Book Image

Modern Scala Projects

By : Ilango gurusamy
Book Image

Modern Scala Projects

By: Ilango gurusamy

Overview of this book

Scala is both a functional programming and object-oriented programming language designed to express common programming patterns in a concise, readable, and type-safe way. Complete with step-by-step instructions, Modern Scala Projects will guide you in exploring Scala capabilities and learning best practices. Along the way, you'll build applications for professional contexts while understanding the core tasks and components. You’ll begin with a project for predicting the class of a flower by implementing a simple machine learning model. Next, you'll create a cancer diagnosis classification pipeline, followed by tackling projects delving into stock price prediction, spam filtering, fraud detection, and a recommendation engine. The focus will be on application of ML techniques that classify data and make predictions, with an emphasis on automating data workflows with the Spark ML pipeline API. The book also showcases the best of Scala’s functional libraries and other constructs to help you roll out your own scalable data processing frameworks. By the end of this Scala book, you’ll have a firm foundation in Scala programming and have built some interesting real-world projects to add to your portfolio.
Table of Contents (14 chapters)
Title Page
Copyright and Credits
Packt Upsell
Contributors
Preface
Index

Summary


Fraud detection is not a supervised learning problem. We did not use the random forests algorithm, decision trees, or logistic regression (LR). Instead, we leveraged what is known as a Gaussian Distribution equation to build an algorithm that performed classification, which is really an anomaly detection or identification task. The importance of picking an appropriate Epsilon (error term) to enable the algorithm to find the anomalous samples cannot be overestimated. Otherwise, the algorithm could go off the mark and label non-fraudulent examples as anomalies or outliers that indicate a fraudulent transaction. The point is, tweaking the Epsilon parameter does help with a better fraud detection process.

A good part of the computational power required was devoted to finding the so-called best Epsilon. Computing the best Epsilon was one key part. The other part, of course, was the algorithm itself. This is where Spark helped out a lot. The Spark ecosystem provided us with a powerful environment...