Abstract: With the amount of information available on the internet growing at phenomenal rate, research in improving the effectiveness and efficiency of information extraction and knowledge discovery has become crucial. Text mining is one of the most important ways of extracting meaningful information from a large collection of text documents, leaving aside the information which is not useful to the ordinary user. In this study, we propose a method for automatically extracting key elements from a collection of text documents by extracting a set of features using a machine learning technique. We have used the Genetic algorithms for classifying the features those are selected by the machine learning technique. We also compared the results produced by the Genetic algorithm with 10 folds cross-validation at C4.8, Rain Forest, Raintree and NB Tree methods and we have found C4.8 has produced better precision and recall and also the Genetic algorithm is an effective classifier and is quite competitive even though the concept increases in complexity.