How to identify the emotions from blog documents using Weka tool Clustering methods

Hi all,

I am new to weka. I need to identify the set of emotions from the blog documents based on Weka tool clustering method. For the emotion detection, I am using different feature sets values (My features are represented as attributes). I have the doubt in the creation of .arff file. How I create the data set for this work. For example, my data set will be as:

@relation emotion

@attribute pos real ->total no of times each part-of-speech(noun,verb,adjective,adverb) occur in the document
@attribute Positive_Words real ->Count of positive words occur in the document
@attribute Negative_Words real ->Count of Negative words occur in the document
@attribute Emotion_Words real ->Count of Emotion words occur in the document
@attribute Emotion_Words real ->Count of Emotion words occur in the document
@attribute First_Sent_Weight real ->Weight given to first sentence in each blog
@attribute Emotions {happy, sad, love, angry, fear, surprise}

@data
//My doubt is how to generate data values by using the above attribute. For given each data set, I need to represent which emotion is it. So, please suggest any idea to represent the data set values. Thanks in advance...