You could think of machine learning as (one of)  the main set of tools for performing data mining.

Data Mining is applied Machine Learning.  The former focuses more on the practical aspects of deploying machine learning algorithms on large datasets, whereas the later focuses on modeling, prediction, classification and computational efficiency for large datasets. Also, please check the following question in RG with lots of good answers.

https://www.researchgate.net/post/What_is_the_difference_between_machine_learning_and_data_mining

The best short answer, in my opinion, was given by Sam Roweis during the Machine Learning Summer School in 2005 (fifth slides, minute 8:10):

http://videolectures.net/mlss05au_roweis_pgm/

I'd suggest a couple of interesting readings that are almost 20 years apart:

- Heiki Mannila, "Data mining: machine learning, statistics, and databases"

http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=505910

- Johannes Fernkranz et alii, "Foundations of Rule Learning", Chapter 1.

http://www.springer.com/us/book/9783540751960

http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=505910

http://www.springer.com/us/book/9783540751960

http://videolectures.net/mlss05au_roweis_pgm/

In machine learning ,we hear about supervised learning,unsupervised learning and semi- supervised learning.In data mining ,we learn few techniques such as classification,association,and clustering. As per my knowledge,while introducing  these classification ,association  and clustering techniques  to students ,will link these  concepts to supervised and unsupervised learning techniques.so in  data mining few concepts are inherited from machine learning

You can look at CRISP-DM methodology (Cross-Industry Standard Process for Data Mining). Data mining is a few-stage process. Machine learning special algorithms and methods that are used in one step of this process - modeling (see Fig 3. - table of tasks during Data Mining here http://www.dataprix.net/en/reference-model-crisp-dm )

If you use weka you will have to convert data file into arff format. With R, csv file can be used directly. R has very good functions for visualization of associations rule. Documentation of the package "arules" is available. 

With simple function interface, It is easier to use than weka GUI.