Main research field are consitute on Theory and Algorithm for Data Mining and Data Stream and other cross research on data mining. 

1、Theory and Algorithm study for Data Mining

As one of early scholar for researching theory and algorithm for data mining from 1998, conitinually research and coorperate with a lot of noted expert from home and aboard, expectrally in 2011 as senior vistiting scholar for Institute of Software Academy of Sciences .

Main research puclications included：

（1）      After mention the concept of （Set of Item-sequences）at first in 2002，published in 《Chinese Journal of Computers》(2002-4）；then expanded and refined algebraic lattice space and knowledge evolution model based on the concept in《Chinese Journal of Electronics》、《ACTA AUTOMATICA SINICA》. The work is obtained the domestic profession expert's affirmation by more than 200 accumulative citations.  

（2）     In  2003，doctoral dissertation“Research on data mining technology and association rule mining algorithm”，concentrate on the research on theory , is one of the early systematically study , which is cited more than 170 papers, and was downloaded about ten thousands by internet database.  

2、Theory and Algorithm study for Data Stream Mining

Focus on Data Stream study from 2005 as a visiting scholar of UVM, coorperated with international famous data mining expert professor Wu Xindong （Dean、Chief Editor of IEEE Trans and TKDE journal ），became one of scholars involved in early study one data stream. Based on internaltional study, focus more on Online maximum frequency set mining problem for data stream.  Typical results included in ：

（3）  In    2005，Proposed On - line maximum frequency set mining method on data stream INSTANT. After publicated as technical  report of University of Vermount ，recieved a lot of comments on the result among on international conference and journal papers. For example,Data processing top-level meetings SIGMOD’06 paper“Research Issues in Data Stream Association Rule Mining”comments on：“A few of the proposed algorithms generate exact mining results by maintaining a small subset，┄. Another way is to maintain only special itemsets such as short frequent itemsets, closed frequent itemsets or maximal frequent itemsets as in [Yang, 2004；Mao, 2005]”. [Mao，2005] refer to algorithm of INSTANT，is one of typical methord on the research area. 

（4）    In  2007，using algebraic lattice theory, the problem of operators and models required for online mining of data streams is solved. The results on publicated on 《Journal of Information Science》。The paper is on TOP 500 of  citation frequency single paper of global information science in 2007.（No.286，statistics on 《 Science in the news》）

3、Cross research on data mining

Late 5 years,exploratory work on cross research on data mining methord and other subjects or research as  economics, SNS(Social Networking Services), information security, big data etc. With the suport of NSFC(Natural Science Foundation of China),  a few results on Applied model prototype system development publicated on interntional professional journals  such as IJCIS、IJFCC etc，which is one of important direction of future research.