Okay, now we are going into the test classification. So what is test classification? In test classification stage, we usually try to classify or predict a variable. To predict a variable we train our classifier prediction model from training data. The classifier can be a regression or a machine learning algorithm. In training if the algorithm is linear regressions, we are training and calculating the MMC in the y equals mx plus C. After training the classifier or model we then use the model to predict a variable y in a new data set.
To get the accuracy of the model, we usually do evaluation using tempo crossover validation testing. So, in our poll programmer application, we will be using the Stanford NLP classifier. So we can download the Stanford NLP classifier at this web address. Okay, so the low for the dialogue okay. Okay, I open the file. Guy copy this to my D drive.
Okay, then I import these stem for classify into my libraries and jaw folder. Stem file classifier. Okay, then we'll use the following course. Oh, these are a Stanford NLP course you can buy online. All sama forum. So if option equal equal phi e option equal equal phi Yeah, she'll be asking the user for some input.
So enter the properties file and then entering file and test file. System dot O, green line and profit the K stream str in one equals SC dot next System dot out dot print line and find the train. Street stream STI n SC dot next. Then System dot out dot println and test as far as streaming SDI entry equal See? I may want to change this to answer and then optionee Okay. So we will be using the column data classifier.
So, column data classifier. sci fi Yeah. Can you call the class sci fi okay. Call them data classify the property fast. So SDI and so on. So, classifier dot train thought tree should be str.
To me tempo streamlines in object bang. Ghana eatery. She'll be on the surgery UTF eight datum stream stream D the classifier.me did come from lie System dot out dot println libre system No Oh, no green light light brass ah classifier classify the dog get call classifier class on D Okay, now we can run this program is C, KB just by text classification and I will carry the dry stem for NLP stem for classified Yeah, example. So I will choose cheese, cheese balls have on top property then trying to choose cheese dz stock trading choose dz stock pass This is a line this is the predictor variable predict across and then this is the score so for more information you may want to look into stamper NLP classifier. Guy our documentation. So this is an example of how I implement text classification using standby now PK, we've come to the end of chapter six, we have successfully created our own test mining application.
In this chapter we have Kaabah, import test pass, test transformation, test exploration and test classification. So we can actually save this whole project as a Java application or Java. So we do run Okay, so the location of the job is here. When we create and build NetBeans will create this distribution folder. So DC, this is where our jar file is. So when we want to distribute our software we can distribute why Saw inside this folder