Knowledge: Making a Quality Choice

We discuss in this paper how we teach an intro to data science course. Knowledge is a matter of choice and choosing the best course of action is through computer programming this is the way which we consider how to teach this course. We begin with an introduction to big data, data mining, machine learning and data visualization. We use python as the language to clean big data files and to select the right attributes for classification when mining the data. We discuss the Bayes method for classification, then we move to java, specifically Weka, to show how to program the algorithms, some applications are also discussed specifically using Bayes for text processing, we then move to other supervised classification algorithm such as J4.8 and Random Forest trees. And then move to clustering, where we discuss both KMeans and EM algorithms. We discuss truth and knowledge and it is where we use cross validation as the most reasonable approach to find truth based on historical data. Later we discuss supervised and unsupervised Learning, briefly talking about Neural Networks collective and reinforced learning. Briefly talk about current deep learning using CNNs. We finish the course talking about Genetic Algorithms and soft computing.