Python Social Media Analytics电子书

Title Page

Copyright

Python Social Media Analytics

Credits

About the Authors

Acknowledgments

About the Reviewer

www.PacktPub.com

Why subscribe?

Customer Feedback

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

Errata

Piracy

Questions

Introduction to the Latest Social Media Landscape and Importance

Introducing social graph

Notion of influence

Social impacts

Platforms on platform

Delving into social data

Understanding semantics

Defining the semantic web

Exploring social data applications

Understanding the process

Working environment

Defining Python

Selecting an IDE

Illustrating Git

Getting the data

Defining API

Scraping and crawling

Analyzing the data

Brief introduction to machine learning

Techniques for social media analysis

Setting up data structure libraries

Visualizing the data

Getting started with the toolset

Summary

Harnessing Social Data - Connecting, Capturing, and Cleaning

APIs in a nutshell

Different types of API

RESTful API

Stream API

Advantages of social media APIs

Limitations of social media APIs

Connecting principles of APIs

Introduction to authentication techniques

What is OAuth?

User authentication

Application authentication

Why do we need to use OAuth?

Connecting to social network platforms without OAuth

OAuth1 and OAuth2

Practical usage of OAuth

Parsing API outputs

Twitter

Creating application

Selecting the endpoint

Using requests to connect

Facebook

Creating an app and getting an access token

Selecting the endpoint

Connect to the API

GitHub

Obtaining OAuth tokens programmatically

Selecting the endpoint

Connecting to the API

YouTube

Creating an application and obtaining an access token programmatically

Selecting the endpoint

Connecting to the API

Pinterest

Creating an application

Selecting the endpoint

Connecting to the API

Basic cleaning techniques

Data type and encoding

Structure of data

Pre-processing and text normalization

Duplicate removal

MongoDB to store and access social data

Installing MongoDB

Setting up the environment

Starting MongoDB

MongoDB using Python

Summary

Uncovering Brand Activity, Popularity, and Emotions on Facebook

Facebook brand page

The Facebook API

Project planning

Scope and process

Data type

Analysis

Step 1 – data extraction

Step 2 – data pull

Step 3 – feature extraction

Step 4 – content analysis

Keywords

Extracting verbatims for keywords

User keywords

Brand posts

User hashtags

Noun phrases

Brand posts

User comments

Detecting trends in time series

Maximum shares

Brand posts

User comments

Maximum likes

Brand posts

Comments

Uncovering emotions

How to extract emotions?

Introducing the Alchemy API

Connecting to the Alchemy API

Setting up an application

Applying Alchemy API

How can brands benefit from it?

Summary

Analyzing Twitter Using Sentiment Analysis and Entity Recognition

Scope and process

Getting the data

Getting Twitter API keys

Data extraction

REST API Search endpoint

Rate Limits

Streaming API

Data pull

Data cleaning

Sentiment analysis

Customized sentiment analysis

Labeling the data

Creating the model

Model performance evaluation and cross-validation

Confusion matrix

K-fold cross-validation

Named entity recognition

Installing NER

Combining NER and sentiment analysis

Summary

Campaigns and Consumer Reaction Analytics on YouTube – Structured and Unstructured

Scope and process

Getting the data

How to get a YouTube API key

Data pull

Data processing

Data analysis

Sentiment analysis in time

Sentiment by weekday

Comments in time

Number of comments by weekday

Summary

The Next Great Technology – Trends Mining on GitHub

Scope and process

Getting the data

Rate Limits

Connection to GitHub

Data pull

Data processing

Textual data

Numerical data

Data analysis

Top technologies

Programming languages

Programming languages used in top technologies

Top repositories by technology

Comparison of technologies in terms of forks, open issues, size, and watchers count

Forks versus open issues

Forks versus size

Forks versus watchers

Open issues versus Size

Open issues versus Watchers

Size versus watchers

Summary

Scraping and Extracting Conversational Topics on Internet Forums

Scope and process

Getting the data

Introduction to scraping

Scrapy framework

How it works

Related tools

Creating a project

Creating spiders

Teamspeed forum spider

Data pull and pre-processing

Data cleaning

Part-of-speech extraction

Data analysis

Introduction to topic models

Latent Dirichlet Allocation

Applying LDA to forum conversations

Topic interpretation

Summary

Demystifying Pinterest through Network Analysis of Users Interests

Scope and process

Getting the data

Pinterest API

Step 1 - creating an application and obtaining app ID and app secret

Step 2 - getting your authorization code (access code)

Step 3 - exchanging the access code for an access token

Step 4 - testing the connection

Getting Pinterest API data

Scraping Pinterest search results

Building a scraper with Selenium

Scraping time constraints

Data pull and pre-processing

Pinterest API data

Bigram extraction

Building a graph

Pinterest search results data

Bigram extraction

Building a graph

Data analysis

Understanding relationships between our own topics

Finding influencers

Conclusions

Community structure

Summary

Social Data Analytics at Scale – Spark and Amazon Web Services

Different scaling methods and platforms

Parallel computing

Distributed computing with Celery

Celery multiple node deployment

Distributed computing with Spark

Text mining With Spark

Topic models at scale

Spark on the Cloud – Amazon Elastic MapReduce

Summary