Using Machine Learning Algorithms with KNIME/Apache Spark’s MLLib/TensorFlow.

Rescheduling to June 15th since our presenter has some family matters to attend to.

Bio:

Jimmy Koyan is a programmer and data engineer working at Big Q Labs a data engineering company that he founded as of October 2015, he develops statistical models using Supervised Machine learning algorithms to predict future outcomes, most recently he developed a model to predict cashier fraud in QSR stores (Quick Service Restaurants). The System was built using Open Source Frameworks and Tools such as KNIME, Kibana , Apache Spark’s MLLib and Apache Hadoop. He has a degree in Electronics Engineering and post graduate diploma in Advance Computing from CDAC (Center for Development of Advanced Computing - Credit with creating the first Super-Computer PARAM in India).

Talk Abstract:

Machine learning algorithms are broadly classified under two categories, Supervised and Unsupervised Classification. We will take a look at these two categories.
Naive Bayes is a machine learning algorithm, that uses probabilistic inference to predict an outcome of an event. Jimmy will breakdown the mechanics of the algorithm and then look at using open source frameworks and tools to solve some real world examples.
The talk is aimed at Programmers who are learning the ropes with Machine Learning, we take a deep dive by starting off with KNIME which is an Open Source Analytics platform, the IDE supports drag and drop of components, Users can create, test and visualize a statistical model in a matter of minutes without any programming. Next we will construct a Naive Bayes model using Python and the scikit (sklearn) framework, Finally we will solve a problem using Apache Spark’s MLLib (Machine learning library).
In closing we will look at Neural Networks and Deep Learning and to that we will explore a few examples with TensorFlow. (TensorFlow is an Open Source Software Library for Machine Intelligence, originally developed at Google).