DATA MINING AND MACHINE LEARNING. PREDICTIVE TECHNIQUES: REGRESSION, GENERALIZED LINEAR MODELS, SUPPORT VECTOR MACHINE AND NEURAL NETWORKS
()
About this ebook
Read more from César Pérez López
DATA MINING and MACHINE LEARNING. PREDICTIVE TECHNIQUES: ENSEMBLE METHODS, BOOSTING, BAGGING, RANDOM FOREST, DECISION TREES and REGRESSION TREES.: Examples with MATLAB Rating: 0 out of 5 stars0 ratingsDEEP LEARNING TECHNIQUES: CLUSTER ANALYSIS and PATTERN RECOGNITION with NEURAL NETWORKS. Examples with MATLAB Rating: 0 out of 5 stars0 ratingsDATA MINING and MACHINE LEARNING: CLUSTER ANALYSIS and kNN CLASSIFIERS. Examples with MATLAB Rating: 0 out of 5 stars0 ratings
Related to DATA MINING AND MACHINE LEARNING. PREDICTIVE TECHNIQUES
Related ebooks
MATLAB Machine Learning Recipes: A Problem-Solution Approach Rating: 0 out of 5 stars0 ratingsAdvanced Forecasting with Python: With State-of-the-Art-Models Including LSTMs, Facebook’s Prophet, and Amazon’s DeepAR Rating: 0 out of 5 stars0 ratingsSoftware Modeling A Complete Guide - 2020 Edition Rating: 0 out of 5 stars0 ratingsIntroduction to Reliable and Secure Distributed Programming Rating: 0 out of 5 stars0 ratingsFoundations of Data Intensive Applications: Large Scale Data Analytics under the Hood Rating: 0 out of 5 stars0 ratingsReadings in Computer Vision: Issues, Problem, Principles, and Paradigms Rating: 0 out of 5 stars0 ratingsEmbedded Systems Design using the Rabbit 3000 Microprocessor: Interfacing, Networking, and Application Development Rating: 0 out of 5 stars0 ratingsA Primer on Statistical Distributions Rating: 0 out of 5 stars0 ratingsBeginning T-SQL Rating: 0 out of 5 stars0 ratingsGraphics Gems III (IBM Version): Ibm Version Rating: 3 out of 5 stars3/5Bayesian Optimization and Data Science Rating: 0 out of 5 stars0 ratingsFuzzy Modeling and Genetic Algorithms for Data Mining and Exploration Rating: 5 out of 5 stars5/5Diffuse Algorithms for Neural and Neuro-Fuzzy Networks: With Applications in Control Engineering and Signal Processing Rating: 0 out of 5 stars0 ratingsHashing Rating: 0 out of 5 stars0 ratingsMachine Learning: A Bayesian and Optimization Perspective Rating: 3 out of 5 stars3/5Data Modeling Fundamentals: A Practical Guide for IT Professionals Rating: 0 out of 5 stars0 ratingsSingular Spectrum Analysis for Time Series Rating: 0 out of 5 stars0 ratingsDeploy Machine Learning Models to Production: With Flask, Streamlit, Docker, and Kubernetes on Google Cloud Platform Rating: 0 out of 5 stars0 ratingsStatistics for Experimentalists Rating: 0 out of 5 stars0 ratingsOptimization Techniques in Statistics Rating: 0 out of 5 stars0 ratingsMarkov Processes: An Introduction for Physical Scientists Rating: 1 out of 5 stars1/5Hands-on Time Series Analysis with Python: From Basics to Bleeding Edge Techniques Rating: 5 out of 5 stars5/5Excel Custom Functions: Straight to the Point Rating: 0 out of 5 stars0 ratingsMachine Learning for Beginners: Learn to Build Machine Learning Systems Using Python (English Edition) Rating: 0 out of 5 stars0 ratingsQuantum Theory of Collective Phenomena Rating: 0 out of 5 stars0 ratingsJulia for Data Analysis Rating: 0 out of 5 stars0 ratingsArchitecting CSS: The Programmer’s Guide to Effective Style Sheets Rating: 0 out of 5 stars0 ratingsComplex Binary Number System: Algorithms and Circuits Rating: 0 out of 5 stars0 ratingsMachine Learning - A Comprehensive, Step-by-Step Guide to Intermediate Concepts and Techniques in Machine Learning: 2 Rating: 0 out of 5 stars0 ratings
