Skip to Main Content
The problem of pattern classification has two highly interactive aspects: (1) the selection of numerical measurements to ‘represent’ the patterns, and (2) the specification of an algorithm to identify patterns, based upon the numerical values of these measurements. The present paper presents the mathematical framework for one attack upon these problems and gives results obtained in some experiments in character recognition.
Note: The Institute of Electrical and Electronics Engineers, Incorporated is distributing this Article with permission of the International Business Machines Corporation (IBM) who is the exclusive owner. The recipient of this Article may not assign, sublicense, lease, rent or otherwise transfer, reproduce, prepare derivative works, publicly display or perform, or distribute the Article.