finelybook
Python Tour In Machine Learning
by Md. Akramul Hossain(Author)
ASIN: B099LD5PSX
Publisher Finelybook 出版社: Independently published (July 19, 2021)
Language 语言: English
pages 页数: 141 pages
ISBN-13 书号: 9798536444870
Book Description
An easy and step by step implementation of machine learning problem is shown in python. You will find 6 machine learning problems and their step by step solutions.
Among 6 problems, 4 are supervised learning problems and 2 are unsupervised learning problems.
There are 2 problems taken kaggle competitions to get started as beginners.
The 6 problems are listed below:
Prediction on iris plants dataset (data is taken from sklearn.datasets.load_iris())
California Housing dataset (data is taken from (sklearn.datasets.fetch_california_housing())
Titanic – Machine Learning from Disaster (kaggle link: https://www.kaggle.com/c/titanic)
House Prices Advanced Regression Techniques (kaggle link: https://www.kaggle.com/c/house-prices-advanced-regression-techniques )
An artificial dataset made by sklearn.datasets.make_blobs() to understand unsupervised learning
Market basket analysis (kaggle link: https://www.kaggle.com/vjchoudhary7/customer-
segmentation-tutorial-in-python )
In chapter 1, some basic machine learning concepts is defined easily. In chapter 2, popular used python libraries is introduced. How to install, how to use etc. In chapter 3, Implementation of ML classification technique in iris plants dataset. In chapter 4, Implementation of ML regression technique in california housing dataset. In chapter 5, Prediction of survived and dead based on Titanic - Machine Learning from Disaster data. In chapter 6, Training on House Prices - Advanced Regression Techniques dataset. In chapter 7, A KMeans clustering model is built on artificial dataset to understand unsupervised learning. In chapter 8, Customer segmentation is performed by KMeans clustering technique.
The following steps are implemented step by step as necessary in each problem:
Data Preprocessing [Checking data leakage, Handling Categorical variables, Handling missing values, Handling class imbalance]
Building model and prediction
Cross validation
Various Evaluation techniques
Besides these, best feature selection technique, plotting decision region boundary etc will be found also.
Hope, you will love this book. If you have any questions or suggestions regarding this book, please let me know at my email address ikraminf.mat@gmail.com.
