-
41
MPFM-VC: A Voice Conversion Algorithm Based on Multi-Dimensional Perception Flow Matching
Published 2025-05-01Get full text
Article -
42
Deep Learning-Based Detection of Depression and Suicidal Tendencies in Social Media Data with Feature Selection
Published 2025-03-01“…This study proposes a machine learning-based framework that integrates pre-trained language models and advanced feature selection techniques to improve the detection of depression and suicidal tendencies from social media data. …”
Get full text
Article -
43
Diagnosis Method for High-Speed Train Axle Box Bearing Slight Faults Based on Improved SAE and Temperature-Vibration Fusion
Published 2025-04-01“…And the model is validated by the test data collected from the high-speed train rolling bearing test bench. ［Result & Conclusion］ Compared with single-source feature-driven fault diagnosis method, the fault diagnosis accuracy of the diagnosis method based on temperature-vibration feature fusion is higher, with an average diagnosis accuracy rate of over 99%. …”
Get full text
Article -
44
Relationship extraction between entities with long distance dependencies and noise based on semantic and syntactic features
Published 2025-05-01“…Specifically, we leverage contextual semantic features generated by the pre-trained BERT model alongside syntactic features derived from dependency syntax graphs, effectively utilizing the complementary strengths of both sources of information to enhance the model’s performance in long-distance dependency scenarios. …”
Get full text
Article -
45
FEATURES, FORMS AND DISTANCE LEARNING TECHNOLOGIES
Published 2022-05-01“…Introduction of distance learning in educational process considerably increases its efficiency. In article features of distance learning, a form of occupations are stated at distance learning, standard structures of System of on-line training and the Portal of System of on-line training are presented. …”
Get full text
Article -
46
-
47
-
48
Contrastive Feature Bin Loss for Monocular Depth Estimation
Published 2025-01-01“…Recently monocular depth estimation has achieved notable performance using encoder-decoder-based models. These models have utilized the Scale-Invariant Logarithmic (SILog) loss for effective training, leading to significant performance improvements. …”
Get full text
Article -
49
Communication network-based train control system
Published 2024-05-01“…The automatic configuration (inauguration) of train communication networks (TCN) is emphasized as a distinguishing feature of train control and management systems (TCMS). …”
Get full text
Article -
50
-
51
Passenger Demand-Oriented High-Speed Train Stop Planning with Service-Node Features Analysis
Published 2020-01-01“…The current approach to train stop planning in China is based primarily on passenger demand volume information and the preset high-speed railway station level. …”
Get full text
Article -
52
Features of teaching Russian sign language to employees of internal affairs bodies in the process of professional training
Published 2024-10-01“…To consider the features of teaching Russian sign language to employees of internal affairs bodies in the process of professional training.Materials and methods. …”
Get full text
Article -
53
-
54
Reflecting realities: gauging the pulse of simulator-based training in medical minds—resonance of simulator-based ultrasound training in medical education
Published 2025-05-01“…Abstract Background Simulator-based training (SBT) transforms medical education from traditional methods to technology-driven simulations for safer, complex scenario learning. …”
Get full text
Article -
55
-
56
Date recognition based on multi feature extraction
Published 2022-09-01“…Drug production date and expiration date are important indicators to measure the safety and effectiveness of drugs.The production date and validity period of liquid drugs in vertical bags are printed with numbers of 0～9.The recognition of the drug date of the hospital needs to meet the requirements of high speed and accuracy.The conventional template matching method and neural network recognition method have large amount of calculation and complex training.A drug date recognition method was proposed based on multi feature extraction.The combination of vertical line feature and three features for fine feature extraction of different digital characters has advantages of small amount of calculation and fast recognition speed.Compared with the recognition method of single extracted feature, it can effectively distinguish 10 different numbers, especially suitable for numbers of similar shape.It can provide patients with safe drug use guarantee, improve the management mode of the medical staff, improve the work efficiency, so as to improve the level of the pharmaceutical service of the hospital.…”
Get full text
Article -
57
-
58
-
59
-
60
