In this age of Data science where R and Python rule the roost, Let's take a look at another data science tool called Weka. Weka has been around for quite some time and was developed internally at the University of Waikato for research purposes. What makes Weka worth trying is the easy learning curve.
For someone who hasn't coded in a while, Weka with its GUI provides the easiest transition to the world of data science. Being written in Java, those with Java experience can also call the library in their code.
Personally, I had my first chance to Data science when I did a course at the University of Waikato. It was a healthy introduction and gave me a smooth transition to data science.. Later, when I had to tackle bigger issues, walk. So I highly recommend Weka as a learning tool for those looking to delve into the world of data..
Below is the step-by-step learning pedagogy that will help you understand the concepts in a better and more concrete way:
Paso 1 : What is Weka and why use it?
According to Wikipedia :,
Put is a collection of machine learning algorithms for data mining tasks. Algorithms can be applied directly to a data set or can be called from your own Java code. Put contains instruments for preprocessing, classification, regression, grouping, association rules and data visualization.
You may want to take a look at this video de Brandon Weinberg. This video will give you considerable insight into this amazing tool. You may not understand everything through this video, but will surely learn things.
Paso 2: machine configuration
Now that we meet Weka, we can move on to the next stage. To learn more about the tool and the people behind its success, you can take a look at this site about the Weka Project. What's more, you can also download the software and get the latest version for your system from this Link.
Paso 3: learn the basics of Weka
The best way to get started with Weka is by using the MOOCs offered by the University of Waikato. Data Mining with Weka is a reputable course, but it is not available all year round. But nevertheless, do not worry, in such cases one can access the course videos from this Youtube Channel. You can see the official link of this course. here. The data sets to be discussed here can be downloaded from this Link. The page has more links to data sets. Weka uses data in ARFF format. In case the data is not in ARFF format, you can convert them from CSV to ARFF format by taking the help of this video.
Paso 4: data sets
After having tested the data sets provided by the course coordinators, we will test a new data set of DataHack. Since the format would be .csv, convert it to ARFF format, so that we can read it in the Weka interface. After having taken these courses, once you have acquired enough skills to start working and analyzing data sets using the Weka GUI. Those who visited the MOOC link would have seen the course ‘More data mining with Weka’.
Paso 5: more data mining with Weka
Here, some more advanced features of using the software have been discussed. Develop the experience of using the previous course, Thus, is a prerequisite.
YouTube series
In addition to this course, you might want to take a look at this YouTube lecture series from Rushdi Shams.There is a total of 38 conferences. You can omit some of the 2-3 kickoff lectures if you find the content is similar to previous courses. This course has been based on various skills that are complementary to those provided by the previous series.
Weka News
There are some interesting discussions going on in Reddit about Weka. It is advisable to go through the mentioned link to gather news about Weka and how others are using it. This should give sufficient perspective on the next possible step after Weka..
Paso 6: Weka command line
Next step: Hereinafter, we have been relying on using Weka using the Weka GUI. Hereinafter, both courses are GUI based for this purpose, those with experience in JAVA Programming can trust to call Weka from within the JAVA Code. This is useful because when testing or working with large data sets, scripts help automate your work. What's more, since JAVA is used for Hadoop Framework, Weka can also be used for BigData. You can read more using Weka in BigData at here.
Then, Those interested in this aspect of Weka can try this series of lectures by Dr. Noureddin Sadawi. You may want to check out this Weka API tutorial play list what's more. The emphasis is on calling the Weka API from JAVA code, repeats some of the previous concepts, but we use Weka using a command line interface.
Paso 7: Word2Vec Challenge
Having obtained a meaningful insight, now we will see the sentiment analysis. There is a small dataset with a dataset size of around 25 MB. Therefore, these can be processed using the Weka GUI. For data sets of more than 40 MB, we need to use command line method. This discussion it could be useful.
–
This path has been contributed by Abhinav Unnam, who did an internship with us last year. Abhinav is currently on a dual degree course at IIT Roorkee, one of the best engineering universities in India. He started his machine learning journey through Weka and today he enjoys participating in various Kaggle competitions using R and Kaggle.
