A5010520358- Biomedical Text Mining and Ontologies
- Energy Load and Power Forecasting
- Smart Grid Energy Management
- Robotics and Sensor-Based Localization
- Elevator Systems and Control
- SARS-CoV-2 and COVID-19 Research
- Smart Agriculture and AI
- Computational Drug Discovery Methods
- High voltage insulation and dielectric phenomena
- Bioinformatics and Genomic Networks
- Biosensors and Analytical Detection
- Robotic Path Planning Algorithms
- Topic Modeling
- COVID-19 Clinical Research Studies
- Blind Source Separation Techniques
- COVID-19 diagnosis using AI
- Water Quality Monitoring Technologies
- Advanced Manufacturing and Logistics Optimization
- Smart Grid Security and Resilience
- Pharmacovigilance and Adverse Drug Reactions
Shandong University
2020-2021
Iowa State University
2020
IBM Research - Almaden
2018
Jiangsu University
2018
IBM (United States)
2018
Shanghai University
2017
Significance We adapted natural language processing to the biological literature and demonstrated end-to-end automated knowledge discovery by exploring subtle word connections. General text mining scanned 21 million publication abstracts selected a reliable 130,000 from which hypothesis generation algorithms predicted kinases not known phosphorylate p53, but likely do so. Six of these p53 kinase candidates passed experimental validation. Among them NEK2 was examined in depth shown repress...
Regulations are increasing the scope of activities that fall under remit drug safety. Currently, individual case safety report (ICSR) collection and collation is done manually, requiring pharmacovigilance professionals to perform many transactional before data available for assessment aggregated analyses. For a biopharmaceutical company meet its responsibilities patients regulatory bodies regarding safe use distribution products, improved business processes must be implemented drive industry...
Effectively and efficiently diagnosing COVID-19 patients with accurate clinical type is essential to achieve optimal outcomes of the as well reducing risk overloading healthcare system. Currently, severe non-severe types are differentiated by only a few features, which do not comprehensively characterize complicated pathological, physiological, immunological responses SARS-CoV-2 invasion in different types. In this study, we recruited 214 confirmed 148 type, from Wuhan, China. The patients'...
Nonintrusive load monitoring (NILM) has become a key technology in the power Internet of Things as well an important information source for characteristics analysis. Whether selected features are appropriate determines effectiveness identification NILM. In order to reduce redundancy and improve relevancy features, feature selection method that balances is proposed. Based on approximate Markov blanket decision, this article puts forward basis set elimination. This used determine priority by...
The objective of this research was to eliminate the influence environmental stray light on measured spectral reflectance in a sensor system used weeding teleoperated robot for rape fields. mainly consists source, signal acquisition, and data processing. In study, optical modulation technology discrete Fourier transform method were carry out experiments based four characteristic wavelengths. results show that experimental can obtain stable reflectivity value no matter whether changes slowly...
Non-intrusive load monitoring(NILM) has become a key technology in the power Internet of Things and an important information source characteristics analysis. Whether selected features are appropriate determines effectiveness identification NILM. In order to reduce redundance improve relevance features, feature selection method which integrates mutual Relief-F algorithm is proposed. Based on analysis consumption various electrical appliances, prospective components initial set for NILM...
This paper presents a navigation system customized for the automatic path tracking movement of tracked stacking vehicle working in shipping container. The is equipped with 2D laser scanner localization and navigation. Returned data continuously sent back to be processed order identify location relative For localization, feature extraction algorithm based on IEPF priori geometric features container proposed. navigation, fuzzy logic controller designed by implementing error as input track...
Partial Discharge (PD) detection of gas insulated substation (GIS) must be done before commissioning, this is an important guarantee to reduce the incidence post-commissioning faults. Fast and accurate partial discharge location key improve test efficiency. Therefore, a two-stage method based on pre-screening defect recognition proposed in paper. The convolution neural network (CNN) used supervise train signal atlas recorded by GIS withstand voltage test. By identifying discharging atlas,...