Manual Categorization Has Accuracy Challenges

Often, fully automatic categorization of electronically stored information is insufficient. The alternative, manual classification, has challenges when it comes to accuracy and keeping up with the pace at which ESI is created.

If it's being done for a user's own purposes, such as when someone stores email in a folder, the user will generally be pretty careful. However, users are often unable to keep up with the flow of information, so much remains unclassified.

When manual categorization is done to serve third-party or more nebulous needs, such as compliance or data loss prevention, people are sloppy:

  • They probably aren't properly familiar with the appropriate taxonomy.
  • Even if they understand the different categories on choice, they often just choose the first option so as to minimize the distraction to them.

A good approach to help increase the accuracy of classification is to have the computer make suggestions. Recommind's Decisiv Email is a nice example of such technology.

... David Ferris

Post a comment

You must be logged in to post a comment. To comment, first join our community.