibowo
Keyword(s):  
Machine Learning ◽  
Exchange Rates ◽  
Support Vector Regression ◽  
Short Term Memory ◽  
Political Stability ◽  
Support Vector ◽  
Short Term ◽  
Term Memory ◽  
Long Short Term Memory ◽  
Gated Recurrent Unit

- Foreign Exchange (Forex) is the exchange / trading of currencies from different countries with the aim of making profit. Exchange rates on Forex markets are always changing and it is hard to predict. Many factors affect exchange rates of certain currency pairs like inflation rates, interest rates, government debt, term of trade, political stability of certain countries, recession and many more. Uncertainty in Forex prediction can be reduced with the help of technology by using machine learning. There are many machine learning methods that can be used when predicting Forex. The methods used in this paper are Long Short Term Memory (LSTM), Gated Recurrent Unit (GRU), Support Vector Regression (SVR). XGBOOST, and ARIMA. The outcome of this paper will be comparison results that show how other major currency pairs have influenced the performance and accuracy of different methods. From the results, it was proven that XGBoost outperformed other models by 0.36% compared to ARIMA model, 4.4% compared to GRU model, 8% compared to LSTM model, 9.74% compared to SVR model. Keywords— Forex Forecasting, Long Short Term Memory, Gated Recurrent Unit, Support Vector Regression, ARIMA, Extreme Gradient Boosting

scholarly journals Short-Term Load Forecasting of Microgrid via Hybrid Support Vector Regression and Long Short-Term Memory Algorithms

2020 ◽  
Vol 12 (17) ◽  
pp. 7076 ◽  
Author(s):  
Arash Moradzadeh ◽  
Sahar Zakeri ◽  
Maryam Shoaran ◽  
Behnam Mohammadi-Ivatloo ◽  
Fazel Mohammadi
Keyword(s):  
Support Vector Regression ◽  
Performance Metrics ◽  
Short Term Memory ◽  
Load Forecasting ◽  
Support Vector ◽  
Short Term ◽  
Term Memory ◽  
Short Term Load Forecasting ◽  
Long Short Term Memory ◽  
Input Variables

Short-Term Load Forecasting (STLF) is the most appropriate type of forecasting for both electricity consumers and generators. In this paper, STLF in a Microgrid (MG) is performed via the hybrid applications of machine learning. The proposed model is a modified Support Vector Regression (SVR) and Long Short-Term Memory (LSTM) called SVR-LSTM. In order to forecast the load, the proposed method is applied to the data related to a rural MG in Africa. Factors influencing the MG load, such as various household types and commercial entities, are selected as input variables and load profiles as target variables. Identifying the behavioral patterns of input variables as well as modeling their behavior in short-term periods of time are the major capabilities of the hybrid SVR-LSTM model. To present the efficiency of the suggested method, the conventional SVR and LSTM models are also applied to the used data. The results of the load forecasts by each network are evaluated using various statistical performance metrics. The obtained results show that the SVR-LSTM model with the highest correlation coefficient, i.e., 0.9901, is able to provide better results than SVR and LSTM, which have the values of 0.9770 and 0.9809, respectively. Finally, the results are compared with the results of other studies in this field, which continued to emphasize the superiority of the SVR-LSTM model.

scholarly journals Perbandingan Algoritma Long Short-Term Memory dengan SVR Pada Prediksi Harga Saham di Indonesia

Petir ◽  
2020 ◽  
Vol 13 (1) ◽  
pp. 33-43
Author(s):  
Adhib Arfan ◽  
Lussiana ETP
Keyword(s):  
Support Vector Regression ◽  
Short Term Memory ◽  
Data Preprocessing ◽  
Support Vector ◽  
Short Term ◽  
Term Memory ◽  
Long Short Term Memory

