Buzz Words for Clustering Abound
December 21, 2012
As the end of the year 2012 approaches, media mavens are abuzz with new buzz words. We have sighted another in the recent Clearwell Systems blog post on e-discovery 2.0. The article, “Q&A with Predictive Coding Guru Maura R. Grossman, Esq.” points out the idea of technology assisted review(TAR).
Technology assisted review is described as being synonymous with clustering, concept search or other early case assessment tools.
One question addressed is the number one mistake practitioners should aim to avoid when using these tools. The article tells us that accuracy can be misleading because it is usually impacted by the number of relevant documents in the overall total collection.
Delving deeper into this scenario, the article states:
“Consider, for example, a document collection containing one million documents, of which ten thousand (or 1%) are relevant. A search or review effort that identified 100% of the documents as non-relevant, and therefore, found none of the relevant documents, would have 99% accuracy, belying the failure of that search or review effort to identify a single relevant document.”
Many vendors report that their tools boast 99% accuracy. This word should obviously be taken lightly, or at least within the proper context.
Megan Feil, December 21, 2012
Sponsored by ArnoldIT.com, developer of Augmentext
Comments
-
- Subscribe to Beyond Search
Feature archive
News archive
-
