Semantics: Hot Again?
January 11, 2011
We hear this each year: Semantics will be hot in [fill in the year].
Kazeon’s The Future eDiscovery Arms Race: It is all about the Semantics investigates where the eDiscovery market must go in order to handle the growing volume of ESI while progressing the efficiency, accuracy and reliability of the search result process. After posing this question, where does Kazeon end up? At the doorstep of Semantics Future Institute.
Those who only speak English may not realize what a complicated language we really have. Words used regularly often carry several meanings, at times even four or more definitions to a single word. This is known as polysemy, and it throws a huge wrench in the search method.
We are beginning to see some workarounds to this issue finally realized. One of the major players is Latent Semantic Analysis, which plainly “searches documents for themes within the language usage and extracts the concepts, which are common to the documents”, helping to alleviate false positive results. Other aides include Word Search Disambiguation, which focuses on word meaning rather than merely matching character strings and Local Co-Occurrence Statistics, which counts how often sets of terms appear together within a predetermined period.
These tools will no doubt be helpful in refining search techniques, but our main question is, where was Kazeon five years ago when the semantic buzz began?
Sarah Rogers, January 11, 2011
Freebie
Comments
-
- Subscribe to Beyond Search
Feature archive
News archive
-
