Poly-linear regression with augmented long short term memory neural network: Predicting time series data

Ahmed, Supriyo; Chakrabortty, Ripon K.; Essam, Daryl; Ding, Weiping

doi:10.1016/j.ins.2022.05.078

Cited by 19 publications

(3 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ahmed etc. [26] presented a novel DL approach for time series prediction using a combination of poly-linear regression with Long Short-Term Memory (LSTM) and data augmentation. It is consequently named Polylinear Regression with Augmented Long Short-Term Memory Neural Network (PLR-ALSTM-NN).…”

Section: Related Workmentioning

confidence: 99%

Research on Fresh Produce Sales Prediction Algorithm for Store Based on Multidimensional Time Series Data Analysis

Li,

Yu,

Zhu

et al. 2023

ICST Transactions on Scalable Information Systems

View full text Add to dashboard Cite

INTRODUCTION: Fresh produce is a daily necessity; however, offline stores often rely on personal experience for purchase, which is highly subjective and may result in inaccurate estimation of purchase quantities. This can lead to produce wastage and subsequently impact the profitability of business. This paper introduces a fresh produce sales prediction model, which can predict fresh produce sales by analyzing multidimensional time series data that influence sales. This model aims to provide guidance for fresh produce purchase in offline stores.OBJECTIVES: The purpose of this study is to predict fresh produce sales by analyzing multidimensional time series data that influence sales. This aims to provide a basis for fresh produce purchase in stores, reduce produce wastage, and enhance business profitability.METHODS: This paper proposes a fresh produce sales prediction model by analyzing multidimensional time series data that affect store sales of fresh produce. An essential component of this model is the ARIMA-LSTM combined prediction model. In this study, the weighted reciprocal of errors averaging method is selected as the weight determination method for the ARIMA-LSTM combined prediction model.RESULTS: In this paper, the ARIMA-LSTM combined model is used for prediction in two scenarios: when the single-model prediction accuracy is superior and when it is inferior. Experimental results indicate that in the case of lower accuracy in single-model prediction, the combined prediction model outperforms, improving prediction accuracy by 3.86% as measured by MAPE. Comparative experiments are conducted between the fresh produce sales prediction model proposed in this paper and time series prediction framework Prophet, traditional LSTM model, and ARIMA model. The experimental results indicate that the proposed model outperforms the others.CONCLUSION: The fresh produce sales prediction model proposed in this paper is based on multidimensional time series data to predict fresh produce sales in stores. This model can accurately predict the sales of fresh produce, providing purchase guidance for fresh produce stores, reducing fresh produce wastage caused by subjective purchasing factors, and increase business profits.

show abstract

Section: Related Workmentioning

confidence: 99%

Research on Fresh Produce Sales Prediction Algorithm for Store Based on Multidimensional Time Series Data Analysis

Li,

Yu,

Zhu

et al. 2023

ICST Transactions on Scalable Information Systems

View full text Add to dashboard Cite

show abstract

“…Machine learning methods, on the other hand, establish nonlinear relationships using models like the Support Vector Machine (SVM) or the more powerful Deep Neural Network (DNN) (Mana et al, 2023). DNN, particularly Recurrent Neural Networks (RNN) like Long Short-Term Memory (LSTM), have shown great success in learning feature representations and improving prediction accuracy for time series data (Ahmed et al, 2022).…”

Section: Introductionmentioning

confidence: 99%

Winds of Change: Enhancing Wind Power Generation Forecasting with LSTM Models and Advanced Techniques in a Single Turbine Wind Farm

Mana

2024

AIA

View full text Add to dashboard Cite

This article presents a comprehensive analysis of wind power generation using a Long Short-Term Memory (LSTM) model in the context of a wind farm with a single turbine. The novelty of the research lies in the integration of a technical model, predictive model, heat map analysis, and LSTM architecture to enhance forecasting accuracy. The performance of the forecasted model is evaluated using standard metrics, including Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). In the estimation phase, wind power generation is predicted based on current weather conditions and historical trends. Leveraging multivariate time series forecasting with LSTM through the Kera library, different look-back values are explored to optimise prediction accuracy. In particular, the study achieves favourable performance metrics, including MSE of 27.781, RMSE of 5.271, MAE of 3.281, and variance of 0.886 with a 70% training and 30% test split. Furthermore, a 60% training and 40% test split yield improved accuracy with MSE of 28.791, RMSE of 5.366, MAE of 3.376, and variance of 0.888. In the prediction phase, the LSTM model is employed to forecast power generation without relying on future weather information. Through various experiments, the optimal look-back period and number of neurones for the LSTM model are determined. Notably, the achieved mean absolute percent errors (MAPE) of 11.433% and 11.158% for 24 and 48 hours of data respectively showcase the impact of varying neuron counts on prediction accuracy. To further enhance predictive capabilities, optimisation techniques are implemented, involving adjustments to the LSTM model, including increased input batch size, batch normalisation, and dense layers. This optimisation leads to a substantial decrease in MAPE from 92.53% to 87.14% in a monthly prediction experiment. Furthermore, forward predictions are made for multiple days in the future, where LSTM successfully captures patterns and fluctuations, resulting in MAPEs of 70.74% for 24 hours, 39.26% for 2 days, and 51.48% for 1 week. Additionally, a comparison of two LSTM-based models, Autoencoder LSTM and FFT-Encoder-Decoder-LSTM, demonstrates the superior performance of the latter in terms of RMSE. This study emphasizes the potential of LSTM models in wind power generation prediction while introducing novel insights into optimizing model performance for accurate forecasting. The integration of technical and predictive models, along with heat map analysis and LSTM architecture, sets the groundwork for advancing wind power prediction methodologies.

show abstract

“…Intelligent fault diagnosis [2] refers to the application of machine learning methods such as support vector machine [3], logistic regression [4], linear regression [5], random forest [6], k-means [7], and other machine learning methods are applied to machine fault diagnosis [8]. In 2009, Duan et al [9] proposed a new cross-domain kernel learning method.…”

Section: Introductionmentioning

confidence: 99%

An unsupervised transfer learning bearing fault diagnosis method based on depthwise separable convolution

Li,

Yuan,

Wang

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

Bearings are an essential component of rotating mechanical equipment. Traditional signal processing-based fault diagnosis methods usually require a massive labeled data for training, but bearings generally operate in the equipment under normal fault-free conditions. This paper proposes an improved Adaptive Batch Normalization (AdaBN) transfer learning bearing fault diagnosis method for Batch Normalization (BN) in traditional deep learning architecture. The AdaBN network preprocesses the raw vibration signals, and then the preprocessed features are input to a depthwise sparable convolution neural model for training. Features are extracted by depthwise convolution and point convolution in the network. AdaBN can freeze all the parameters in the network except the BN layer. Finally, a small amount of labeled data is classified using transfer learning methods. A laboratory data set was used for validation, and the experimental validation showed that the accuracy of the bearing fault diagnosis method using AdaBN reached 85%.

show abstract

Poly-linear regression with augmented long short term memory neural network: Predicting time series data

Cited by 19 publications

References 34 publications

Research on Fresh Produce Sales Prediction Algorithm for Store Based on Multidimensional Time Series Data Analysis

Research on Fresh Produce Sales Prediction Algorithm for Store Based on Multidimensional Time Series Data Analysis

Winds of Change: Enhancing Wind Power Generation Forecasting with LSTM Models and Advanced Techniques in a Single Turbine Wind Farm

An unsupervised transfer learning bearing fault diagnosis method based on depthwise separable convolution

Contact Info

Product

Resources

About