A hybrid forecasting model using LSTM and Prophet for energy consumption with decomposition of time series data

Arslan, Serdar

doi:10.7717/peerj-cs.1001

Cited by 28 publications

(17 citation statements)

References 48 publications

(51 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Here's where Bayesian optimization comes in provides a refreshing change from other tuning techniques. In Bayesian optimization, all previous estimates of the function f (x) are used instead of local gradients to obtain the most accurate estimates [23]. Bayesian optimization (BO) was used in this study to analyze the optimal parameters of the studied methods.…”

Section: B Bayesian Optimizationmentioning

confidence: 99%

Predicting Energy Consumption in Wastewater Treatment Plants through Light Gradient Boosting Machine: A Comparative Study

Alali

Harrou

Sun

2022

2022 10th International Conference on Systems and Control (ICSC)

View full text Add to dashboard Cite

Section: B Bayesian Optimizationmentioning

confidence: 99%

Predicting Energy Consumption in Wastewater Treatment Plants through Light Gradient Boosting Machine: A Comparative Study

Alali

Harrou

Sun

2022

2022 10th International Conference on Systems and Control (ICSC)

View full text Add to dashboard Cite

“…LSTM is widely used in time series forecasting (Lindemann et al 2021, Arslan 2022). However, a few studies in the literature applied LSTM in FTS.…”

Section: Introductionmentioning

confidence: 99%

Gated recurrent unit network-based fuzzy time series forecasting model

Arslan

2023

Afyon Kocatepe University Journal of Sciences and Engineering

View full text Add to dashboard Cite

Time series forecasting has lots of applications in various industries such as weather, business, retail and energy consumption forecasting. Accurate prediction in these applications is very important and also difficult task because of complexity and uncertainty of time series. Nowadays, using deep learning methods is a popular approach in time series forecasting and shows better performance than classical methods. However, in the literature, there are few studies which use deep learning methods in fuzzy time series (FTS) forecasting. In this study, we propose a novel FTS forecasting model which is based upon hybridization of Recurrent Neural Networks with FTS to deal with complexity and also uncertanity of these series. The proposed model utilizes Gated Recurrent Unit (GRU) to make prediction by using combination of membership values and also past value from original time series data as model input and produce real forecast value. Moreover, the proposed model can handle first order fuzzy relations as well as high order ones. In experiments, we have compared our model results with those of state-of-art methods by using two real world datasets; The Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX) and Nikkei Stock Average. The results indicate that our model outperforms or performs similar to other methods. The proposed model is also validated by using Covid-19 active case dataset and shows better performance than Long Short-term Memory (LSTM) networks.

show abstract

“…In [25], the authors propose a hybrid model consisting of Prophet providing seasonality information followed by a bidirectional LSTM for a deseasonalized version of the data. This data consists of monthly energy consumption values of 7 countries for ten and a half years.…”

Section: Introductionmentioning

confidence: 99%

Time Series Prediction: Comparative Study of ML Models in the Stock Market

Goverdhan

Khare

Manoov

2022

Preprint

View full text Add to dashboard Cite

Analyzing the Stock Market is a perpetual process and hard to grasp, especially for newcomers looking to invest in the market. This paper will be useful for novice investors to learn to invest in the stock market based on various factors that dictate prices. The paper’s target is to create a program that analyses previous stock data of companies. This also includes identifying factors affecting the share market. We generate the patterns from large data sets of data of the stock market and predict an approximate value of share price. The stock market can have a significant impact on individuals and the economy as a whole. Therefore, accurately predicting stock trends can minimize loss risk and maximize profit. The primary objective of this paper is to generate an approximate forecasting output and a general idea of future values by generating a pattern from historical data. The scope of this paper extends to model a suggestion tool based on ARIMA, XGBoost, and LSTM and give critical insights into their relative advantages and disadvantages, finally determining the best tool for forecasting the trend or even the future. The trend chart will adequately guide all prospective investors. We have analyzed data from the stock market collected from Yahoo Finance. The performance of the models was evaluated using metrics such as MSE, MAE, RMSE, and MAPE. During our trials, we found XGBoost to perform the best with an accuracy of 98.92%.

show abstract

A hybrid forecasting model using LSTM and Prophet for energy consumption with decomposition of time series data

Cited by 28 publications

References 48 publications

Predicting Energy Consumption in Wastewater Treatment Plants through Light Gradient Boosting Machine: A Comparative Study

Predicting Energy Consumption in Wastewater Treatment Plants through Light Gradient Boosting Machine: A Comparative Study

Gated recurrent unit network-based fuzzy time series forecasting model

Time Series Prediction: Comparative Study of ML Models in the Stock Market

Contact Info

Product

Resources

About