Help me find the dependency on the data and put them in the spss algorithm?

#1
The topic of my thesis is about bug prediction using data mining algorithms in spss. I found the required data from the article and had to run algorithms on that data. One of the algorithms is binary logistic or multinomial logistic. But my problem is finding the dependency and putting it in the algorithm. How to find and paste dependent data and factors?
 
#2
Click on:

Analyse
Regression
Binary logistic

And click in the 0/1 variabel in the "dependent variable. "
and the explanatory variables in "Covariates"

The you can click on "OK" to run it or try "Paste". Try both of them-
 
#4
If your variable is only coded as 1 and 2, then enter that and see what will happen.

Otherwise, click on
Recode
recode into different variables


But it would be good for you to play around a little. Make up 20 values and enter them (in a new data set) and test what will happen. Get to know the software and what the result would be for different models and data sets.
 
#5
Do you mean to create a new column and add code 0 and 1 as desired and then execute it?
Does this not cause problems in executing pre-created data and replacing it with new data?
 
#6
Do you mean to create a new column and add code 0 and 1 as desired and then execute it?
yes.

Does this not cause problems in executing pre-created data and replacing it with new data?
No. It will create a new variable and the other ones will remain the same.

Try it. Play around with it. Get used to it. Do example data sets where you know the answer and check how it works. Search for many data set on the internet. Run them. Then when you understand the system, then you can conclude something about your own data.
 
#8
Start with the binary. That is easier. You will need to learn this anyway.

But it depends on what the dependent variable is in the data set that you are interested in.

As I said. Train with lots of data. Try explain to a few friends. If they don't understand then you don't know it well enough. Do more training.

Good luck.

1) I took a book
2) I read 500 pages,
3) I went back to 1)

Keep on like that.
 
#9
I created a new column and converted the numbers to 0 and 1. Again, when using the binary algorithm, it displays the following error:
Due to redundancies, degrees of freedom have been reduced for one or more variables.
 
#10
You must have variation in your data. One variable can not just have one value. Try to do descriptives and means and standard deviations.

Make up som data. Test it with different procedures. Means, correlations, linear regression...
 
#14
I don't think you should expect too much from SPSS.

Where do I run navie bayes algorithm and svm algorithm in spss?
Did you mean naive Bayes? (BTW Thomas Bayes was a person so I guess that he should be spelled with B in Bayes in English, right?)

If you are not so familiar with regression and logit models, then maybe you should wait a little bit with the Bayesian stuff.