Xi, ZHANG
Smart Assistant
[Summary]
Schedule exists everywhere in our daily life. The process of scheduling includes putting forward tasks, evaluating, organizing and optimizing. It does cost much energy and time to complete the process manually. However the number of events that need carefully consideration is actually very small. Most of the time, it is only a mechanical work that need only arranging the events follow certain rules. Our goal is to set up an expert system to help users to schedule their events. Through an optional simple training, the system could schedule the events, offer related information such as advice and ads, as well as help gather people to participate in special activities. And the system will also learn by itself through users’ feed back of their behaviors.[Details]
- Statistical Learning for Automatically Arrange Events
- Question Training
- Sample Training
- Self-Learning
- Expert Knowledge Base
- Official Expert Base
- Baidu zhidao / wiki mode
- Self learning/ Self Grow Knowledge Base
- Matching Module
- Match with Ads (Commercial/Commonweal)
- Match with Information/Guide（Organize activity/knowledge base）
- Event Description Language：
- User Interface
- Information Passage inner program
[Results]
[Reference]
[Machine Learning]Supervised Learning Approach towards Ranking and Sorting Problems
[Summary]
Recently, classification problems have been well developed. The popular methods for classification, Neural Network, SVM and Boosting have also been applied to related fields such as ranking and sorting. Although ranking and sorting problems have attracted the attention of researchers in the machine learning community, most of the solutions simply treat the ranking/sorting problems as multi-class classification problems. In this paper, we will propose a specified approach for the ranking / sorting problems based on 2-class classification SVM method.[Details]
[Results]
A Supervised Learning Approach towards Ranking and Sorting Problems[Reference]
[Machine Learning]Image Land Cover Classification
[Summary]
- Segmented image and Extract Features through matrix points
- Analysis the SVM classifier by adjust kernel functions and their parameters
- Optimized the methods used in common image process to apply on the remote sensing images of three different bands
[Details]
[Results]
[Reference]
[Parallel]Parallel Computing in the Process of Remote Sensing Image
[Summary]
- Parallel Registration algorithms and raise new method
[Details]
[Results]
[Reference]
[Parallel]Maze (The largest P2P network of CERNet)
[Summary]
- Analysis the system log and presented the results in easy-to-learn manners automatically through the monitor and record of the status of the system
- Traced the efficiency of distributed system
[Details]
[Results]
[Reference]
[Other]InSAR(China High Tech 863 Project)
[Summary]
- Processed InSAR images follow steps of filter etc.
- Design and implement (with VC++) the User Interface, integrate the UI and functional DLLs
[Details]
[Results]
[Reference]
[Other]MPMES
[Summary]
- Mission Planning/Monitoring/Evaluation System on Aerial Remote Sensing
- Design and implement (with VC++) the framework of the software