Data Science Day 26
When I was cleaning my home, I found a brand new book of Fundamentals of Machine Learning for Predictive Data Analytics. Therefore I decided to read the book and share some exercise problems.
Chapter 1. Machine Learning for Predictive Data Analytics
-
What is predictive data analytics?
Predictive data analytics is a subfield of data analytics that focuses on building and using models that make predictions based on insights/patterns extracted from historical data. To build these models, we use machine learning algorithms to extract patterns from datasets.
Applications:
- Price Prediction: hotel chains, airlines, or online retailers
- Dosage Prediction: how much dose for treatment
- Risk Assessment: Loan/ Insurance policy
- Propensity Modeling: Future customers action
- Diagnosis: doctor, engineer, and scientists make decisions
- Document Classification: automatically filtering spam, sentiment analysis
2. What is supervised machine learning
Supervised machine learning automatically learn the relationship between a set of descriptive features and a target feature from a set of historical instances.
We use Supervised machine learning to build models that can make predictions based on patterns extracted from historical data.
Historical Data(Training set)-> Prediction Model (Machine Learning Algorithm)-> Prediction
3. Machine Learning is often referred to as an ill-posed Problem. What does this mean?
If there is not enough information in the training data to choose a single best model then the inductive machine learning is referred to as an ill-posed problem. Machine Learning algorithms essentially search through all the possible patterns that exists between a set of descriptive features and a target feature to find the best model for the training data. It is possible to find multiple models that are consistent with a certain training set.
Reference:
Book:
http://www.machinelearningbook.com
Answers:
http://machinelearningbook.com/wp-content/uploads/2015/07/FMLPDA_freely_avail_solutions.pdf