A5051586546- Imbalanced Data Classification Techniques
- Anomaly Detection Techniques and Applications
- Machine Learning and Data Classification
- Infrastructure Maintenance and Monitoring
- Microbial Community Ecology and Physiology
- Remote Sensing and Land Use
- Tea Polyphenols and Effects
- Traditional Chinese Medicine Analysis
- Blind Source Separation Techniques
- Neuroscience and Neural Engineering
- Electricity Theft Detection Techniques
- Gut microbiota and health
- Artificial Intelligence in Healthcare
- Metabolomics and Mass Spectrometry Studies
- Machine Learning in Healthcare
- EEG and Brain-Computer Interfaces
Shanxi Agricultural University
2025
Dalian Maritime University
2022-2024
The grade of Daqu significantly influences the quality Baijiu. To address issues high subjectivity, substantial labor costs, and low detection efficiency in evaluation, this study focused on light-flavor proposed a two-layer classification structure model based computer vision machine learning. Target images were extracted using three image segmentation methods: threshold segmentation, morphological fusion, K-means clustering. Feature factors selected through methods including mean decrease...
Hongxin (HX) is an indispensable Daqu in the production of light-flavor Baijiu (LFB). However, classification method HX highly subjective, and functional differences microorganisms different grades are still unclear. In this study, metagenomics physiochemical analysis were used to compare three (top, first, second) clarify their brewing functions LFB. The results showed that a total 1556 genera 5367 species detected all samples. Bacteria fungi main HX, relative abundance bacteria above...
Diabetes has become one of the seven major diseases affecting human death, so early prediction disease to prevent it is critical. Several existing works literature, however, make predictions about diabetes with few considerations missing and imbalanced data proper. To overcome these problems, in this paper, we propose an efficient Probabilistic Ensemble classification algorithm for handling class Imbalance Missing values (PE_DIM) which can effectively handle issue imbalances improve...
Epilepsy, a chronic brain disease with recurrent attacks, is the most common neurological in humans. It extremely important for patients epilepsy to accurately and timely diagnose seizures. A new method based on Electroencephalogram (EEG) rhythm mixture features an ensemble classifier proposed seizure detection. The fixed frequency range empirical wavelet transform filter-bank EEG rhythms (EREWT) first introduced into sub-band decomposition of signals. Then, these sub-bands are reconstructed...