Hello All,
I am a SAS newbie and had a question regarding defifining rules for people names. What I am trying to do is write a rule which identifies a person's name from a document. There are few patterns (I am assuming) that have been defined by studying the sample documents. For example. Mr. John Doe (the first two letters with a period after it). Also, if it is a email cc: and then the next two words would be a name. I understand that defining the names are very chanllenging which is why we are not looking at 100% accuracy. But something lie a 70% accuracy would be good. Can somebody help me with how I can proceed with wiriting the rules? Please! Thanks!!!