A5087988017- Gene expression and cancer classification
- Advanced Multi-Objective Optimization Algorithms
- Metaheuristic Optimization Algorithms Research
- Bioinformatics and Genomic Networks
- Stock Market Forecasting Methods
- Financial Markets and Investment Strategies
- Machine Learning in Bioinformatics
- Evolutionary Algorithms and Applications
- Emotion and Mood Recognition
- Machine Learning and Data Classification
- Speech and Audio Processing
- Cell Image Analysis Techniques
- Explainable Artificial Intelligence (XAI)
- Anomaly Detection Techniques and Applications
- Single-cell and spatial transcriptomics
- Speech Recognition and Synthesis
- MicroRNA in disease regulation
- Big Data and Business Intelligence
- Complex Systems and Time Series Analysis
- Financial Reporting and Valuation Research
- Molecular Biology Techniques and Applications
- Solar and Space Plasma Dynamics
- Cancer-related molecular mechanisms research
- Circular RNAs in diseases
- Time Series Analysis and Forecasting
Baylor University
2021-2025
A.T. Still University
2024
Rogers (United States)
2022
Fordham University
2013-2021
Qinghai Normal University
2020-2021
Long Island University
2020
St. John's University
2016
Columbia University
2013-2014
Eastern Michigan University
2010-2011
University of Michigan
2011
The t-distributed stochastic neighbor embedding t-SNE is a new dimension reduction and visualization technique for high-dimensional data. rarely applied to human genetic data, even though it commonly used in other data-intensive biological fields, such as single-cell genomics. We explore the applicability of data make these observations: (i) similar previously techniques principal component analysis (PCA), able separate samples from different continents; (ii) unlike PCA, more robust with...
Although high-throughput microarray based molecular diagnostic technologies show a great promise in cancer diagnosis, it is still far from clinical application due to its low and instable sensitivities specificities pattern recognition. In fact, high-dimensional heterogeneous tumor profiles challenge current machine learning methodologies for small number of samples large or even huge variables (genes). This naturally calls the use an effective feature selection data classification. We...
MicroRNAs (miRNAs) are a class of important non-coding RNAs, which play roles in tumorigenesis and development by targeting oncogenes or tumor suppressor genes. One miRNA can regulate multiple genes, one gene be regulated miRNAs. To promote the clinical application miRNAs, two fundamental questions should answered: what's regulatory mechanism to gene, miRNAs for specific type cancer. In this study, we propose influence capturing (miRNAInf) decipher regulation relations on target genes...
Purpose Studies on mining text and generating intelligence human resource documents are rare. This research aims to use artificial machine learning techniques facilitate the employee selection process through latent semantic analysis (LSA), bidirectional encoder representations from transformers (BERT) support vector machines (SVM). The also compares performance of different learning, vectorization sampling approaches (HR) resume data. Design/methodology/approach LSA BERT used discover...
Compared with multi-objective optimization, solving many-objective optimization problems usually require more strong selection pressure to improve convergence. However, too leads the loss of diversity, and weak often results in slow How control balance convergence diversity remains a challenge optimization. To tackle this challenge, evolutionary algorithm based on hyper-dominance degree is proposed paper. A conception quantify solutions population, which solution higher indicates it has than...
Abstract Feature selection can be seen as a multi-objective task, where the goal is to select subset of features that exhibit minimal correlation among themselves while maximizing their with target label. Multi-objective particle swarm optimization algorithm (MOPSO) has been extensively utilized for feature and achieved good performance. However, most MOPSO-based methods are random lack knowledge guidance in initialization process, ignoring certain valuable prior information data, which may...
The utilization of both constrained and unconstrained-based optimization for solving multi-objective problems (CMOPs) has become prevalent among recently proposed multiobjective evolutionary algorithms (CMOEAs). However, the constrained-based which adopted by many CMOEAs typically gives priority to feasible solutions over infeasible ones regardless their objective values, potentially leading degraded performance due elimination promising with strong convergence diversity. Furthermore,...
Abstract Motivation Wuhan pneumonia is an acute infectious disease caused by the 2019 novel coronavirus (COVID-19). It being treated as a Class A though it was classified B according to Infectious Disease Prevention Act of China. Accurate estimation incubation period essential prevention and control. However, remains unclear about its exact believed that symptoms COVID-19 can appear in few 2 days or long 14 even more after exposure. The accurate calculation requires original...
The scheduling of disassembly lines is great importance to achieve optimized productivity. In this paper, we address the Hybrid Disassembly Line Balancing Problem that combines linear and U-shaped lines, considering multi-skilled workers, targeting profit carbon emissions. contrast common approaches in reinforcement learning typically employ weighting strategies solve multi-objective problems, our approach innovatively incorporates non-dominated ranking directly into reward function....