A5053918335- Muon and positron interactions and applications
- Traffic Prediction and Management Techniques
- Gear and Bearing Dynamics Analysis
- Human Mobility and Location-Based Analysis
- Tribology and Lubrication Engineering
- Advancements in Battery Materials
- Semiconductor materials and devices
- Advanced Battery Materials and Technologies
- Ammonia Synthesis and Nitrogen Reduction
- Advanced Adaptive Filtering Techniques
- Geographic Information Systems Studies
- Silicon Nanostructures and Photoluminescence
- Advanced Battery Technologies Research
- Graphene research and applications
- Data Management and Algorithms
- Non-Destructive Testing Techniques
- Vehicle Noise and Vibration Control
- Anomaly Detection Techniques and Applications
- Infrastructure Maintenance and Monitoring
- Advanced Welding Techniques Analysis
- Complex Network Analysis Techniques
- Radiation Effects in Electronics
- Machine Learning and Algorithms
- Structural Integrity and Reliability Analysis
- Acoustic Wave Phenomena Research
Preferred Networks (Japan)
2021
Tokyo Institute of Technology
2020
National Institute of Informatics
2014-2019
The University of Tokyo
2014-2017
Hitotsubashi University
2015-2016
Osaka Electro-Communication University
2012
Nissan (United Kingdom)
2012
Sanyo (Japan)
2001-2008
Kumamoto University
2006
Tokyo Denki University
1993-1996
Elevating the charging voltage of lithium-ion batteries with a cathode is investigated to develop them toward high capacity and energy density. Three countermeasures are found be essential overcome side reactions subsequent cycle degradations caused by higher potential: limiting cut-off potential below vs for cathode, modification particles element, controlling ratio ethylene carbonate in electrolyte, which major cause degradation an elevated condition. It suggested that oxidized dissolves...
Traffic congestion occurs frequently in urban settings, and is not always caused by traffic incidents. In this paper, we propose a simple method for detecting incidents from probe-car data identifying unusual events that distinguish spontaneous congestion. First, introduce state model based on probabilistic topic to describe the states variety of roads. Formulas estimating parameters are derived, so usual can be learned using an expectation–maximization algorithm. Next, several divergence...
Abstract We built a machine learning model (ML model) which input the number of daily infection cases and other information related to COVID-19 over past 24 days in each 17 provinces South Korea, output total increase coming days. employ combination XGBoost MultiOutputRegressor as model). For province, we conduct binary classification whether our ML can classify where is more than 100. The result Sensitivity = 3/3 100%, Specificity 11/14 78.6%, False Positive Rate 3/11 21.4%, Accuracy 14/17...
The peak frequency of the dielectric loss angle gas molecules adsorbed in a porous silicon sensor having 2.4 nm pore radius is found to vary inversely proportional third power molecular radius. Peak extremely sensitive radius, becoming about one order magnitude higher for 1 nm. temperature dependence range 0 25 °C was also determined.
We have developed a real-time traffic incident detection system for the Tokyo Metropolitan Expressway. This monitors current using probe-car data and compares actual in real time with usual traffic, which is estimated advance batch processing.
This paper proposes a latent statistical model for analyzing global positioning system (GPS) trajectory data. Because of the rapid spread GPS-equipped devices, numerous GPS trajectories have become available, and they are useful various location-aware systems. To better utilize data, number sensor data mining techniques been developed. discusses application model to two closely related problems, namely, moving mode estimation and interpolation observation. The proposed...
There is a possibility that alternating current will cause corrosion in defect of the coating on buried steel pipeline. In this research, we developed an evaluating method with which could evaluate short time compared conventional by mesuring weight loss. The results research showed under cathodic protection be estimated measuring IR-drop-free potential, especially peak (maximum) potential using probe. It was also found achieved when over 100mV less noble than free steel.
Summary Deep learning has the potential to estimate velocity models directly from shot gathers, which would reduce turn-around time of seismic inversion. Our study addresses two challenges in implementing deep techniques for inversion: practical generation a large amount training data and search best neural network architecture. First, we propose flexible system parametrically generates create large-scale, complex fully synthetic dataset, without using target subsurface model. Using this...