Dan Wang, Zhu Han

Research output: Chapter in book / Conference proceedingChapter in an edited book (as author)Academic researchpeer-review


In February 2010, National Centers for Disease Control and Prevention (CDC) identified an outbreak of flu in the mid-Atlantic regions of the United States. However, 2 weeks earlier, Google Flu Trends [1] had already predicted such an outbreak. By no means does Google have more expertise in the medical domain than the CDC. However, Google was able to predict the outbreak early because it uses big data analytics. Google establishes an association between outbreaks of flu and user queries, e.g., on throat pain, fever, and so on. The association is then used to predict the flu outbreak events. Intuitively, an association means that if event A (e.g., a certain combination of queries) happens, event B (e.g., a flu outbreak) will happen (e.g., with high probability). One important feature of such analytics is that the association can only be established when the data is big. When the data is small, such as a combination of a few user queries, it may not expose any connection with a flu outbreak. Google applied millions of models to the huge number of queries that it has. The aforementioned prediction of flue by Google is an early example of the power of big data analytics, and the impact of which has been profound.

Original languageEnglish
Title of host publicationSpringerBriefs in Computer Science
Number of pages7
Publication statusPublished - 1 Jan 2015

Publication series

NameSpringerBriefs in Computer Science
ISSN (Print)2191-5768
ISSN (Electronic)2191-5776


  • Great houston area
  • Smart city
  • Smart grid
  • Theoretical computer science
  • User query

ASJC Scopus subject areas

  • General Computer Science


Dive into the research topics of 'Introduction'. Together they form a unique fingerprint.

Cite this