Machine Learning for the Web

图书信息

作者	Andrea Isoni
出版社	Packt Publishing
ISBN	9781785888724
出版时间	2016-07-01
字数	172.9万
分类	进口书,外文原版书,电脑,网络

读书简介

Explore the web and make smarter predictions using Python About This Book Targets two big and prominent markets where sophisticated web apps are of need and importance. Practical examples of building machine learning web application, which are easy to follow and replicate. A comprehensive tutorial on Python libraries and frameworks to get you up and started. Who This Book Is For The book is aimed at upcoming and new data scientists who have little experience with machine learning or users who are interested in and are working on developing smart (predictive) web applications. Knowledge of Django would be beneficial. The reader is expected to have a background in Python programming and good knowledge of statistics. What You Will Learn Get familiar with the fundamental concepts and some of the jargons used in the machine learning community Use tools and techniques to mine data from websites Grasp the core concepts of Django framework Get to know the most useful clustering and classification techniques and implement them in Python Acquire all the necessary knowledge to build a web application with Django Successfully build and deploy a movie recommendation system application using the Django framework in Python In Detail Python is a general purpose and also a comparatively easy to learn programming language. Hence it is the language of choice for data scientists to prototype, visualize, and run data analyses on small and medium-sized data sets. This is a unique book that helps bridge the gap between machine learning and web development. It focuses on the difficulties of implementing predictive analytics in web applications. We focus on the Python language, frameworks, tools, and libraries, showing you how to build a machine learning system. You will explore the core machine learning concepts and then develop and deploy the data into a web application using the Django framework. You will also learn to carry out web, document, and server mining tasks, and build recommendation engines. Later, you will explore Python’s impressive Django framework and will find out how to build a modern simple web app with machine learning features. Style and approach Instead of being overwhelmed with multiple concepts at once, this book provides a step-by-step approach that will guide you through one topic at a time. An intuitive step-by step guide that will focus on one key topic at a time. Building upon the acquired knowledge in each chapter, we will connect the fundamental theory and practical tips by illustrative visualizations and hands-on code examples.

目录

Machine Learning for the Web

Table of Contents

Machine Learning for the Web

Credits

Foreword

About the Author

About the Reviewers

www.PacktPub.com

eBooks, discount offers, and more

Why subscribe?

Preface

What this book covers

What you need for this book

Who this book is for

Conventions

Reader feedback

Customer support

Downloading the example code

Downloading the color images of this book

Errata

Piracy

Questions

1. Introduction to Practical Machine Learning Using Python

General machine-learning concepts

Machine-learning example

Installing and importing a module (library)

Preparing, manipulating and visualizing data – NumPy, pandas and matplotlib tutorials

Using NumPy

Arrays creation

Array manipulations

Array operations

Linear algebra operations

Statistics and mathematical functions

Understanding the pandas module

Exploring data

Manipulate data

Matplotlib tutorial

Scientific libraries used in the book

When to use machine learning

Summary

2. Unsupervised Machine Learning

Clustering algorithms

Distribution methods

Expectation maximization

Mixture of Gaussians

Centroid methods

k-means

Density methods

Mean – shift

Hierarchical methods

Training and comparison of the clustering methods

Dimensionality reduction

Principal Component Analysis (PCA)

PCA example

Singular value decomposition

Summary

3. Supervised Machine Learning

Model error estimation

Generalized linear models

Linear regression

Ridge regression

Lasso regression

Logistic regression

Probabilistic interpretation of generalized linear models

k-nearest neighbours (KNN)

Naive Bayes

Multinomial Naive Bayes

Gaussian Naive Bayes

Decision trees

Support vector machine

Kernel trick

A comparison of methods

Regression problem

Classification problem

Hidden Markov model

A Python example

Summary

4. Web Mining Techniques

Web structure mining

Web crawlers (or spiders)

Indexer

Ranking – PageRank algorithm

Web content mining

Parsing

Natural language processing

Information retrieval models

TF-IDF

Latent Semantic Analysis (LSA)

Doc2Vec (word2vec)

Word2vec – continuous bag of words and skip-gram architectures

Mathematical description of the CBOW model

Doc2Vec extension

Movie review query example

Postprocessing information

Latent Dirichlet allocation

Model

Example

Opinion mining (sentiment analysis)

Summary

5. Recommendation Systems

Utility matrix

Similarities measures

Collaborative Filtering methods

Memory-based Collaborative Filtering

User-based Collaborative Filtering

Item-based Collaborative Filtering

Simplest item-based Collaborative Filtering – slope one

Model-based Collaborative Filtering

Alternative least square (ALS)

Stochastic gradient descent (SGD)

Non-negative matrix factorization (NMF)

Singular value decomposition (SVD)

CBF methods

Item features average method

Regularized linear regression method

Association rules for learning recommendation system

Log-likelihood ratios recommendation system method

Hybrid recommendation systems

Evaluation of the recommendation systems

Root mean square error (RMSE) evaluation

Classification metrics

Summary

6. Getting Started with Django

HTTP – the basics of the GET and POST methods

Installation and server creation

Settings

Writing an app – most important features

Models

URL and views behind HTML web pages

HTML pages

URL declarations and views

Admin

Shell interface

Commands

RESTful application programming interfaces (APIs)

Summary

7. Movie Recommendation System Web Application

Application setup

Models

Commands

User sign up login/logout implementation

Information retrieval system (movies query)

Rating system

Recommendation systems

Admin interface and API

Summary

8. Sentiment Analyser Application for Movie Reviews

Application usage overview

Search engine choice and the application code

Scrapy setup and the application code

Scrapy settings

Scraper

Pipelines

Crawler

Django models

Integrating Django with Scrapy

Commands (sentiment analysis model and delete queries)

Sentiment analysis model loader

Deleting an already performed query

Sentiment reviews analyser – Django views and HTML

PageRank: Django view and the algorithm code

Admin and API

Summary

Index