Participated Projects

ASU Coordination Tracker (Sep 2009-Current)

The ASU Coordination Tracker (ACT) is an innovative tool from Arizona State University that helps non-profit organizations  collectvisualize, and act on requests for humanitarian aid. It is an event response coordination system that facilitates multi-organization responses to an event, including natural disasters, disease epidemics, and unrest. ACT provides a fast, effective, and open way to streamline communication and coordination. It leverages crowdsourced information, and simplifies complex data analysis with the form of a groupsourced response. ACT helps visualize available and deployed resources on interactive maps according to their distribution, with the increasing efficiency by avoiding duplicate responses to the same incidents. [more details]


Online Streaming Supervising System (Jun 2008–Sep 2008)

The Online Streaming Supervising System is a project in Pattern Recognition and Intelligent System Lab of Beijing University of Posts and Telecommunications. It is developed for SARFT (the State Administration of Radio File and Television), during the period of my master study. The system is designed to detect the online streaming (i.e., videos) that contain illegal information in real time during the 2008 Beijing Olympics. It considers various attributes of the online streaming, eg., title, comments, and contents, and classifies them into several security levels. Streamings with higher security level, i.e., containing more illegal information, are then labeled by experts for further processing.


Campus Object Search Engine (Sep 2007–Apr 2008)

The Campus Object Search Engine (COSE) is an information retrieval system of Pattern Recognition and Intelligent System Lab in Beijing University of Posts and Telecommunications (BUPT). It leverages a campus-oriented web crawler to collect the web pages through the whole campus network, aiming at building an informational center like wiki for all the students in BUPT. The retrieval results are clustered to various relevant entities before retuening to users. Students in BUPT can then search information around the campus, such as news of restaurants, notification of schools, and look up the results through relevant topics.


See-Saw (Oct 2006–Apr 2007)

See-Saw is an intelligent assistant system to improve learning performance for primary students. It helps a student generate an efficient and healthy learning habit, by monitoring his studying status and then giving feedbacks to him. The system extracts various features of a user’s studying status, including sitting posture and studying concentration level, while the former utilizes physical sitting posture sensors with analysis module, and the latter is judged by processing user’s eye image with a “perclos”-based algorithm. These features are then leveraged to compute the user's study efficiency and return feedbacks with pre-stored expert judgements. The system has ranked the world top 200 in “Microsoft Imagine Cup-Embedded Development Competition”, 2007.


Resume Parsing System (Nov 2006–Jun 2007)

The Resume Parsing System is part of my bachelor's thesis. It is proposed to help companies analyze millions of resumes in a short time, therefore faciliate them in efficiently finding the best candidates. It analyzes the layout of resumes, extract users’ information, and then store the analysis results. A four-Layer extracting strategy is devised together with an entry transition probability matrix, which greatly improves the recognition and extracting performance over traditional methods. A testing tool is also developed to evaluate the results with the consideration of human knowledge.


First Aid Expert System (Mar 2006-Aug 2006)

The First Aid Expert System (FAES) is an embedded system based on Intel XScale platform. Due to the current first aid situation in China, most people can not be cured in time when sent to hospital with ambulance. We established this embedded system located in the ambulance, aimed at collecting and analyzing electrocardiosignal of the patients, therefore make sure the patients is safe during the transition. FAES motinors the patient’s health status via analyzing his physiological data, especially the electrocardiosignal which is collected by the corresponding medical equipments. The system has won the first prize in “2006 ‘INTEL Cup’ - National College Students Embedded System Competition”.