Computer Science Engineer

Shanchan Wu

Department of Computer Science, University of Maryland, College Park, MD 20742, USA

H 301-***-**** B ***@**.***.***

Research Interests

Social Media and Social Network Analysis, Information Retrieval, Machine Learning and

Data Mining

Education

University of Maryland, College Park 20062012

Ph.D. Candidate, Computer Science expected graduation: August 2012

Dissertation Topic: Prediction in Social Media for Monitoring and Recommendation

(tentative)

University of Maryland, College Park 20062008

M.S., Computer Science

Tsinghua University, Beijing, China 20012004

Master of Engineering, Department of Automation

Tsinghua University, Beijing, China 19972001

Bachelor of Engineering, Department of Automation

Research and Work Experience

University of Maryland, College Park2008presentResearch Assistant

Advisor: Louiqa Raschid

{ Built a prediction model in microblog (twitter) by learning from a hybrid network.

{ Developed a scalable tool to crawl the user follower network and user information from

Twitter API. It can run coordinately

on multiple servers, efficiently speeding up the data collection.

{ Built a prediction system for blogs. Analyzed the links between different blogs. Built

the profiles for blogs. Developed

multiple solutions and trained a Ranking SVM. Identified a number of features that

improve the prediction accuracy.

{ Processed the blog data and built a retrieval system based on Lucene open source.

Implemented a state-of-art document

similarity metric.

IBM T.J. Watson Research Center05/200908/2009

Summer Research Intern in Information Retrieval

{ Designed and implemented a novel algorithm and a similarity metric to search for Top K

similar sentences from the transla-

tion memory which stores segments of text that have been previously translated. Built a

demo for Top K search for similar

sentences. Implemented the novel search algorithm based on the Lucene lower level API.

Conducted experimental evalua-

tions on real datasets.

{ Set up a Hadoop cluster configuration for cloud computing. Built up the translation

memory index using the Hadoop

cluster.

IBM T.J. Watson Research Center 05/2008 08/2008

Summer Research Intern in Database Systems

{ Studied DB2 z/OS Table Space Allocation Algorithm, which finds available free space to

insert a new record while trying

to maintain data clustering, minimize waste of space and unnecessary space extension.

Developed a simulation tool to

simulate the algorithm. Provided the product and development teams a convenient tool to

optimize the current algorithm.

Built a platform for further development of the simulation of new table space allocation

algorithms. Produced a published

research paper from this work.

{ Improved the current algorithm. Identified several techniques to improve the algorithm.

Implemented some of the tech-

niques and demonstrated better performance. Planed a direction for further improvement.

Yahoo! China 07/2004 07/2006Software Engineer

1/3

{ Led a tool development team. Led a project of building a popular search words ranking

system and the project of building

the new customer center.

{ Applied Yahoo Search Technology to the web publishing system. Improved the system

efficiency.

{ Improved the fetch program for yahoo china, which automatically fetched some particular

data from external web pages,

inserted it into database, and pushed the generated HTML files to the Yahoo China front

servers.

{ Technical support for yahoo china portal revision. Developed some editor tools for the

portal editors, including slide show,

live text/picture broadcasting.

{ Administrated and maintained Yahoo China web publishing system (Yahoo Jake), which was

distributed on a cluster of

Unix servers.

Tsinghua University, Beijing, China 09/200107/2004Research Assistant

{ Designed the technique to analyze the web log sequential patterns. Built a web log

mining and recommendation demo

system. Developed a model to predict web requests.

{ Applied the Rough Set theory to analyze the attribute reduction and discretization of

attributes with real values. Discretiza-

tion is an important preprocessing step of the data mining to make the data mining

techniques useful for the real applications

containing attributes with continuous values.

Academic Honors

{ Privilege to enter the Master program at Tsinghua with Entrance Exams free, 2001

{ General Electric Scholarship, Tsinghua University, 1999-2000

{ Sanwa Bank International Fund Scholarship, Tsinghua University, 1998-1999

{ Excellent Student Scholarship of Tsinghua University, 1997-1998

{ Excellent Freshman Scholarship, Tsinghua University, 1997

Publications

{

Computer Skills

Languages: Java(proficient),C++(proficient), C, Python, Perl, PHP, Unix Shell Script, SQL

Tools and OS: Eclipse, Microsoft Visual Studio, Unix/Linux, Windows

Databases: MySql, SQL Server, DB2References

Available upon request.

3/3

Contact this candidate