A5114549706- Face and Expression Recognition
- Metaheuristic Optimization Algorithms Research
- Gene expression and cancer classification
- Evolutionary Algorithms and Applications
- Machine Learning in Bioinformatics
- Advanced Image Processing Techniques
- Financial Distress and Bankruptcy Prediction
- Imbalanced Data Classification Techniques
- Voice and Speech Disorders
- Machine Learning and ELM
- Image Retrieval and Classification Techniques
- Vehicle License Plate Recognition
- Artificial Immune Systems Applications
- Advanced Multi-Objective Optimization Algorithms
- Fuzzy Logic and Control Systems
- Advanced Algorithms and Applications
- Advanced Statistical Modeling Techniques
- Advanced Data Compression Techniques
- Video Coding and Compression Technologies
Jilin University
2011-2018
Jilin Province Science and Technology Department
2018
Jilin Medical University
2013
Feature selection enhances classification accuracy by removing irrelevant and redundant feature. plays an important role in data mining pattern recognition. In this paper, we propose a hybrid feature subset algorithm called the maximum Pearson distance improved whale optimization (MPMDIWOA). First, based on Pearson's correlation coefficient distance, filter is proposed named (MPMD). Two parameters are MPMD to adjust weights of relevance redundancy. Second, modified can act as wrapper...
It is of great clinical significance to establish an accurate intelligent model diagnose the somatization disorder community correctional personnel. In this study, a novel machine learning framework proposed predict severity in correction The core adopt improved bacterial foraging optimization (IBFO) optimize two key parameters (penalty coefficient and kernel width) extreme (KELM) build IBFO-based KELM (IBFO-KELM) for diagnosis patients. main innovation point IBFO-KELM introduction...
Microarray data are highly redundant and noisy, most genes believed to be uninformative with respect studied classes, as only a fraction of may present distinct profiles for different classes samples. This paper proposed novel hybrid framework (NHF) the classification high dimensional microarray data, which combined information gain(IG), F-score, genetic algorithm(GA), particle swarm optimization(PSO) support vector machines(SVM). In order identify subset informative embedded out large...
Abstract Retraction: Zheng Y, Li Wang G, et al. A hybrid feature selection algorithm for microarray data. Concurrency Computat Pract Exper. 2018;e4716. https://doi.org/10.1002/cpe.4716 . The above article, published online on 25th October 2018 in Wiley Online Library (wileyonlinelibrary.com), has been retracted by agreement between the authors, journal Editor Chief Professor Geoffrey Fox, and John Sons Ltd. retraction agreed as an alternative was used than that described manuscript resulting...
In this paper, an ant colony optimization based method (AM) is proposed for gene selection. AM consists of two stages. the first stage, some redundant genes are filtered by information gain (IG). second a fuzzy adaptive applied to We evaluate performance on five expression datasets, which have dimensions varying from 7129 12000. also compare with results obtained four existing well-known algorithms. The comparison details show that could get better classification accuracy.
The video coding standard of a new generation, high efficiency ( HEVC ), is JCT-VT under planning, mainly orienting toward definition television （HDTV）and system. From the start basic structure HEVC, this paper not only introduces comprehensively key technologies in intra-frame and inter-frame predictive estimation, orthogonal transformation, filter compensation entropy but also points out hot issues latest research direction.