Applications & Software For You
Adobe Photoshop: A Complete Course and Compendium of Features Rating: 5 out of 5 stars5/5The Best Hacking Tricks for Beginners Rating: 4 out of 5 stars4/5Mastering ChatGPT Rating: 0 out of 5 stars0 ratingsKodi User Manual: Watch Unlimited Movies & TV shows for free on Your PC, Mac or Android Devices Rating: 0 out of 5 stars0 ratingsYouTube Channels For Dummies Rating: 3 out of 5 stars3/5How to Create Cpn Numbers the Right way: A Step by Step Guide to Creating cpn Numbers Legally Rating: 4 out of 5 stars4/5Adobe Illustrator: A Complete Course and Compendium of Features Rating: 0 out of 5 stars0 ratingsBlender 3D Basics Beginner's Guide Second Edition Rating: 5 out of 5 stars5/5Adobe Premiere Pro: A Complete Course and Compendium of Features Rating: 0 out of 5 stars0 ratings2022 Adobe® Premiere Pro Guide For Filmmakers and YouTubers Rating: 5 out of 5 stars5/5iPhone Photography For Dummies Rating: 0 out of 5 stars0 ratingsLogic Pro X For Dummies Rating: 0 out of 5 stars0 ratingsAffinity Photo How To Rating: 0 out of 5 stars0 ratingsFL Studio Cookbook Rating: 4 out of 5 stars4/5Learn to Code. Get a Job. The Ultimate Guide to Learning and Getting Hired as a Developer. Rating: 5 out of 5 stars5/5iPhone Photography: A Ridiculously Simple Guide To Taking Photos With Your iPhone Rating: 0 out of 5 stars0 ratingsAdobe InDesign CC: A Complete Course and Compendium of Features Rating: 0 out of 5 stars0 ratingsHilarious Jokes for Minecrafters: Mobs, Creepers, Skeletons, and More Rating: 1 out of 5 stars1/5GarageBand Basics: The Complete Guide to GarageBand: Music Rating: 0 out of 5 stars0 ratingsCanon EOS Rebel T3/1100D For Dummies Rating: 5 out of 5 stars5/5Six Figure Blogging In 3 Months Rating: 4 out of 5 stars4/5Excel : The Ultimate Comprehensive Step-By-Step Guide to the Basics of Excel Programming: 1 Rating: 5 out of 5 stars5/5iPhone X Hacks, Tips and Tricks: Discover 101 Awesome Tips and Tricks for iPhone XS, XS Max and iPhone X Rating: 3 out of 5 stars3/5Book Design Made Simple Rating: 0 out of 5 stars0 ratingsGarageBand For Dummies Rating: 5 out of 5 stars5/5The Unofficial Guide to Open Broadcaster Software: OBS: The World's Most Popular Free Live-Streaming Application Rating: 0 out of 5 stars0 ratingsVocal Rescue: Rediscover the Beauty, Power and Freedom in Your Singing Rating: 4 out of 5 stars4/5Mastering QuickBooks 2020: The ultimate guide to bookkeeping and QuickBooks Online Rating: 0 out of 5 stars0 ratings
Reviews for DATA MINING AND MACHINE LEARNING. PREDICTIVE TECHNIQUES
0 ratings0 reviews
Book preview
DATA MINING AND MACHINE LEARNING. PREDICTIVE TECHNIQUES - César Pérez López
DATA MINING AND MACHINE LEARNING. PREDICTIVE TECHNIQUES: REGRESSION, GENERALIZED LINEAR MODELS, SUPPORT VECTOR MACHINE AND NEURAL NETWORKS
