Book Image

F# for Machine Learning Essentials

By : Sudipta Mukherjee
Book Image

F# for Machine Learning Essentials

By: Sudipta Mukherjee

Overview of this book

The F# functional programming language enables developers to write simple code to solve complex problems. With F#, developers create consistent and predictable programs that are easier to test and reuse, simpler to parallelize, and are less prone to bugs. If you want to learn how to use F# to build machine learning systems, then this is the book you want. Starting with an introduction to the several categories on machine learning, you will quickly learn to implement time-tested, supervised learning algorithms. You will gradually move on to solving problems on predicting housing pricing using Regression Analysis. You will then learn to use Accord.NET to implement SVM techniques and clustering. You will also learn to build a recommender system for your e-commerce site from scratch. Finally, you will dive into advanced topics such as implementing neural network algorithms while performing sentiment analysis on your data.

Related Content you might be interested in

Current Title:

Small book image
F# for Machine Learning Essentials
Sudipta Mukherjee
Feb, 2016 | 194 pages
No titles found
Table of Contents (16 chapters)
F# for Machine Learning Essentials
F# for Machine Learning Essentials
Credits
Credits
Foreword
Foreword
About the Author
About the Author
Acknowledgments
Acknowledgments
About the Reviewers
About the Reviewers
www.PacktPub.com
www.PacktPub.com
Preface
Preface
Free Chapter
1
Introduction to Machine Learning
Introduction to Machine Learning
Objective
Why use F#?
Unsupervised learning
Machine learning frameworks
Machine learning for fun and profit
Recognizing handwritten digits – your "Hello World" ML program
Summary
2
Linear Regression
Linear Regression
Objective
Different types of linear regression algorithms
APIs used
The basics of matrices and vectors (a short and sweet refresher)
QR decomposition of a matrix
Linear regression method of least square
Finding linear regression coefficients using F#
Finding the linear regression coefficients using Math.NET
Putting it together with Math.NET and FsPlot
Multiple linear regression
Multiple linear regression and variations using Math.NET
Weighted linear regression
Plotting the result of multiple linear regression
Ridge regression
Multivariate multiple linear regression
Feature scaling
Summary
3
Classification Techniques
Classification Techniques
Objective
Different classification algorithms you will learn
Some interesting things you can do
Understanding logistic regression
Multiclass classification using logistic regression
Multiclass classification using decision trees
Predicting a traffic jam using a decision tree: a case study
Challenge yourself!
Summary
4
Information Retrieval
Information Retrieval
Objective
Different IR algorithms you will learn
What interesting things can you do?
Information retrieval using tf-idf
5
Collaborative Filtering
Collaborative Filtering
Objective
Different classification algorithms you will learn
Vocabulary of collaborative filtering
Baseline predictors
Item-item collaborative filtering
Top-N recommendations
Evaluating recommendations
Ranking accuracy metrics
Working with real movie review data (Movie Lens)
Summary
6
Sentiment Analysis
Sentiment Analysis
Objective
What you will learn
A baseline algorithm for SA using SentiWordNet lexicons
Handling negations
Identifying praise or criticism with sentiment orientation
Pointwise Mutual Information
Using SO-PMI to find sentiment analysis
Summary
7
Anomaly Detection
Anomaly Detection
Objective
Detecting point anomalies using IQR (Interquartile Range)
Detecting point anomalies using Grubb's test
Grubb's test for multivariate data using Mahalanobis distance
Chi-squared statistic to determine anomalies
Detecting anomalies using density estimation
Strategy to convert a collective anomaly to a point anomaly problem
Dealing with categorical data in collective anomalies
Summary
Index
Index

Pointwise Mutual Information

PMI between two words is calculated using the following formula:

represent the number of occurrences of the word word in the entire document collection. The original article that proposed this idea used the number of articles returned for the search word word from the AltaVista search engine. But you can safely use a probability (the number of documents in which the word word appeared divided by the total number of documents). The & operator in refers to the number of documents containing both words word1 and word2 divided by the total number of documents.

The following function finds the probability of the word in a document collection represented by list:

The following function finds the probability of the words w1 and w2 in a document collection represented by list:

The following function calculates the PMI between w1 and w2:

Previous Section
End of Section 7
Next Section