Supply chain sales forecasting based on lightGBM and LSTM combination model

Weng, Tingyu; Liu, Wenyang; Xiao, Jun

doi:10.1108/imds-03-2019-0170

Cited by 55 publications

(37 citation statements)

References 14 publications

(20 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In contrast, leaf-wise growth strategy is more efficient because it only split the leaf that has the largest information gain on the same layer. Furthermore, considering this strategy may cause trees with high depth, resulting in overfitting, a maximum depth limitation is adopted during the growth of trees [39]. The level-wise and leaf-wise growth strategies are shown in Figure 5.…”

Section: Lightgbm Algorithmmentioning

confidence: 99%

Predicting Hard Rock Pillar Stability Using GBDT, XGBoost, and LightGBM Algorithms

et al. 2020

View full text Add to dashboard Cite

Predicting pillar stability is a vital task in hard rock mines as pillar instability can cause large-scale collapse hazards. However, it is challenging because the pillar stability is affected by many factors. With the accumulation of pillar stability cases, machine learning (ML) has shown great potential to predict pillar stability. This study aims to predict hard rock pillar stability using gradient boosting decision tree (GBDT), extreme gradient boosting (XGBoost), and light gradient boosting machine (LightGBM) algorithms. First, 236 cases with five indicators were collected from seven hard rock mines. Afterwards, the hyperparameters of each model were tuned using a five-fold cross validation (CV) approach. Based on the optimal hyperparameters configuration, prediction models were constructed using training set (70% of the data). Finally, the test set (30% of the data) was adopted to evaluate the performance of each model. The precision, recall, and F1 indexes were utilized to analyze prediction results of each level, and the accuracy and their macro average values were used to assess the overall prediction performance. Based on the sensitivity analysis of indicators, the relative importance of each indicator was obtained. In addition, the safety factor approach and other ML algorithms were adopted as comparisons. The results showed that GBDT, XGBoost, and LightGBM algorithms achieved a better comprehensive performance, and their prediction accuracies were 0.8310, 0.8310, and 0.8169, respectively. The average pillar stress and ratio of pillar width to pillar height had the most important influences on prediction results. The proposed methodology can provide a reliable reference for pillar design and stability risk management.

show abstract

Section: Lightgbm Algorithmmentioning

confidence: 99%

Predicting Hard Rock Pillar Stability Using GBDT, XGBoost, and LightGBM Algorithms

et al. 2020

View full text Add to dashboard Cite

show abstract

“…They applied a developed model to the sales data of the Coca-Cola company and showed its accuracy. Even though ARIMA has been widely used in a variety of supply chain areas, the linear characteristic of ARIMA makes it difficult to forecast real-world demand changes [3].…”

Section: Literature Reviewmentioning

confidence: 99%

“…Recently, development of the neural network (NN) allowed for effectively solving untouchable real-world problems such as self-driving cars, image recognition, the game of go, etc. Using NN is also expected to increase the accuracy of demand forecasting [3]. Thus,…”

Section: Introductionmentioning

confidence: 99%

Gated Recurrent Unit with Genetic Algorithm for Product Demand Forecasting in Supply Chain Management

et al. 2020

View full text Add to dashboard Cite

Product demand forecasting plays a vital role in supply chain management since it is directly related to the profit of the company. According to companies’ concerns regarding product demand forecasting, many researchers have developed various forecasting models in order to improve accuracy. We propose a hybrid forecasting model called GA-GRU, which combines Genetic Algorithm (GA) with Gated Recurrent Unit (GRU). Because many hyperparameters of GRU affect its performance, we utilize GA that finds five kinds of hyperparameters of GRU including window size, number of neurons in the hidden state, batch size, epoch size, and initial learning rate. To validate the effectiveness of GA-GRU, this paper includes three experiments: comparing GA-GRU with other forecasting models, k-fold cross-validation, and sensitive analysis of the GA parameters. During each experiment, we use root mean square error and mean absolute error for calculating the accuracy of the forecasting models. The result shows that GA-GRU obtains better percent deviations than other forecasting models, suggesting setting the mutation factor of 0.015 and the crossover probability of 0.70. In short, we observe that GA-GRU can optimally set five types of hyperparameters and obtain the highest forecasting accuracy.

show abstract

“…At the same time, some ensemble learning models have also been applied to the application and research of time series data forecasting [4][5]. In the research of many scholars, it is not difficult to find that the use of a single LSTM model to predict time series data shows better model prediction performance [7][8]. Although a single LSTM model has shown strong momentum in the prediction of nonlinear time series data, stock time series data has highly nonlinear characteristics, and there are still some shortcomings in predicting only a single LSTM model.…”

Section: Introductionmentioning

confidence: 99%

“…In this paper, the combined prediction method of LSTM and LightGBM is used [8]. Construct LSTM_LightGBM combined model.…”

Section: Introductionmentioning

confidence: 99%

Study on the application of LSTM-LightGBM Model in stock rise and fall prediction

Li²,

2021

MATEC Web Conf.

View full text Add to dashboard Cite

This paper proposes a hybrid financial time series forecast model based on LSTM and LightGBM, namely LSTM_LightGBM model. Use the LightGBM model to train the processed stock historical data set, and save the training results. Then the opening price, closing price, highest price, lowest price, trading volume and adjusted closing price are separately input into the LSTM model for prediction. The prediction result of each attribute is used as the test set of the prediction after LightGBM training. Constantly adjust the parameters of each model, and finally get the optimal stock price forecast model. The model is validated with the rise and fall of AAPL stock. Through the comparison of evaluation index root mean square error RMSE, mean absolute error MAE, prediction accuracy Accuracy and f1_score. It is found that the LSTM_LightGBM model exhibits stable and better prediction performance in the stock prediction. That is to say, the LSTM_LightGBM model proposed in this paper is stable and feasible in the stock price fluctuation forecast.

show abstract

Supply chain sales forecasting based on lightGBM and LSTM combination model

Cited by 55 publications

References 14 publications

Predicting Hard Rock Pillar Stability Using GBDT, XGBoost, and LightGBM Algorithms

Predicting Hard Rock Pillar Stability Using GBDT, XGBoost, and LightGBM Algorithms

Gated Recurrent Unit with Genetic Algorithm for Product Demand Forecasting in Supply Chain Management

Study on the application of LSTM-LightGBM Model in stock rise and fall prediction

Contact Info

Product

Resources

About