Weekend Learning Series: ML and AI
Altair EmployeeIntroducing Weekend Learning Series by Dr. Mamdouh Refaat
Please feel free to ask any questions that you in the comments below. I hope you enjoy this series.
Data Science Basics: (1) Decision Trees
This article introduces decision trees and their applications. It also explores their advantages versus neural networks.
https://medium.com/@Mamdouh.Refaat/data-science-basics-1-decision-trees-80e870609974
Data Science Basics: (2) Decision Trees for Data Exploration and Transformation
Decision trees are powerful tools for modeling and explaining data. This article focuses on using decision trees to explore data and discover meaningful transformations that could lead to better predictive models.
Data Science Basics: (3) Decision Trees and the Simpson’s Paradox
Simpson’s paradox is a common problem found in many real-life datasets. This article explores the paradox and shows how to demonstrate it and explain it using decision trees.
Data Science Basics: (4) Understanding Statistical Distributions
This article explores four commonly used statistical distributions: the normal, student-t, the Chi-Square, and the F-distribution. It shows their relationship and provides a quick recipe for selecting the appropriate distribution for common situations.
Data Science Basics: (5) the Least Squares vs. the Maximum Likelihood Methods
This article explains the difference between the two main methods used to fit statistical and data science models: the maximum likelihood and the least squares methods. It also shows that the two methods lead to the same model in the case of linear regression under certain conditions.
Data Science Basics: (6) Sampling
This article explores the practical application of sampling in data science, explaining the difference between sampling and partitioning. It also provides a comprehensive overview of the various types of sampling. Additionally, it addresses critical questions about sample size and the validity of the sample in inferring population properties.
https://medium.com/@Mamdouh.Refaat/data-science-basics-6-sampling-3792244f520a
Data Science Basics: (7) The Weight of Evidence (WOE) Transformation
The Weight of Evidence transformation converts categorical and binned continuous variables into numeric values using a binary dependent variable. It solves two problems: (1) it converts categorical variables into numeric values, and (2) it allows the development of standard credit risk scorecards. This article explains the definition of the WOE and how to use it.
Data Science Basics: (8) Correlation Analysis
This article discusses two correlation coefficients: the Pearson Correlation Coefficient and the Spearman rank correlation coefficient. The article also explains when to apply each and the difference between correlation and causation.
https://medium.com/@Mamdouh.Refaat/data-science-basics-7-correlation-analysis-6c4e923b29a9
AI Will Not Kill Us
This article presents a counter response to many articles and presentations that propose that AI will lead to the destruction of civilization and mankind. It discusses the limitations of AI and shows how AI has so far delivered good value to society in many fields.
https://medium.com/@Mamdouh.Refaat/ai-will-not-kill-us-2c5568983b33
Spider-Man Meets Machine Learning … and Physics
This is a fun article that uses Machine Learning and basic mechanics to test some validity of some of the scenes in Spider-Man movies. It extrapolates the properties of real spiders’ silk to calculate the possible strength of Spider-Man’s silk. It also analyzes the mechanics of him swinging between tall buildings.
https://medium.com/@Mamdouh.Refaat/spider-man-meets-machine-learning-and-physics-8681a1b12ed6
The Amazing Log Function
The log function is used in many disciplines, including engineering, science, physics, astronomy, … etc. This article explores the basic mathematical properties of this function and explains why it is so popular.
https://medium.com/@Mamdouh.Refaat/the-amazing-log-function-64d4137cb146
Back to Programming in SAS!
This article discusses why SAS is an important language to learn and use in machine learning. It explores the areas where SAS is more powerful than Python and how it should be used by data scientists and analysts.
https://medium.com/@Mamdouh.Refaat/back-to-sas-2959c5fdf785
Deep Learning Is Regression
In this article, the roots of deep learning and neural networks are traced back to linear regression. It demonstrates how linear regression is in fact the building block of modern deep learning algorithms.
https://medium.com/@Mamdouh.Refaat/deep-learning-is-regression-f0cbf9a0aa07
SLC Bits: (1) Introduction to the Altair Analytics Workbench
The Altair Analytics workbench is a powerful tool for developing and running programs in SAS. It implements the language of SAS using Altair SLC. This article presents an introduction to the software and how to use it.
SLC Bits: (2) Introduction to Data Step Programming
This article introduces the user of the Altair Analytics Workbench and the Altair SLC to the programming of data step. It shows how to create data libraries and datasets, as well as importing data from CSV files. It also shows how to use SQL within SAS programs and explores the representation of missing values in the SAS language.
https://medium.com/@Mamdouh.Refaat/slc-bits-2-introduction-to-data-step-programming-d1f7d467a1bd
Thanks,
Altair Community Team