A5023517939- Data Mining Algorithms and Applications
- Recommender Systems and Techniques
- Diabetes Management and Education
- Complex Network Analysis Techniques
- Neuroscience of respiration and sleep
- Obstructive Sleep Apnea Research
- Computational Geometry and Mesh Generation
- Sleep and related disorders
- Blood Pressure and Hypertension Studies
- Chronic Disease Management Strategies
- Diabetes Management and Research
- Video Analysis and Summarization
- Geoscience and Mining Technology
- Hormonal Regulation and Hypertension
- Imbalanced Data Classification Techniques
- Diabetes Treatment and Management
- Iron and Steelmaking Processes
- Digital Platforms and Economics
- Advanced Algorithms and Applications
- Consumer Market Behavior and Pricing
- Rough Sets and Fuzzy Logic
- Sodium Intake and Health
- Cardiac Health and Mental Health
Alibaba Group (China)
2024-2025
Zhuhai People's Hospital
2021-2024
Zhejiang University
2020-2021
Shantou University
2016
Shantou University Medical College
2016
Click-Through Rate prediction aims to predict the ratio of clicks impressions a specific link, which is challenging due (1) extremely high-dimensional categorical features; (2) both important original features and their interactions; (3) reliance on different interactions in time periods. To overcome these difficulties, we propose new feature interaction detection method based idea frequent itemset mining, named Online Random Intersection Chains (ORIC), detects informative with high...
To investigate the engagement and health outcomes of community-based intervention for obstructive sleep apnea (OSA) in general population.
With the growing size of datasets, feature selection becomes increasingly important. Taking interactions original features into consideration will lead to extremely high dimension, especially when are discrete and one-hot encoding is applied. This makes it more worthwhile mining useful as well their interactions. Association rule aims extract interesting correlations between items, but difficult use rules a qualified classifier themselves. Drawing inspiration from association mining, we come...
Blast furnace is the main equipment in modern ironmaking industry. To achieve hot metal of high quality, temperature blast rector, often indicated by silicon content metal, should be controlled within a proper range. This brief captures multiscale features system and proposes predictor based on nonuniform delay-coordinate embedding for prediction. We use discrete binary particle swarm optimization method to optimize parameters, including dimension time lags. Then, trained with under...
The aim of this study is to evaluate the effect diabetes disease management program (DMP) on glycemic control in type 1 mellitus (T1DM) patients Shantou China. A sample 240 participants recruited from 3C subgroup was followed up DMP for 3 years. provided self-management education, individualized therapy plan, complications screening, and laboratory examination periodical according clinical practice guidelines. Primary outcomes were changes hemoglobin A1C (HbA1c). Two hundred one completed...
Interactions between several features sometimes play an important role in prediction tasks. But taking all the interactions into consideration will lead to extremely heavy computational burden. For categorical features, situation is more complicated since input be high-dimensional and sparse if one-hot encoding applied. Inspired by association rule mining, we propose a method that selects of called Random Intersection Chains. It uses random intersections detect frequent patterns, then most...
Abstract Objective Midlife in women is associated with an increase prevalence of hypertension. Little known on the risk factors new-onset hypertension among middle-aged women. Methods In this nested case-control study, 1,430 aged 40 to 60 years repeated physical examinations between 2009 and 2019 were recruited. Data included age, body mass index, blood pressure (BP), a series biomarkers. Participants divided into two samples: 388 cases episodic (ie, one normal BP at first visit abnormal...
With the growing size of data sets, feature selection becomes increasingly important. Taking interactions original features into consideration will lead to extremely high dimension, especially when are categorical and one-hot encoding is applied. This makes it more worthwhile mining useful as well their interactions. Association rule aims extract interesting correlations between items, but difficult use rules a qualified classifier themselves. Drawing inspiration from association mining, we...
Click-Through Rate prediction aims to predict the ratio of clicks impressions a specific link. This is challenging task since (1) there are usually categorical features, and inputs will be extremely high-dimensional if one-hot encoding applied, (2) not only original features but also their interactions important, (3) an effective may rely on different in time periods. To overcome these difficulties, we propose new interaction detection method, named Online Random Intersection Chains. The...
article: Effect of intervention on self-management level diabetic patients based Metabolic Disease Management Center target management - Minerva Surgery 2022 December;77(6):602-4 Medica Journals