pattern.en sentiment analysis accuracy percentage

[Daan Harsya]

First thing first, I'm new in this field of (data mining) research. I think I only have the basic knowledge. I hope I can gain more insight here.

My research is sentiment analysis for stock tweet. I use TextBlob Sentiment Analyzers, which based on Pattern. I have done it, I've got the sentiment and subjectivity score. 

So, It's been known in text mining/sentiment analysis research, there are various method such as naive bayes, SVM, maximum enthropy and etc. In TextBlob, we can choose to do sentiment analysis using default library (pattern library) and NaiveBayesAnalyzer. I choose pattern library. Eventho I found out later that both have been tested from movie review corpus. 

So, my first question is; what is the name of the method used in pattern.en for sentiment analysis? Is it Naive Bayes? What is the different between pattern library and NaiveBayesAnalyzer (Is it the same? considering both were trained in movie reviews corpus. CMIIW)

Second, in pattern.en for Sentiment analysis, it said that the accuracy is about 75% for movie reviews. What if I want to do sentiment analysis for stock tweet (or for something else?) What is the accuracy? Or how can I test the accuracy? 

I hope I make my self clear. 

Thank you :)