## Logistic Regression and Odds

Logistic regression motivated by analogy to information gain and decision trees. Mutual information, odds, and maximum entropy are all discussed....

From Math4IQB

Logistic regression motivated by analogy to information gain and decision trees. Mutual information, odds, and maximum entropy are all discussed....

From Math4IQB

Decision trees, probability trees, and many other types of trees are useful in machine learning because they are intuitive. Also, they provide reasons for their predictions....

From Math4IQB

Information gain is used to construct decision trees, although Gini impurity is also a possibility. Examples from scikit learn and from the r package rattle....

From Math4IQB

Conditional Probability is sometimes considered difficult to understand, whereas conditional self-information is highly intuitive. In this video, we define conditional information and show how other "conditionals" can be obtained from it. The goal, ultim...

From Math4IQB

Predictive models use data to infer a probability distribution for the process that produced the data. They are literally what a machine "learns" when doing machine learning. This is an introduction to the concept of predictive modeling....

From Math4IQB

Nearest neighbor graphs are used for both unsupervised learning (e.g., clustering) and supervised learning (e.g., imputing missing values). Supervised learning is often necessary, as for instance when statistical bias is an issue....

From Math4IQB

Complex networks can also be "Big Data." When they are, they tend to be scale free....

From Math4IQB

A complex system is a high dimensional interaction of chaotic and regular variables. However, complex systems are not random, and likewise, Big Data -- while containing some randomness -- is better thought of in terms of its complexity....

From Math4IQB

Big Data is complex. In this video, we also introduce and explore pre-processing of data as well as principal components analysis....

From Math4IQB

An application of basic probability to user preferences...

From Math4IQB

A basic tutorial on probability...

From Math4IQB

A description of the structure and expectations of the ETSU MATH 5830 course, Analytics and Predictive Modeling....

From Math4IQB

AIC, Mutual Information, and the importance of entropy and information....

From Math4IQB

Example and a Derivation of Boosting in Machine Learning....

From Math4IQB

Boosting and Bagging with Decision Trees leads to Random Forests -- and the end of this course!...

From Math4IQB

Loss Functions in Machine learning, and why they are not always useful....

From Math4IQB

Demonstration of Togaware rattle as a "rapid prototyping" tool for the data sciences. Often, rattle can be used to get a project up and running; and their excellent logging feature allows you to move from quick prototype to hands on R-coding to implement...

From Math4IQB

Softmax and maximum entropy models, along with KL divergence and a brief look at the AIC criterion....

From Math4IQB

Brief discussion of Entropy, Mutual Information, and relative entropy....

From Math4IQB

The Odds ratio, contingency tables, frequency tables, and examples....

From Math4IQB

Odds, log odds, information, and the beginnings of Logistic Regression....

From Math4IQB

An overview of the midterm for ETSU MATH 5830, Analytics and Predictive Modeling....

From Math4IQB

ROC curves and Area under the Curve....

From Math4IQB

Training, Validation, Testing, and cross-validation. Instability and other issues with decision trees....

From Math4IQB

The confusion matrix, false postives, false negatives, and various measures....

From Math4IQB

Overview of Assignment 2 Notebook for ETSU MATH 5830, Analytics and Predictive Modeling...

From Math4IQB

The normal distribution is extremely important, especially as it relates to the Central Limit theorem. However, it is not a panacea, and it is important when discussing it that we don't gloss over when it applies -- in particular, the crucial assumptio...

From Math4IQB

Simple stochastic model of user preference. Or alternately, a completely random classifier problem....

From Math4IQB

Review of Basic Probability. Good for introductory foundation "to it all". But nothing very advanced....

From Math4IQB

Edited version of same video ( https://youtu.be/ynCkUHPEDOI ) from November 7, 2013. Information, Shannon Entropy, and Maximum Entropy....

From Math4IQB

A review of the ideas in Probability necessary for information theory....

From Math4IQB

Discussion of Small World Networks in preparation for explaining why neighbor-based network algorithms (kNN) are not necessarily good models of real world networks....

From Math4IQB

Nearest neighbor networks and imputation of missing values...

From Math4IQB

Final thoughts on the multiscale nature of complex systems....

From Math4IQB

The relationship of chaotic dynamics to Big Data...

From Math4IQB

Multiscale as related to Fractals and Chaos...

From Math4IQB

The Curse of Dimensionality and the problem of Phantom Information....

From Math4IQB

A first look at classifiers and why classification problems with large, complex data sets may be difficult....

From Math4IQB

Complex systems and big data often deal with emergent properties, and emergence tends to require multiple scales....

From Math4IQB

Discusses how the concept of Big Data is related to complexity and high dimensionality....

From Math4IQB

A detailed description, with examples, of the structure and requirements of ETSU MATH 5830, Analytics and Predictive Modeling...

From Math4IQB

A description of the structure and expectations of the ETSU MATH 5830 course, Analytics and Predictive Modeling....

From Math4IQB

An introduction to the ETSU MATH 5830 "Analytics and Predictive Modeling" course, an introductory data sciences course that focuses on "Big Data."...

From Math4IQB

Video capture of the upload of a couple of jupyter notebooks -- one Python 2.7 and the other R -- and how to work through them using the features of the Jupyter Notebook and the Sage Math Cloud....

From Math4IQB

Sixth in a series of elementary examples of the singular value decomposition....

From Math4IQB

Fifth in a series of elementary examples of the singular value decomposition....

From Math4IQB

Fourth in a series of elementary examples of the singular value decomposition....

From Math4IQB

Third in a series of elementary examples of the singular value decomposition....

From Math4IQB