[R] Text Pattern Recognition - Model

[Anshuk Pal Chaudhuri]

Hi,

I have a training dataset which has two columns which has around 70 values.

1.       "PNRNo" whose values like UT768G, CXKA, 4IOI59, 4BV7TW...(typical PNR number patterns)

2.       I have created one more factor variable mentioning (IsPNR) - so all the values are 1 (true)

My first objective is to create a model on this training set which would recognize the text pattern.

Second objective: The  model would then be used to predict IsPNR with new set of test values like "Anshuk", "4EL58S"...as 0 and 1...

Which model would be best for recognizing such kind of pattern and having decent accuracy? I tried naiveBayes, but I don't think it is all doing a good job. Its predicting all the test values as true. I suppose bayes is not meant for this.


Regards,
Anshuk Pal Chaudhuri


	[[alternative HTML version deleted]]