How google and microsoft taught search to understand the web
Download
1 / 6

How Google and Microsoft taught search to “understand” the Web - PowerPoint PPT Presentation


  • 72 Views
  • Uploaded on

How Google and Microsoft taught search to “understand” the Web. Austin Granger Chris Hesemann. Knowledge of the Web. String searching does not always convey the true meaning of content. Search by knowledge, not by sub-string matching.

loader
I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
capcha
Download Presentation

PowerPoint Slideshow about ' How Google and Microsoft taught search to “understand” the Web' - amaris


An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
How google and microsoft taught search to understand the web

How Google and Microsoft taught search to “understand” the Web

Austin Granger

Chris Hesemann


Knowledge of the web
Knowledge of the Web the Web

  • String searching does not always convey the true meaning of content.

  • Search by knowledge, not by sub-string matching.

  • Extracting and categorizing concepts allows for knowledge-based searching.


Web of concepts
“Web of Concepts” the Web

  • Extract raw data (phone numbers, addresses, prices, etc.).

  • Link related entities together (e.g., link actor to movie).

  • Categorize information about each entity (what does this store sell, what has this author written, how highly are they reviewed?).


  • Search engines discover webpages, parse them into objects and data, process them and store the data, updating existing entries as needed.

  • “Concept web” stored in vast databases.

    • Not traditional databases.

    • Based on graph theory, not relational model.

    • Database consists of nodes and links.


Memory cloud
Memory Cloud and data, process them and store the data, updating existing entries as needed.

  • To make this efficient we must traverse the entire graph in milliseconds.

  • One solution – “memory cloud.”

    • Store entire database within memory at all times.

  • Example: Google search “blowfish”

    • Results: Show company, encryption algorithm, sushi

    • New results: Suggest “pufferfish”


Limitations
Limitations and data, process them and store the data, updating existing entries as needed.

  • Currently only works in English.

  • Including other languages increases the complexity exponentially, we’ve got a long way to go.

  • Dissecting language to understand searches written in normal language, not just keywords.

The Future of Knowledge Searching


ad