An Augmented Sample Selection Framework for Prediction of Anticancer Peptides.

Anticancer peptides (ACPs) have promising prospects for cancer treatment. Traditional ACP identification experiments have the limitations of low efficiency and high cost. In recent years, data-driven deep learning techniques have shown significant potential for ACP prediction. However, data-driven prediction models rely heavily on extensive training data. Furthermore, the current publicly accessible ACP dataset is limited in size, leading to inadequate model generalization. While data augmentation effectively expands dataset size, existing techniques for augmenting ACP data often generate noisy samples, adversely affecting prediction performance. Therefore, this paper proposes a novel augmented sample selection framework for the prediction of anticancer peptides (ACPs-ASSF). First, the prediction model is trained using raw data. Then, the augmented samples generated using the data augmentation technique are fed into the trained model to compute pseudo-labels and estimate the uncertainty of the model prediction. Finally, samples with low uncertainty, high confidence, and pseudo-labels consistent with the original labels are selected and incorporated into the training set to retrain the model. The evaluation results for the ACP240 and ACP740 datasets show that ACPs-ASSF achieved accuracy improvements of up to 5.41% and 5.68%, respectively, compared to the traditional data augmentation method.

Multi-Stream Convolution-Recurrent Neural Networks Based on Attention Mechanism Fusion for Speech Emotion Recognition.

The quality of feature extraction plays a significant role in the performance of speech emotion recognition. In order to extract discriminative, affect-salient features from speech signals and then improve the performance of speech emotion recognition, in this paper, a multi-stream convolution-recurrent neural network based on attention mechanism (MSCRNN-A) is proposed. Firstly, a multi-stream sub-branches full convolution network (MSFCN) based on AlexNet is presented to limit the loss of emotional information. In MSFCN, sub-branches are added behind each pooling layer to retain the features of different resolutions, different features from which are fused by adding. Secondly, the MSFCN and Bi-LSTM network are combined to form a hybrid network to extract speech emotion features for the purpose of supplying the temporal structure information of emotional features. Finally, a feature fusion model based on a multi-head attention mechanism is developed to achieve the best fusion features. The proposed method uses an attention mechanism to calculate the contribution degree of different network features, and thereafter realizes the adaptive fusion of different network features by weighting different network features. Aiming to restrain the gradient divergence of the network, different network features and fusion features are connected through shortcut connection to obtain fusion features for recognition. The experimental results on three conventional SER corpora, CASIA, EMODB, and SAVEE, show that our proposed method significantly improves the network recognition performance, with a recognition rate superior to most of the existing state-of-the-art methods.

Strong Generalized Speech Emotion Recognition Based on Effective Data Augmentation.

The absence of labeled samples limits the development of speech emotion recognition (SER). Data augmentation is an effective way to address sample sparsity. However, there is a lack of research on data augmentation algorithms in the field of SER. In this paper, the effectiveness of classical acoustic data augmentation methods in SER is analyzed, based on which a strong generalized speech emotion recognition model based on effective data augmentation is proposed. The model uses a multi-channel feature extractor consisting of multiple sub-networks to extract emotional representations. Different kinds of augmented data that can effectively improve SER performance are fed into the sub-networks, and the emotional representations are obtained by the weighted fusion of the output feature maps of each sub-network. And in order to make the model robust to unseen speakers, we employ adversarial training to generalize emotion representations. A discriminator is used to estimate the Wasserstein distance between the feature distributions of different speakers and to force the feature extractor to learn the speaker-invariant emotional representations by adversarial training. The simulation experimental results on the IEMOCAP corpus show that the performance of the proposed method is 2-9% ahead of the related SER algorithm, which proves the effectiveness of the proposed method.

Mapping the virtual water trade in water-scarce basin: an environmentally extended input-output analysis in the Yellow River Basin of China.

China's Yellow River, the nation's second-longest, grapples with severe water scarcity, impeding the high-quality development of its basin. Our study meticulously examines the intricate virtual water trade network inside and outside the basin, providing essential insights to combat its acute water scarcity. We calculated water consumption coefficients for seven pivotal sectors across diverse Chinese provinces, forming the foundational data for quantifying virtual water trade both inside and outside the basin. Utilizing the 2015 Multi-Regional Input-Output Table, we assessed the Yellow River Basin's reliance on external water resources. Despite enduring chronic water scarcity, the basin annually exports a substantial 27.2 billion m3 of virtual water, equivalent to half of its yearly runoff. This outflow predominantly flows to the economically advanced eastern coastal region, with Agriculture and Manufacturing sectors dominating. Significantly, an irrational industrial layout leads to a substantial transfer of virtual water from economically disadvantaged areas to more affluent regions, exacerbating water scarcity in the basin's less privileged areas. Our study yields critical insights for mitigating water shortages in the Yellow River Basin and provides a transferrable framework for regions worldwide grappling with analogous challenges.

Cross-Corpus Speech Emotion Recognition Based on Transfer Learning and Multi-Loss Dynamic Adjustment.

In this paper, we do research on cross-corpus speech emotion recognition (SER), in which the training and testing speech signals come from different speech corpus. The mismatched feature distribution between the training and testing sets makes many classical algorithms unable to achieve better results. To deal with this issue, a transfer learning and multi-loss dynamic adjustment (TLMLDA) algorithm is initiatively proposed in this paper. The proposed algorithm first builds a novel deep network model based on a deep auto-encoder and fully connected layers to improve the representation ability of features. Subsequently, global domain and subdomain adaptive algorithms are jointly adopted to implement features transfer. Finally, dynamic weighting factors are constructed to adjust the contribution of different loss functions to prevent optimization offset of model training, which effectively improve the generalization ability of the whole system. The results of simulation experiments on Berlin, eNTERFACE, and CASIA speech corpora show that the proposed algorithm can achieve excellent recognition results, and it is competitive with most of the state-of-the-art algorithms.

Improved semi-supervised autoencoder for deception detection.

Existing algorithms of speech-based deception detection are severely restricted by the lack of sufficient number of labelled data. However, a large amount of easily available unlabelled data has not been utilized in reality. To solve this problem, this paper proposes a semi-supervised additive noise autoencoder model for deception detection. This model updates and optimizes the semi-supervised autoencoder and it consists of two layers of encoder and decoder, and a classifier. Firstly, it changes the activation function of the hidden layer in network according to the characteristics of the deception speech. Secondly, in order to prevent over-fitting during training, the specific ratio dropout is added at each layer cautiously. Finally, we directly connected the supervised classification task in the output of encoder to make the network more concise and efficient. Using the feature set specified by the INTERSPEECH 2009 Emotion Challenge, the experimental results on Columbia-SRI-Colorado (CSC) corpus and our own deception corpus show that the proposed model can achieve more advanced performance than other alternative methods with only a small amount of labelled data.