If it has a syntax, it isn't user friendly

-- /usr/game/fortune

1. Project Overview

Keyword search is a dominant way of accessing information from unstructured data sources (such as the Web, emails, etc.). This project investigates technologies to enable free-form (google-style) keyword search on structured data sources (such as a relational database) and semi-structured data sources (such as an XML document repository).

There are many technical challenges to support this simple and intuitively plausible idea. For one thing, how to define the search results? On the Web, each result is just a page (or document). On (semi-)structured data sources, it is typically a collection of elements that are semantically meaningful and collectively relevant to the query.

As an example, we show the results of searching "tom cruise nicole kidman" on our system (SPARK) against the IMDB movie info database, and on other systems (Google and WolframAlpha).

SPARK Google Google Wolfram Alpha

SPARK screenshot

Google screenshot 1

Google screenshot 2

Wolfram Alpha screenshot

This research project is partly funded by ARC Discovery Projects DP0881779 and DP130103405.

2. People

3. Publications

  1. Xiang Wang, Ying Zhang, Wenjie Zhang, Xuemin Lin, Wei Wang. AP-Tree: Efficiently Support Continuous Spatial-Keyword Queries over Stream. ICDE 2015.

  2. Shiyu Yang, Muhammad Aamir Cheema, Xuemin Lin, Wei Wang. Reverse k Nearest Neighbors Query Processing: Experiments and Analysis. PVLDB 2015.

  3. Xiaoyang Wang, Ying Zhang, Wenjie Zhang, Xuemin Lin, Wei Wang. Selectivity Estimation On Streaming Spatio-Textual Data Using Local Correlations. VLDB 2015.

  4. Junfeng Zhou, Xingmin Zhao, Wei Wang, Ziyang Chen, Jeffrey Xu Yu. Top-Down Keyword Query Processing on XML Data. CIKM 2013.

    • check out the full TR version here.

  5. Junfeng Zhou, Zhifeng Bao, Wei Wang, Jinjia Zhao, and Xiaofeng Meng. Efficient query processing for XML keyword queries based on the IDList index. VLDB Journal, 2013.

  6. Jianxin Li, Chengfei Liu, Rui Zhou, Wei Wang. XML keyword search with promising result type recommendations. World Wide Web, 2013.

  7. Xiaoling Zhou, Yifei Lu, Yifang Sun, Muhammad Aamir Cheema. Improved Spatial Keyword Search Based on IDF Approximation. APWeb 2013.

  8. Junfeng Zhou, Zhifeng Bao, Wei Wang, Tok Ling Wang, Ziyang Chen, Xudong Lin, Jingfeng Guo. Fast SLCA and ELCA Computation for XML Keyword Queries Based on Set Intersection. ICDE 2012.

  9. Yi Chen, Wei Wang, Ziyang Liu. Searching, Analyzing and Exploring Databases. DASFAA 2011.

  10. Yi Chen, Wei Wang, Ziyang Liu. Keyword-based Search and Exploration on Databases. ICDE 2011.

  11. Yifei Lu, Wei Wang, Jianxin Li, Chengfei Liu. XClean: Providing Valid Spelling Suggestions for XML Keyword Queries. ICDE 2011.

  12. Jianxin Li, Chengfei Liu, Rui Zhou, Wei Wang. Top-k Keyword Search over Probabilistic XML Data. ICDE 2011.

  13. Yi Luo, Wei Wang, Xuemin Lin, Xiaofang Zhou, Jianmin Wang, Keqiu Li. SPARK2: Top-k Keyword Query in Relational Databases. TKDE.

  14. Jianxin Li, Chengfei Liu, Rui Zhou, Wei Wang. Suggestion of Promising Result Types for XML Keyword Search. EDBT 2010.

  15. Yi Chen, Wei Wang, Ziyang Liu, Xuemin Lin. Keyword Search on Structured and Semi-structured Data. SIGMOD 2009 (tutorial).

  16. Yi Luo, Wei Wang, Xuemin Lin. SPARK: A Keyword Search Engine on Relational Databases. ICDE 2008. (demo)

  17. Yi Luo, Xuemin Lin, Wei Wang, Xiaofang Zhou. SPARK: Top-k Keyword Query in Relational Databases. SIGMOD 2007.

  18. Wei Wang. Keyword Search in Databases. Tutorial at APWeb 2006.

4. Online Demo and Download of the Prototype

Please visit here.

5. Commercial Product

SPARK - Enterprise Graph Search: An enterprise graph search product by Lumanetix Software founded by Nino Svonja featuring state-of-the-art technology inspired by this research.