Banyak investor masih ragu dengan risiko dalam berinvestasi, hal ini disebabkan oleh fluktuasi indeks harga saham dalam waktu singkat. Telah banyak dikembangkan metode untuk memperkirakan harga saham yang akan datang namun masih memiliki keterbatasan di antaranya adalah ketergantungan jangka panjang. Tujuan penelitian yang ingin dicapai adalah menghasilkan model peramalan harga saham yang lebih efektif dan memberikan hasil yang akurat. Tahapan yang dilakukan terdiri dari pengumpulan data, preprocessing data, pembagian data, perancangan LSTM, pelatihan LSTM dan melakukan pengujian. Berdasarkan hasil pengujian, LTSM mampu memprediksi harga saham pada tahun 2017-2019 dengan performa yang baik dan tingkat kesalahan yang relatif kecil. Sedangkan pengujian menggunakan metode Support Vector Regression (SVR), LSTM memiliki nilai loss lebih baik dari algoritma SRV. Rentang data pada LSTM mempengaruhi waktu latih yang digunakan, semakin besar rentang data maka semakin lama waktu latih yang digunakan. Rentang data pada SVR mempengaruhi nilai loss, semakin besar rentang data maka semakin besar nilai loss yang dihasilkan. Dengan demikian dapat disimpulkan bahwa LSTM mampu menanggulangi ketergantungan jangka panjang dan mampu memprediksi harga saham dengan hasil yang akurat.

scholarly journals Stock Prediction using Long Short-Term Memory, Support Vector Regression and Linear Regression Algorithms

2021 ◽  
Vol 9 (VII) ◽  
pp. 3632-3638
Author(s):  
Mr. V. Manoj Kumar
Keyword(s):  
Linear Regression ◽  
Support Vector Regression ◽  
Short Term Memory ◽  
Support Vector ◽  
Stock Prediction ◽  
Short Term ◽  
Data Set ◽  
Term Memory ◽  
Long Short Term Memory ◽  
Memory Support

Prediction is most important for stock market not only for traders but also for computer engineers who analyses stock data. We can perform this prediction by two ways one is using historical stock data and other by analyzing by information gathered from social media. It is based on model/pattern used to predict stock dataset, there are so many models are available for predicting stocks, simply model is algorithm that’s from machine learning and deep learning. In the data set the two main parameters open and close value are used for stock prediction mostly but we can also predict by its volume too. So that data is preprocessed before it is used for prediction. In this paper we used various algorithm like linear regression, support vector regression and long short-term memory for better accuracy and to compare how it different from other algorithm and for predicting future stock.

scholarly journals Comparison of Machine Learning Methods for Image Reconstruction Using the LSTM Classifier in Industrial Electrical Tomography

Energies ◽  
2021 ◽  
Vol 14 (21) ◽  
pp. 7269
Author(s):  
Grzegorz Kłosowski ◽  
Tomasz Rymarczyk ◽  
Konrad Niderla ◽  
Magdalena Rzemieniak ◽  
Artur Dmowski ◽  
...  
Keyword(s):  
Machine Learning ◽  
Support Vector Machine ◽  
Short Term Memory ◽  
Support Vector ◽  
Short Term ◽  
Electrical Tomography ◽  
Term Memory ◽  
Tomographic Images ◽  
Memory Network ◽  
Long Short Term Memory

Electrical tomography is a non-invasive method of monitoring the interior of objects, which is used in various industries. In particular, it is possible to monitor industrial processes inside reactors and tanks using tomography. Tomography enables real-time observation of crystals or gas bubbles growing in a liquid. However, obtaining high-resolution tomographic images is problematic because it involves solving the so-called ill-posed inverse problem. Noisy input data cause problems, too. Therefore, the use of appropriate hardware solutions to eliminate this phenomenon is necessary. An important cause of obtaining accurate tomographic images may also be the incorrect selection of algorithmic methods used to convert the measurements into the output images. In a dynamically changing environment of a tank reactor, selecting the optimal algorithmic method used to create a tomographic image becomes an optimization problem. This article presents the machine learning method’s original concept of intelligent selection depending on the reconstructed case. The long short-term memory network was used to classify the methods to choose one of the five homogenous methods—elastic net, linear regression with the least-squares learner, linear regression with support vector machine learner, support vector machine model, or artificial neural networks. In the presented research, tomographic images of selected measurement cases, reconstructed using five methods, were compared. Then, the selection methods’ accuracy was verified thanks to the long short-term memory network used as a classifier. The results proved that the new concept of long short-term memory classification ensures better tomographic reconstructions efficiency than imaging all measurement cases with single homogeneous methods.

Sign in / Sign up

Export Citation Format

Share Document