Graph Neural Network-Guided Contrastive Learning for Sequential Recommendation.

Sequential recommendation uses contrastive learning to randomly augment user sequences and alleviate the data sparsity problem. However, there is no guarantee that the augmented positive or negative views remain semantically similar. To address this issue, we propose graph neural network-guided contrastive learning for sequential recommendation (GC4SRec). The guided process employs graph neural networks to obtain user embeddings, an encoder to determine the importance score of each item, and various data augmentation methods to construct a contrast view based on the importance score. Experimental validation is conducted on three publicly available datasets, and the experimental results demonstrate that GC4SRec improves the hit rate and normalized discounted cumulative gain metrics by 1.4% and 1.7%, respectively. The model can enhance recommendation performance and mitigate the data sparsity problem.

Speech Recognition of Accented Mandarin Based on Improved Conformer.

The convolution module in Conformer is capable of providing translationally invariant convolution in time and space. This is often used in Mandarin recognition tasks to address the diversity of speech signals by treating the time-frequency maps of speech signals as images. However, convolutional networks are more effective in local feature modeling, while dialect recognition tasks require the extraction of a long sequence of contextual information features; therefore, the SE-Conformer-TCN is proposed in this paper. By embedding the squeeze-excitation block into the Conformer, the interdependence between the features of channels can be explicitly modeled to enhance the model's ability to select interrelated channels, thus increasing the weight of effective speech spectrogram features and decreasing the weight of ineffective or less effective feature maps. The multi-head self-attention and temporal convolutional network is built in parallel, in which the dilated causal convolutions module can cover the input time series by increasing the expansion factor and convolutional kernel to capture the location information implied between the sequences and enhance the model's access to location information. Experiments on four public datasets demonstrate that the proposed model has a higher performance for the recognition of Mandarin with an accent, and the sentence error rate is reduced by 2.1% compared to the Conformer, with only 4.9% character error rate.

Nomogram to Predict Overall and Cancer-Specific Survival in Patients with Synovial Sarcoma in the Extremities: A Population-Based Study.

Background: Synovial sarcoma is a rare disease, and synovial sarcoma that first appears in the extremities accounts for more than 80% of cases. We established two nomograms to predict the overall survival (OS) and cancer-specific survival (CSS) rates of patients with synovial sarcoma. Methods: A total of 227 patients diagnosed with synovial sarcoma in the extremities between 2010 and 2015 were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Univariate and multivariate Cox analyses were performed to explore independent prognostic factors and to create two separate nomograms for OS and CSS. The C-index, the area under the curve (AUC), calibration curve, decision curve analysis (DCA), and Kaplan-Meier (KM) curve were used to evaluate the column line graphs and analyze prognostic factors. Results: Age, Stage M, and surgery were identified as independent prognostic factors for OS and CSS. The ROC curve showed good discriminative power for the nomogram. Calibration curves and DCA curves showed that the nomogram had a satisfactory ability to predict OS and CSS. The KM curve showed that chemotherapy alone did not affect patient survival. Conclusion: Age, Stage M, and surgery are variables that affect OS and CSS in patients with synovial sarcoma in the extremities. Two nomograms were established based on the above variables to provide patients with more accurate individual survival predictions and to help physicians make appropriate clinical decisions.

Nomogram-Based Prediction of Overall and Cancer-Specific Survival in Patients with Primary Bone Diffuse Large B-Cell Lymphoma: A Population-Based Study.

Background: Primary bone diffuse large B-cell lymphoma (PD-DLBCL) accounts for more than 80% of primary bone lymphoma. We created two nomograms to predict overall survival (OS) and cancer-specific survival (CSS) in patients with PD-DLBCL for this rare disease. Methods: In total, 891 patients diagnosed with PB-DLBCL between 2007 and 2016 were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Univariate and multivariate Cox analyses were performed to explore independent prognostic factors and create nomograms for OS and CSS. The area under the curve (AUC), the calibration curve, decision curve analysis (DCA), and Kaplan-Meier (K-M) curve analysis were used to evaluate the nomograms. Results: Four variables were identified as independent prognostic factors for OS, and three variables were identified as independent prognostic factors for CSS. The receiver operating characteristic (ROC) curves demonstrated the strong discriminatory power of the nomograms. The calibration and DCA curves showed that the nomograms had a satisfactory ability to predict OS and CSS. The K-M curves showed that age, gender, primary site, chemotherapy, and tumor stage affected patient survival. Conclusions: In patients with PD-DLBCL, age, race, primary site, and chemotherapy affected OS, while age, race, and chemotherapy affected CSS. The two nomograms created based on the aforementioned variables provided more accurate individual survival predictions for PD-DLBCL patients and can help physicians make appropriate clinical decisions.