Funding

Self-Support

Website

http://192.168.3.17/Searcher2/index.html

Introduction

Studies the way to automaticly create facets for results returned by search engines and organize facets and results in user favored interface

Funding

China National Natural Science Foundation

Grant No.

NO.60473122

Introduction

Studies the charateristics of the graph structure of China Web Graph and its evolution over time

Funding

IBM China Research Laboratory

Website:

http://apex.sjtu.edu.cn/projects/orient/News.htm

Introduction

This project aims at researching on ontology building and evolution methods and developing an integrated ontology engineering environment -- ORIENT. ORIENT has the following objectives/features:
*Supports manipulation of very large ontology that stored in a database
*Supports ontology mapping and evolution methods
*Based on the Eclipse tool platform

Funding

Microsoft Research Asia

Introduction

On the basis of China Web mining and the co-analysis of retrieved results and whole collection, we hope to provide a high-quality, visualized and interactive Chinese Web search platform.

Funding

Google Inc.

Introduction

Chinese Bulletin Board System (BBS) and Weblog have been expanded in an incredible speed, and also contain a huge amount of information. Therefore, we need to propose some advanced methods to facilitate users to better utilize those resources. The project consists of two main components:
*Discover user interests by mining personal Blog data for personalized search and automated advertisements recommendation.
*Classify BBS posters according to the category predefined, such as performance, price, quality of a product etc. An end user can see what others review on a product, and an enterprise user, he/she can obtain feedback from the statistic results.

Funding

Self-support

Website

http://apex.sjtu.edu.cn/projects/semsearch

Introduction

Semantic information makes it possible to utilize formal reasoning to answer user's query more acurately. However, what is the relationship between reasoning and retrieval? Can they help each other? How to integrate them? Because of the inherent uncertainty in judging whether a piece of information satisfies a user's need, and of the difference between formal inference mechanism and common-sense human reasoning, these problems are quite unique and difficult. This research tries to solve the problem by establishing models and methods for semantic search. We will test the solution by applying it in searching enterprise knowledge portals.

Funding

Self-support

Introduction

P2P system is a good platform for resource-sharing. As more documents are shared, especially texts, there has to be a good method to solve text retrieval problem. As a branch of distributed information retrieval, P2P system is facing lots of difficulties, which is a challanging research topic. Out research tries to improve current P2P platforms, such as JXTA, and integrate P2P community mechanisms so that search task efficiency can be improved.