A study on limited pre-sale strategy with consideration of consumer regret.

The effect of regret on consumers' purchasing behavior is more and more obvious. The limited pre-sale can make retailers with limited production capacity allocate two periods of stock effectively and increase their income. This paper considers the heterogeneous consumers with regret behavior in the market and constructs a model to study the retailer's optimal limited pre-sale strategy. The results show that the high price regret sensitivity negatively affects the higher price of the products in the pre-sale strategy, while the out-of-stock regret sensitivity negatively affects the retailer's profit When the production capacity is relatively low, the proportion of rational consumers is large and the high price regret sensitivity coefficient is small, the retailer should pre-sell at a limited discount and the lowest valuation, and the highest valuation is on sale, otherwise, it should be sold at a price slightly lower than the highest valuation, but when the capacity is very sufficient, the sensitive coefficient of stock-out regret is small and the proportion of rational consumers is small, the retailer should pre-sell at an unlimited premium, and a price slightly lower than the highest valuation of the pre-sale, the lowest valuation of the sale, or should be pre-sold at the highest valuation.

A Deep Learning Method Approach for Sleep Stage Classification with EEG Spectrogram.

The classification of sleep stages is an important process. However, this process is time-consuming, subjective, and error-prone. Many automated classification methods use electroencephalogram (EEG) signals for classification. These methods do not classify well enough and perform poorly in the N1 due to unbalanced data. In this paper, we propose a sleep stage classification method using EEG spectrogram. We have designed a deep learning model called EEGSNet based on multi-layer convolutional neural networks (CNNs) to extract time and frequency features from the EEG spectrogram, and two-layer bi-directional long short-term memory networks (Bi-LSTMs) to learn the transition rules between features from adjacent epochs and to perform the classification of sleep stages. In addition, to improve the generalization ability of the model, we have used Gaussian error linear units (GELUs) as the activation function of CNN. The proposed method was evaluated by four public databases, the Sleep-EDFX-8, Sleep-EDFX-20, Sleep-EDFX-78, and SHHS. The accuracy of the method is 94.17%, 86.82%, 83.02% and 85.12%, respectively, for the four datasets, the MF1 is 87.78%, 81.57%, 77.26% and 78.54%, respectively, and the Kappa is 0.91, 0.82, 0.77 and 0.79, respectively. In addition, our proposed method achieved better classification results on N1, with an F1-score of 70.16%, 52.41%, 50.03% and 47.26% for the four datasets.