A5041698452- Advanced Graph Neural Networks
- Recommender Systems and Techniques
- Topic Modeling
- Reservoir Engineering and Simulation Methods
- Spectroscopy and Chemometric Analyses
- Radiomics and Machine Learning in Medical Imaging
- Oil and Gas Production Techniques
- Skin Protection and Aging
- Machine Fault Diagnosis Techniques
- melanin and skin pigmentation
China University of Petroleum, East China
2024
Taiyuan University of Science and Technology
2021-2022
Abstract Recommendation system is a technology that can mine user's preference for items. Explainable recommendation to produce recommendations target users and give reasons at the same time reveal recommendations. The explainability of improve transparency probability choosing recommended merits about are obvious, but it not enough focus solely on in field explainable Therefore, essential construct an framework items while maintaining accuracy diversity. An based knowledge graph...
Summary Electric submersible pump (ESP) is one of the common artificial lift technologies in offshore production systems. ESP failures are main cause decline efficiency oil wells. Early warning and diagnosis crucial to improve well efficiency. In this study, a hybrid model long short-term memory neural network convolutional (LSTM-CNN) for accurate early faults proposed, based on electrical data as basis analysis. Using hyper-parameters optimize LSTM structure highly fit field so that it can...
Nowadays, artificial intelligence promotes the rapid development of skin cancer detection technology, and federated model (FSDM) dual generative adversarial network (DGANM) solves fragmentation privacy data to a certain extent.To overcome problem that many-objective evolutionary algorithm (MaOEA) cannot guarantee convergence diversity population when solving above models, based on integrated strategy (MaOEA-IS) is proposed.First, idea learning introduced into mutation, new parents are...
In order to further obtain the best explanation paths and recommendation items that satisfy preferences of users accurately efficiently, a many-objective explainable model GrEA-ERS based on is proposed, which designs pruning strategies delete unsatisfied in knowledge graph users, so as reduce search paths. Combined with optimization algorithm GrEA, are used decision variables, at different explained correspond recommended items, accuracy, diversity, novelty explainability optimized their...