A5101493129- Recommender Systems and Techniques
- Hepatocellular Carcinoma Treatment and Prognosis
- Cholangiocarcinoma and Gallbladder Cancer Studies
- Robotic Path Planning Algorithms
- Reproductive System and Pregnancy
- Ectopic Pregnancy Diagnosis and Management
- Topic Modeling
- Reproductive Biology and Fertility
- Advanced Graph Neural Networks
- Caching and Content Delivery
- Advanced Computational Techniques and Applications
- Pharmacological Effects of Natural Compounds
- Gynecological conditions and treatments
- Software Engineering Research
- Expert finding and Q&A systems
- Ovarian function and disorders
- Robotics and Sensor-Based Localization
- Wnt/β-catenin signaling in development and cancer
- Image and Video Quality Assessment
- Engineering and Test Systems
- Software System Performance and Reliability
- Artificial Immune Systems Applications
- Endometrial and Cervical Cancer Treatments
- Guidance and Control Systems
- Mental Health via Writing
Sun Yat-sen University
2012-2024
Second Affiliated Hospital of Nanchang University
2024
Nanchang University
2024
Sixth Affiliated Hospital of Sun Yat-sen University
2019-2024
China Mobile (China)
2017-2023
Joint Global Change Research Institute
2023
Third Affiliated Hospital of Sun Yat-sen University
2012
Chinese Academy of Sciences
2008
Harbin Engineering University
2004
Harbin University
2004
Rationale: It has been emergingly recognized that apoptosis generates plenty of heterogeneous apoptotic vesicles (apoVs), which play a pivotal role in the maintenance organ and tissue homeostasis.However, it is unknown whether apoVs influence postnatal ovarian folliculogenesis.Methods: Apoptotic pathway deficient mice including Fas mutant (Fas mut ) ligand (FasL were used with apoV replenishment to evaluate biological function during folliculogenesis.Ovarian was characterized by...
This paper presents a cooperative path planning algorithm using improved Rapidly-exploring Random Trees (RRTs) to generate paths for multiple unmanned air vehicles (UAVs) in real time, from given starting locations goal the presence of unknown pop-up obstacles. Generating no conflicting obstacle environments group DAVs within short time is challenging task. Firstly, we propose an RRT by taking maneuvering constraints into account and simple efficient pruning method designed delete redundant...
This paper is concerned with TV recommendation, where one major challenge the coupling behavior issue that behaviors of multiple users are coupled together and not directly distinguishable because share same account. Unable to identify current watching user use could lead sub-optimal recommendation results due noise introduced by other users. Most existing methods deal this either unsupervised clustering algorithms or depending on latent representation learning strong assumptions. However,...
Calibration in recommender systems ensures that the user's interests distribution over groups of items is reflected with their corresponding proportions recommendation, which has gained increasing attention recently. For example, a user who watched 80 entertainment videos and 20 knowledge expected to receive recommendations comprising about 80% 20% as well. However, calls for responsible it become inadequate just match users' historical behaviors especially when are grouped by qualities,...
Item representation learning (IRL) plays an essential role in recommender systems, especially for sequential recommendation. Traditional recommendation models usually utilize ID embeddings to represent items, which are not shared across different domains and lack the transferable ability. Recent studies use pre-trained language (PLM) item text (text-based IRL) that universally applicable domains. However, existing text-based IRL is unaware of important collaborative filtering (CF)...
Particle swarm optimization is an tool which inspired by the birds social behavior. To solve premature problem of basic particle (bPSO), a novel modified based on particles equilibriums distribution (PSOED) was proposed. In order to effectively avoiding clustering within sub-area scope, we proposed new parameter can measure equilibrium degree. It introduced guarantee escaping from sub-optimum trap. addition, flying strategy presented direct rational Finally, four function optimizations are...
Background: Laparoscopic hepatectomy has been widely accepted owing to its advantages as a minimally invasive surgery; however, laparoscopic segment 7 (S7) (LSH) rarely reported. We aimed explore the safety and feasibility of LSH by comparing it with open surgical approaches. Methods: Twenty-nine patients who underwent S7 between January 2016 2023 were enrolled in this study. The patients’ characteristics, intraoperative details, postoperative outcomes compared 2 groups. Results: No...
Laparoscopic hepatectomy has minimally invasive advantages, but reports on laparoscopic right anterior sectionectomy (LRAS) are rare. Herein, we try to explore the benefits and drawbacks of LRAS by comparing it with open (ORAS). Between January 2015 September 2023, 39 patients who underwent (n = 18) or ORAS 21) were enrolled in study. The patients' characteristics, intraoperative details, postoperative outcomes compared between two groups. No significant differences preoperative data...
Abstract Study question Do patients with symptomatic cesarean scar defect (CSD) have a higher risk of chronic endometritis (CE) and worse early pregnancy outcomes after hysteroscopic reconstruction for CSD? Summary answer Symptomatic CSD has positive correlation CE. Antibiotic administration is promising option CE to obtain better outcomes. What known already Cesarean (CSD), niche-shaped myometrium defect, can be noticed sections (CS). Studies about the impact on fertility are limited. The...
Item representation learning (IRL) plays an essential role in recommender systems, especially for sequential recommendation. Traditional recommendation models usually utilize ID embeddings to represent items, which are not shared across different domains and lack the transferable ability. Recent studies use pre-trained language (PLM) item text (text-based IRL) that universally applicable domains. However, existing text-based IRL is unaware of important collaborative filtering (CF)...
A UAV formation is the essential air combat unit which consists of two or three UAVs. An autonomous decision making algorithm in an adversarial environment based on Plan Goal Graph (PGG) proposed. Firstly, a tactics knowledge structure PGG designed, and intelligent experience engine process Secondly, tactical method using described to work out sequences optimal maneuvering action at discrete time moment. Finally, simulation experiments different scenarios are realized, results show that...
The purpose of this paper is to use machine learning technology build more user profiles in the process customer segmentation assist marketing decision-making. Specifically, by introducing cross features into construction Feature Engineering, accuracy target group prediction can be further improved and subgroup after feature crossing mined. In addition, also proposes an decision tree model, which comprehensive profiles, then intelligent marketing. Firstly, both methods need preprocess data....
With the rapid development of smart TV, TV recommendation is attracting more and users. users usually distribute in multiple regions with different cultures hence have diverse program preferences. From perspective engineering practice performance improvement, it's very essential to model from one single model. In previous work, Multi-gate Mixture-of-Expert (MMoE) has been widely adopted multi-task multi-domain scenarios. practice, however, we first observe embeddings generated by experts...