A5101501429- Data Quality and Management
- Educational Technology and Pedagogy
- Educational Technology and Assessment
- Traditional Chinese Medicine Studies
- Natural Language Processing Techniques
- E-commerce and Technology Innovations
- Construction Project Management and Performance
- Big Data Technologies and Applications
- Topic Modeling
- Stock Market Forecasting Methods
- Machine Learning and Algorithms
- Software Engineering Techniques and Practices
- Advanced Text Analysis Techniques
- Data Mining Algorithms and Applications
- Multi-Criteria Decision Making
- Advanced Decision-Making Techniques
- Face and Expression Recognition
- AI and Big Data Applications
- Biomedical Text Mining and Ontologies
- Resource-Constrained Project Scheduling
- Rough Sets and Fuzzy Logic
- Technology and Security Systems
- Time Series Analysis and Forecasting
- graph theory and CDMA systems
- Water Quality Monitoring and Analysis
China Southern Power Grid (China)
2021
Alibaba Group (United States)
2021
Yantai University
2021
Changsha Normal University
2019
Hunan Normal University
2019
Kunming University of Science and Technology
2012-2014
Soochow University
2009
Named entity recognition is a fundamental task in natural language processing and many studies have done about it recent decades. Previous word representation methods represent words as single vector of multiple dimensions, which ignore the ambiguity character Chinese. To solve this problem, we apply BERT-BiLSTM-CRF model to Chinese electronic medical records named paper. This enhances semantic by using BERT pre-trained model, then combine BiLSTM network with CRF layer, used input for...
Time series forecasting (TSF) is fundamentally required in many real-world applications, such as electricity consumption planning and sales forecasting. In e-commerce, accurate time-series (TSSF) can significantly increase economic benefits. TSSF e-commerce aims to predict future of millions products. The trend seasonality products vary a lot, the promotion activity heavily influences sales. Besides above difficulties, we know some knowledge advance except for historical statistics. Such may...
Many data cleansing algorithms are based on the low dimensional currently, and can't meet requirement of accuracy that warehouse in enterprise processes high data. In this paper idea using CLIQUE algorithm to process was adopted. Aiming at insufficient processing precision algorithm, meshing pruning were improved by dynamic incremental algorithms. The result testing shows can improve clustering effectively judge similar abnormal points which support cleansing.
The core of engineering education Accreditation is to estimate qualification each institute and make a direction for institutes development. purpose improve universities ability educate high quality engineers enhance Chinese pace with international education. Educational resources that affect the accreditation one important factors, Therefore, model educational resource becomes very insure Accreditation. This paper use Cloud Hierarchical Analysis (CHA) deal fuzziness uncertainty in can...
The recurrent convolutional neural network is an advanced that integrates deep structure and convolution calculation. feedforward with operation important method of learning. In this paper, the are combined to establish a model composed anomalies, LSTM (Long Short‐Term Memory), CNN. This study combines principal component analysis predict analyze test results students’ physical fitness standards. innovation lies in introduction function use conduct qualitative research on seven evaluation...
Abstract. This paper focuses on how to establishment of project team highest efficiency in the single which relative complexity and human resource with relatively stable attributes exist. An attempt model organization complex constraints was had. In designing this model, there are three steps:(1) initialization definition via four models, (2) definitions five variables people, project, factors, factors’ weights, people evaluation indexes, (3) implementation model. Then algorithm problem...
Many data cleansing algorithms are based on the low dimensional currently, and can't meet requirement of accuracy that warehouse in enterprise processes high data. In this paper idea using CLIQUE algorithm to process was adopted. Aiming at insufficient processing precision algorithm, meshing pruning were improved by dynamic incremental algorithms. The result testing shows can improve clustering effectively judge similar abnormal points which support cleansing.