Examples with MATLAB
César Pérez López
CONTENTS
DATA MINING TECHNIQUES
1.1 DATA MINING AND MACHNE LEARNING INTRODUCTION
1.1.1 Data Mining and Machine Learning Techniques with Matlab
1.1.2 Train Classification Models in Classification Learner App
1.1.3 Train Regression Models in Regression Learner App
1.1.4 Train Neural Networks for Deep Learning
PREDICTIVE TECHNIQUES. LINEAR REGRESSION
2.1 Linear Regression
2.1.1 Prepare Data
2.1.2 Choose a Fitting Method
2.1.3 Choose a Model or Range of Models
2.1.4 Fit Model to Data
2.1.5 Examine Quality and Adjust the Fitted Model
2.1.6 Predict or Simulate Responses to New Data
2.1.7 Share Fitted Models
2.2 Interpret Linear Regression Results
2.2.1 Load sample data and define predictor variables.
2.2.2 Fit linear regression model.
2.2.3 Perform analysis of variance (ANOVA) for the model.
2.2.4 Decompose ANOVA table for model terms.
2.2.5 Display coefficient confidence intervals.
2.2.6 Perform hypothesis test on coefficients.
PREDICTIVE TECHNIQUES. GENERALIZED LINEAR REGRESSION
3.1 REGRESSION MODELS
3.2 GENERALIZED LINEAR REGRESSION
3.2.1 Prepare Data
3.2.2 Choose Generalized Linear Model and Link Function
3.2.3 Choose Fitting Method and Model
3.2.4 Fit Model to Data
3.2.5 Examine Quality and Adjust the Fitted Model
3.2.6 Plots to Understand Predictor Effects and How to Modify a Model
3.2.7 Predict or Simulate Responses to New Data
3.2.8 Share Fitted Models
3.3 Train Logistic Regression Classifiers Using Classification Learner App
PREDICTIVE TECHNIQUES. SUPPORT VECTOR MACHINE REGRESSION (SVR) AND GAUSSIAN PROCCES REGRESSION (GPR)
4.1 SUPPORT VECTOR MACHINE REGRESSION
4.1.1 fitrsvm
4.1.2 predict
4.1.3 fitrlinear
4.1.4 predict
4.2 Gaussian Process Regression Models
4.2.1 fitrgp
4.2.2 predict
4.2.3 postFitStatistics
PREDICTIVE MODELS WITH NEURAL NETWORKS: FITTING NEURAL NETWORKS. FIT REGRESSION MODELS
5.1 NEURAL NETWORK TOOLBOX (DEEP LEARNING TOOLBOX)
5.2 Using Neural Network Toolbox
5.3 Automatic Script Generation
5.4 Neural Network Toolbox Applications
5.5 Neural Network Design Steps
5.6 Function fitting neural network. EXAMPLES
5.6.1 Construct and Train a Function Fitting Network
5.6.2 Create and train Feedforward Neural Network
5.6.3 Create and Train a Cascade Network
5.7 Network performance
5.7.1 Description
5.7.2 Examples
5.8 Fit Regression Model and Plot Fitted Values versus Targets. EXAMPLES
5.8.1 Description
5.8.2 Examples
5.9 Plot Output and Target Values. EXAMPLES
5.9.1 Description
5.9.2 Examples
5.10 Plot Training State Values. EXAMPLES
5.11 Plot Performances. EXAMPLES
5.12 Plot Histogram of Error Values. EXAMPLES
5.12.1 Syntax
5.12.2 Description
5.12.3 Examples
5.13 Generate MATLAB function for simulating neural network. EXAMPLES
5.13.1 Create Functions from Static Neural Network
5.13.2 Create Functions from Dynamic Neural Network
5.14 A COMPLETE EXAMPLE: House Price Estimation
5.14.1 The Problem: Estimate House Values
5.14.2 Why Neural Networks?
5.14.3 Preparing the Data
5.14.4 Fitting a Function with a Neural Network
5.14.5 Testing the Neural Network
PREDICTIVE MODELS WITH NEURAL NETWORKS: FIT DATA WITH A NEURAL NETWORK. GRAPHICAL INTERFACE
6.1 INTRODUCTION
6.2 Using the Neural Network Fitting Tool
6.3 Using Command-Line Functions
PREDICTIVE MODELS: Neural Network Time-Series Prediction and Modeling. GRAPHICAL INTERFACE
7.1 INTRODUCTION
7.2 Using the Neural Network Time Series Tool
7.3 Using Command-Line Functions
DATA MINING TECHNIQUES
The availability of large volumes of data and the generalized use of computer tools has transformed research and data analysis, orienting it towards certain specialized techniques encompassed under the generic name of Analytics that includes Multivariate Data Analysis (MDA), Data Mining, Machine Learning and other Business Intelligence techniques.
Data Mining and Machine Learning can be defined as a process of discovering new and significant relationships, patterns and trends when examining large amounts of data. The techniques of Data Mining and Machine Learning pursue the automatic discovery of the knowledge contained in the information stored in an orderly manner in large databases. These techniques aim to discover patterns, profiles and trends through the analysis of data using advanced statistical techniques of multivariate data analysis.
The goal is to allow the researcher-analyst to find a useful solution to the problem raised through a better understanding of the existing data.
Data Mining and Machine Learning uses two types of techniques: predictive techniques (supervised techniques), which trains a model on known input and output data so that it can predict future outputs, and descriptive techniques (unsupervised techniques), which finds hidden patterns or intrinsic structures in input data.
