Academic Background

Research Interests

Data Mining, Parallel/Distributed Computing, Stream Mining, Utility Mining, Business Intelligence, Text Mining

Education

Northwestern University, Evanston, IL, USA


Ph.D in Computer Engineering

December 2005

   Dissertation: High Performance Data Mining Techniques for Large Databases
   Advisor: Prof. Alok Choudhary
   GPA:3.82/4.0

Northwestern University, Evanston, IL, USA


M.S. in Computer Engineering

December 2001

   Project: Using WordNet to Disambiguate Word Senses
   Advisor: Prof. Peter Scheuermann
   GPA: 3.82/4.0

Peking University, Beijing, China


B.S. (Major) in Computer Science

July 1999

   GPA: 3.65/4.0
   B.A. (Minor) in Economics
   GPA: 3.88/4.0

Summary of technical skills

Languages: C, C++, OPENMP, MPI, Java, SQL, JDBC, HTML, CGI
Software: Oracle 8i, Oracle JDeveloper, VTune Performance Analyzer, Clementine, Matlab, XLMiner, AMPL
Operating Systems: Unix, Linux, Solaris, Windows, DOS
Database: Oracle

Research Experience

Fictitious Economy and Data Science Research Center, Chinese Academy of Sciences


Associate Professor

January 2006 - present

Individual credit scoring:Conduct research on credit scoring algorithms and implement the credit scoring system. Project funded by the people's bank of China.

Optimization based utility mining. Research funded by Natural Science Foundation of China.

Center for Ultra-Scale Computing and Information Security, Northwestern University


Research Assistant

September 2001 - December 2005

Distributed mining on data streams:Conducting research on distributed data stream mining. Due to security issues, communication, memory and time-related constraints, stream mining algorithms performed on centralized data are not feasible. A distributed stream mining algorithm is being explored, which offloads simple mining to the sensors and discovers global patterns based on the local mining results at minimum communication cost. It is highly desirable for applications with continuous data streams from distributed servers or sensor networks, such as roadway traffic system, stock market transactions, ATM transactions, and clickstreams.

Utility mining: Utility mining identifies the high utility (profit or total value rather than the frequency of the presence of an item in Association Rule Mining) itemsets or most valuable customers from very large transaction databases. Designed and implemented Two-Phase utility mining algorithm, which outperforms the existing algorithms in terms of accuracy, memory space cost, and speed by several orders of magnitude. A number of high profit item combinations and valuable customers were efficiently discovered from our experiments on a grocery store's real data, which guided its business strategy.

Performance evaluation and characterization of scalable data mining algorithms: Developed a data mining Benchmark suite named MineBench (consisting of several representative data mining algorithms), and parallelized the programs on Shared-Memory Parallel machines (SMP) using OpenMP. Evaluated the performance, characterized the computation kernels of MineBench programs and investigated the usage of the memory hierarchy (in data access patterns and data locality) on a real machine and on a simulation tool. The kernels and characteristics observed from this research helped provide guidance for new architecture and algorithm optimizations. Research funded by Intel Corporation.

High performance data mining in large-scale scientific simulations: As the data mining computation in large-scale scientific applications is time-consuming, a parallel clustering algorithm, called HOP, was designed and implemented. This algorithm achieves good speedups by partitioning data into a balanced KD tree and minimizing inter-processor communication through batch data transfer. An online processing model was proposed to integrate the parallel data mining algorithms into scientific simulations so that the entire simulation cycles can execute automatically without human intervention or data input/output when performing data mining applications repeatedly and continuously. Research funded by Department of Energy (DOE).

Database System Group, Northwestern University


Research Assistant

September 1999 - August 2001

Text classification: As each word in a text document may have multiple meanings (senses), it is difficult to classify a document to its appropriate class. Designed and implemented a WordNet-based algorithm to disambiguate word senses in a text document. Each keyword is automatically mapped to its synsets and a counter is maintained for each synset. The sense a keyword uses is determined by the synset containing it with the greatest counter. A document is classified based on the senses of its keywords. (using Java, CGI, JDBC, SQL, Oracle 8i)

Data warehousing: Designed a data warehouse for Mass Storage Performance Information System, implemented the user interface and query system to facilitate the knowledge discovery. (using Java, CGI, JDBC, SQL, Oracle 8i) Research funded by National Aeronautics and Space Administration (NASA).

CASE Research Group, Peking University, Beijing, China


Development Intern

February 1999 - June 1999

Designed and implemented the statistic tools for Qingniao object oriented online database system. (using C++, HTML, CGI, Oracle Database)

Teaching Experience

Kellogg School of Management, Northwestern University


Teaching Assistant

September 2005 - December 2005

   TECH-917: Business Intelligence Technologies and Customer Relationship Management
   Designed and graded homework assignments.

Electrical and Computer Engineering Department, Northwestern University


Teaching Assistant

January 2005 - March 2005

   INF_TECH-423: Database, Information and Decision Support Systems
   Designed and graded homework assignments.

Kellogg School of Management, Northwestern University


Teaching Assistant

March 2004 - June 2004

   TECH-917: Business Intelligence Technologies and Customer Relationship Management
   Designed and graded homework assignments.

Electrical and Computer Engineering Department, Northwestern University


Teaching Associate

January 2002 - March 2002

   CS-339: Introduction to Database Systems
   Gave lectures to students, designed and graded the final project, graded homework assignments, held office hours.

