Logistic Regression, Decision Tree and Neural Network in R

Name: Logistic Regression, Decision Tree and Neural Network in R
Rating: 3.9 (32 reviews)

Created byModeste Atsague

Last updated 8/2017

English

What you'll learn

At the end of this Course, A student will be able to use Predictive analytics ( Decision tree , neural network or Logistic regression) to predict future outcomes. Some areas of application are the following: Actuarial Science, marketing, financial services, insurance, mobility, pharmaceuticals, healthcare, just to name a few

Course content

1 section • 12 lectures • 1h 0m total length

Introduction0:33
Explore how to implement logistic regression, decision trees, and neural networks in R, and uncover their practical value.
How to download R2:28
Learn to download and install R on Mac or Windows by selecting the proper installer, double-clicking to run, and following the continue prompts through the setup.
Welcome2:04
Explore logistic regression, decision trees, and neural networks in R. Learn to build models, interpret a decision tree, and evaluate prediction accuracy for outcomes.
Description of the data4:03
Describe your dataset in R using summary and describe to explore observations, variables, and factors before applying logistic regression, decision trees, and neural networks.
Data transformation2:11
Transform the response by recoding the sex variable to a binary 0/1 in a data frame using an if statement and the dollar-sign notation, preparing the dataset for analysis.
Missing values detection and treatment2:53
Detect and treat missing values in an R dataset by using complete.cases to identify rows with missing data and na.omit to remove them, illustrated with a salary dataset.
Data visualisation3:36
Install and load the R package, then use plot to visualize salary by sex and occupation and interpret the resulting group comparisons.
Training and Testing set4:26
Learn to split data into training and testing sets and build a confusion matrix with R libraries for logistic regression, decision trees, and neural networks.
Decision Tree5:23
Build and interpret a decision tree in R, train and evaluate the model, visualize splits, and predict sex from attributes using the training data.
Logistic regression4:08
Explore logistic regression in R for binary outcomes, estimate predictive probabilities for sex, use a threshold of 0.5 for class prediction, and assess ~90% accuracy on test data.
Neural network26:52
Learn to run a neural network in R for predicting sex from salary data, including library import, data normalization, class encoding, and confusion-matrix evaluation.
Min Max normalization1:51
Min-max normalization scales x to a zero-to-one range, with min zero and max one. A value above 0.5 marks x as 1 for neural network decisions.

Requirements

The only prerequisite for this class is the willingness to learn and some basic knowledge of R but not necessary

Description

In this course, we cover two analytics techniques: Descriptive statistics and Predictive analytics. For the predictive analytic, our main focus is the implementation of a logistic regression model a Decision tree and neural network. We well also see how to interpret our result, compute the prediction accuracy rate, then construct a confusion matrix .

By the end of this course , you will be able to effectively summarize your data , visualize your data , detect and eliminate missing values, predict futures outcomes using analytical techniques described above , construct a confusion matrix, import and export a data.

Who this course is for:

Anyone seeking a career as data scientist, data analyst , finance analyst, statistician , actuary, just to name few