The aim of predictive techniques is to build a model that makes predictions based on evidence in the presence of uncertainty. A predictive algorithm takes a known set of input data and known responses to the data (output) and trains a model to generate reasonable predictions for the response to new data. Predictive techniques uses classification and regression techniques to develop predictive models.
Classification techniques predict categorical responses, for example, whether an email is genuine or spam, or whether a tumor is cancerous or benign. Classification models classify input data into categories. Typical applications include medical imaging, image and speech recognition, and credit scoring.
Regression techniques predict continuous responses, for example, changes in temperature or fluctuations in power demand. Typical applications include electricity load forecasting and algorithmic trading. This book develops predictive regression techniques
Descriptive techniques finds hidden patterns or intrinsic structures in data. It is used to draw inferences from datasets consisting of input data without labeled responses. Clustering is the most common descriptive technique. It is used for exploratory data analysis to find hidden patterns or groupings in data. Applications for clustering include gene sequence analysis, market research, and object recognition.
MATLAB provides tools to help you try out a variety of Data Mining and Machine Learning models and choose the best. To find MATLAB apps and functions to help you solve Data Mining tasks, consult the following table. Some Data Mining tasks are made easier by using apps, and others use command-line features.
The following systematic Data Mining nd Machine Learning workflow can help you tackle Data Mining and Machine Learning challenges. You can complete the entire workflow in MATLAB.
Descripción: http://es.mathworks.com/help/stats/machinelearningoverviewworkflow.jpgTo integrate the best trained model into a production system, you can deploy Statistics and Machine Learning Toolbox machine learning models using MATLAB Compiler. For many models, you can generate C-code for prediction using MATLAB Coder.
Use the Classification Learner app to train models to classify data using predictive Data Miming techniques. The app lets you explore predictive Data Mining interactively using various classifiers.
Automatically train a selection of models and help you choose the best model. Model types include decision trees, discriminant analysis, support vector machines, logistic regression, nearest neighbors, and ensemble classification.
Explore your data, select features, and visualize results.
Export models to the workspace to make predictions with new data.
Generate MATLAB code from the app to create scripts, train with new data, work with huge data sets, or modify the code for further analysis.
By default, the app protects against overfitting by applying cross-validation. Alternatively, you can choose holdout validation.
Descripción: http://es.mathworks.com/help/stats/mlapp_overview.pngFor more options, you can use the command-line interface. See Classification.
Use the Regression Learner app to train models to predict continuous data using predicte Data Mining. The app lets you explore predictive Data Mininig techniques interactively using various regression models.
Automatically train a selection of models and help you choose the best model. Model types include linear regression models, regression trees, Gaussian process regression models, support vector machines, and ensembles of regression trees.
Explore your data, select features, and visualize results.
Export models to the workspace to make predictions with new data.
Generate MATLAB code from the app to create scripts, train with new data, work with huge data sets, or modify the code for further analysis.
By default, the app protects against overfitting by applying cross-validation. Alternatively, you can choose holdout validation.
Descripción: http://es.mathworks.com/help/stats/regressionlearneroverview17a.pngNeural Network Toolbox (Deep Learning Toolbox from version 18) enables you to perform deep learning with convolutional neural networks for classification, regression, feature extraction, and transfer learning. The toolbox provides simple MATLAB commands for creating and interconnecting the layers of a deep neural network. Examples and pretrained networks make it easy to use MATLAB for deep learning, even without extensive knowledge of advanced computer vision algorithms or neural networks.
.
PREDICTIVE TECHNIQUES. LINEAR REGRESSION
To begin fitting a regression, put your data into a form that fitting functions expect. All regression techniques begin with input data in an array X and response data in a separate vector y, or input data in a table or dataset array tbl and response data as a column in tbl. Each row of the input data represents one observation. Each column represents one predictor (variable).
For a table or dataset array tbl, indicate the response variable with the 'ResponseVar' name-value pair:
mdl = fitlm(tbl,'ResponseVar','BloodPressure');
% or
mdl = fitglm(tbl,'ResponseVar','BloodPressure');
The response variable is the last column by default.
You can use numeric categorical predictors. A categorical predictor is one that takes values from a fixed set of possibilities.
For a numeric array X, indicate the categorical predictors using the 'Categorical' name-value pair. For example, to indicate that predictors 2 and 3 out of six are categorical:
mdl = fitlm(X,y,'Categorical',[2,3]);
% or
mdl = fitglm(X,y,'Categorical',[2,3]);
% or equivalently
mdl = fitlm(X,y,'Categorical',logical([0 1 1 0 0 0]));
For a table or dataset array tbl, fitting functions assume that these data types are categorical:
Logical