Google Research

An Integrated Framework for Spatio-Temporal-Textual Search and Mining

  • Bingsheng Wang
  • Haili Dong
  • Arnold Boedihardjo
  • Chang-Tien Lu
  • Harland Yu
  • Ing-Ray Chen
  • Jing Dai
20th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems (ACM SIGSPATIAL GIS 2012), ACM, 2 Penn Plaza, Suite 701, New York, NY 10121, pp. 570-573


This paper presents an integrated framework for Spatio-Temporal-Textual (STT) information retrieval and knowledge discovery system. The proposed ensemble framework contains an efficient STT search engine with multiple indexing, ranking and scoring schemes, an effective STT pattern miner with Spatio-Temporal (ST) analytics, and novel STT topic modeling. Specifically, we design an effective prediction prototype with a third-order linear regression model, and present an innovative STT topic modeling relevance ranker to score documents based on inherent STT features under topical space. We demonstrate the framework with a crime dataset from the Washington, DC area from 2006 to 2010 and a global terrorism dataset from 2004 to 2010.

Learn more about how we do research

We maintain a portfolio of research projects, providing individuals and teams the freedom to emphasize specific types of work