Building an Effective Data Science Practice: A Framework to Bootstrap and Manage a Successful Data Science Practice

© The Author(s), under exclusive license to APress Media, LLC, part of Springer Nature 2022

V. Raina, S. KrishnamurthyBuilding an Effective Data Science Practicehttps://doi.org/10.1007/978-1-4842-7419-4_4

4. Classification

Vineet Raina¹   and Srinath Krishnamurthy¹

(1)

Pune, India

 

Let’s begin with a common class of problems called classification problems. A classification problem requires you to infer/predict the class/category to which a new observation belongs based on values of some observed attributes. For example, infer whether a mail is “Spam” or “Regular” based on the body of the mail, sender’s email address, etc.; infer whether a digital payment transaction is “Fraud” or “Non-Fraud” based on the details of the transaction like the location ...