Electrical and Computer Engineering Department, Northwestern University


Teaching Associate

January 2001 - March 2001

   ECE-230: Introduction to Computer Engineering
   Gave lectures to students, led lab sessions, graded homework assignments and final project, held office hours.

Honors

Dissertation-Year Fellowship Nomination, Northwestern University, 2004
Best Poster Award, Poster session in the Annual Research Symposium at the Electrical and Computer Engineering Department, Northwestern University, 2003
Research Assistantship, ECE Department, Northwestern University, 1999 - Present
Outstanding Undergraduate Award, Peking University, 1997
XI YUE Scholarship to best undergraduate students, Peking University, 1997
GE (General Electronics) China Scholarship to best undergraduate students, Peking University, 1996
Outstanding Freshmen Award, Peking University, 1995

Publications

Book Chapter

Jianwei Li, Ying Liu, Wei-keng Liao, Alok Choudhary, Parallel Data Mining Algorithms for Association Rules and Clustering, Handbook of Parallel Computing: Models, Algorithms and Applications. Sanguthevar Rajasekaran and John Reif, ed., CRC Press, 2006.

Conference & Workshop Papers

• Lin Zhou, Ying Liu, Jing Wang, Yong Shi, Utility-based Web Path Traversal Pattern Mining, Workshop on High Performance Data Mining in conjunction with IEEE International Conference on Data Mining, October 2007, Omaha, USA.
  

• Ying Liu, Peter Scheuermann, Xingsen Li, Xingquan Zhu, Using WordNet to Disambiguate Word Senses for Text Classification, Workshop on Text Data Mining in conjunction with 7th International conference on Computational Science, May 2007, Beijing, China.
  

• Jing Wang, Ying Liu, Lin Zhou, Yong Shi, Xingquan Zhu, Pushing Frequency Constraint to Utility Mining Model, Workshop on High Performance Data Mining in conjunction with 7th International conference on Computational Science, May 2007, Beijing, China.
  

• Xingsen Li, Ying Liu, Jun Li, Yong Shi, Yuejin Zhang, A Knowledge Management Model for Middle and Small Enterprises, 2006 Internatioal Conference on Distributed Computing and Applications for Business Engineering and Sciences, October 2006, Hangzhou, China.
  

• Ying Liu, Alok Choudhary, Jianhong Zhou, Ashfaq Khokhar, A Scalable Distributed Stream Mining System for Highway Traffic Data, The 10th European Conference on Principles and Practice of Knowledge Discovery in Databases (PKDD), Spetember 2006, Berlin, Germany.
  

• Ying Liu, Wei-keng Liao, Alok Choudhary, A Fast High Utility Itemsets Mining Algorithm, Workshop on Utility-Based Data Mining in conjunction with the 11th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD 2005), August 21-24, Chicago, Illinois, USA.
  

• Ying Liu, Wei-keng Liao, Alok Choudhary, A Two-Phase Algorithm for Fast Discovery of High Utility Itemsets, The 9th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD), May 2005, Hanoi, Vietnam.
  

• Ying Liu, Jayaprakash Pisharath, Wei-keng Liao, Gokhan Memiki, Alok Choudhary, Pradeep Dubey, Performance Evaluation and Characterization of Scalable Data Mining Algorithms, The 16th IASTED International Conference on Parallel and Distributed Computing and Systems(PDCS), Novermber 2004, MIT Cambridge, Massachusetts, USA.
  

• Ying Liu, Wei-keng Liao, Alok Choudhary, Steve Chiu, On-Line Processing Model for Data Mining in Large Scientific Simulations, The 7th Workshop on Mining Scientific and Engineering Datasets in conjunction with the SIAM International Conference on Data Mining, April 2004, Lake Buena Vista, Florida, USA.
  

• Wei-keng Liao, Ying Liu, Alok Choudhary, A Grid-based Clustering Algorithm using Adaptive Mesh Refinement, The 7th Workshop on Mining Scientific and Engineering Datasets in conjunction with the SIAM International Conference on Data Mining, April 2004, Lake Buena Vista, Florida, USA.
  

• Ying Liu, Wei-keng Liao, Alok Choudhary, Design and Evaluation of a Parallel HOP Clustering Algorithm for Cosmological Simulation, The 17th International Parallel and Distributed Processing Symposium (IPDPS), April 2003, Nice, France.
  

  Posters

• Ying Liu, Jianwei Li, Wei-keng Liao, Alok Choudhary, Utility Mining and Advanced Association Mining Algorithms, The 2nd Midwest Database Research Symposium, April 2005, Chicago, Illinois, USA
  

• Ying Liu, Jayaprakash Pisharath, Wei-keng Liao, Gokhan Memiki, Alok Choudhary, Design, Development, and Characterization of Data Mining Algorithms on High Performance Computers, The 1st Midwest Database Research Symposium, April 2004, Chicago, Illinois, USA
  

• Jayaprakash Pisharath, Ying Liu, Wei-keng Liao, Alok Choudhary, High Performance Data Mining, The 1st Midwest Database Research Symposium, April 2004, Chicago, Illinois, USA
  

• Ying Liu, Steve Chiu, Jianwei Li, Wei-keng Liao, Alok Choudhary, High-Performance Data Storage, Management, and Mining for Parallel Data-Intensive Computing, Poster session in the Annual Research Symposium at the Electrical and Computer Engineering Department, Northwestern University, April 2003, Evanston, Illinois, USA
  

Contact Information

Email: yingliu@ece.northwestern.edu