Book Image

Python Data Analysis - Third Edition

By : Avinash Navlani, Ivan Idris
5 (1)
Book Image

Python Data Analysis - Third Edition

5 (1)
By: Avinash Navlani, Ivan Idris

Overview of this book

Data analysis enables you to generate value from small and big data by discovering new patterns and trends, and Python is one of the most popular tools for analyzing a wide variety of data. With this book, you’ll get up and running using Python for data analysis by exploring the different phases and methodologies used in data analysis and learning how to use modern libraries from the Python ecosystem to create efficient data pipelines. Starting with the essential statistical and data analysis fundamentals using Python, you’ll perform complex data analysis and modeling, data manipulation, data cleaning, and data visualization using easy-to-follow examples. You’ll then understand how to conduct time series analysis and signal processing using ARMA models. As you advance, you’ll get to grips with smart processing and data analytics using machine learning algorithms such as regression, classification, Principal Component Analysis (PCA), and clustering. In the concluding chapters, you’ll work on real-world examples to analyze textual and image data using natural language processing (NLP) and image analytics techniques, respectively. Finally, the book will demonstrate parallel computing using Dask. By the end of this data analysis book, you’ll be equipped with the skills you need to prepare data for analysis and create meaningful data visualizations for forecasting values from data.
Table of Contents (20 chapters)
Section 1: Foundation for Data Analysis
Section 2: Exploratory Data Analysis and Data Cleaning
Section 3: Deep Dive into Machine Learning
Section 4: NLP, Image Analytics, and Parallel Computing

Spectral clustering

Spectral clustering is a method that employs the spectrum of a similarity matrix. The spectrum of a matrix represents the set of its eigenvalues, and a similarity matrix consists of similarity scores between each data point. It reduces the dimensionality of data before clustering. In other words, we can say that spectral clustering creates a graph of data points, and these points are mapped to a lower dimension and separated into clusters.

A similarity matrix converts data to conquer the lack of convexity in the distribution. For any dataset, the data points could be n-dimensional, and here could be m data points. From these m points, we can create a graph where the points are nodes and the edges are weighted with the similarity between points. A common way to define similarity is with a Gaussian kernel, which is a nonlinear function of Euclidean distance:

The distance of this function ranges from 0 to 1. The fact...