Helpful Resources for ISYE 6501: Intro to Analytics Modeling — Georgia Tech’s OMSCS
This is a deep dive into all the homework resources that helped me secure an A in the ISYE 6501: Intro to Analytics Modeling course. If you have not yet seen my review and other helpful tips for this course, I strongly recommend giving this article a read. That should help gauge if this course is the best fit in your semester.
Weekly Homework
This course has weekly homework due. Yes, even during exam weeks. That being said, they do give you an opportunity to drop your lowest two grades on the weekly homework.
Below are the resources that became super helpful for each week’s homework and I hope it does the same for you. Don’t let that put you in a box to not explore other resources. What maybe helpful for me, might be something that you already knew and vice-versa. The list is curated to help you on those weeks that are too overwhelming for you because life happens.
Week 1 — Classification
Covered concepts mostly on Support vector machine (SVM) and K-nearest neighbor (KNN) algorithms. I found that doing an R Programming Tutorial truly helped me learn the ins-and-outs of not just the language, but also the R-Studio IDE.
- R Programming Tutorial — Learn the Basics of Statistical Computing — YouTube
- machine learning — What is the influence of C in SVMs with linear kernel? — Cross Validated (stackexchange.com)
- algebra precalculus — Obtain coefficients of a line from 2 points — Mathematics Stack Exchange
- (PDF) Application of Machine Learning Algorithms in Credit Card Default Payment Prediction (researchgate.net)
Week 2 — Validation & Clustering
Covered concepts mostly on supervised and unsupervised learning algorithms and cross-validation. I personally loved the concept of cross-validation and tried implementing it with both a for-loop and an in-built library. Don’t forget to explore in this course!
- machine learning — R: How to split a data frame into training, validation, and test sets? — Stack Overflow
- r — Cross-validating KNN using K-fold — Stack Overflow
- machine learning — Question regarding k fold cross validation for KNN using R — Stack Overflow
- Cross-validation using KNN. Understanding cross-validation, it’s… | by Deepak Jain | Towards Data Science
- K-means Cluster Analysis · UC Business Analytics R Programming Guide (uc-r.github.io)
Week 3 — Basic Data Preparation & Change Detection
Covered concepts of CUSUM, outliers and outlier detection using Box-and-Whisker.
- Grubbs Outlier Test — Testing for Outliers with R — YouTube
- r — How to repeat the Grubbs test and flag the outliers — Stack Overflow
- Normality Test in R — Easy Guides — Wiki — STHDA
- R Tutorial: Testing the extremes with Grubbs’ test
- plot — Overlay normal curve to histogram in R — Stack Overflow
- grubbs.test function — RDocumentation
- Melting and Casting in R — DataScience Made Simple
- A quick tour of qcc (r-project.org)
Week 4 — Time Series
Covered concepts of ARIMA, GARCH and Exponential Smoothing.
- Simple exponential smoothing | Towards Data Science
- 7.3 Holt-Winters’ seasonal method | Forecasting: Principles and Practice (2nd ed) (otexts.com)
- HoltWinters function — RDocumentation
Week 5 — Basic Regression
Covered concepts of Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). These are crucial to understand in order to be successful in this course.
- lm function — RDocumentation
- 4 Examples of Using Linear Regression in Real Life — Statology
- How to get the value of Mean squared error in a linear regression in R — Cross Validated (stackexchange.com)
- Explaining the lm() Summary in R — Learn by Marketing
Week 6 — Advanced Data Preparation
Covered concepts of Box-Cox transformation and how PCA can be used to reduce dimensionality.
- prcomp function — RDocumentation
- PCA example using prcomp in R — Python and R Tips (cmdlinetips.com)
- StatQuest: PCA in R
- Scree Plot. Principal Component Analysis (PCA) is a… | by SANCHITA MANGALE | Medium
- R: Screeplots (ethz.ch)
- dimensionality reduction — How to reverse PCA and reconstruct original variables from several principal components? — Cross Validated (stackexchange.com)
Week 7 — Advanced Regression
Covered concepts of Tree-Based Models.
- Regression Trees · UC Business Analytics R Programming Guide (uc-r.github.io)
- Random Forest Regression in R: Code and Interpretation
- Random Forests
- randomForest function — RDocumentation
- 4 Examples of Using Logistic Regression in Real Life
- How to perform a Logistic Regression in R
Week 8 — Variable Selection
Covered concepts of both greedy and non-greedy variable selection methods. Greedy would include Forward selection, Backward elimination and Stepwise regression. Non-greedy would include Lasso, Elastic net, and Ridge regression.
- Model Selection Essentials in R
- An Introduction to glmnet
- K-Fold Cross Validation in R (Step-by-Step) (statology.org)
- Difference between glmnet() and cv.glmnet() in R?
- Cross-validation for glmnet
Week 9 — Design of Experiments (DoE) & Probability-based Models
Concepts covered for DoE were: A/B Testing, Factorial design, Multi-armed Bandits. Concepts covered for Probability-based models were Poisson, Weibull, Exponential, Geometric, and Binomial.
- Design of Experiments Examples and Applications
- FrF2: Function to provide regular Fractional Factorial 2-level designs
- DoE 64: Building Fractional Designs in R — FrF2 package
- Simulation with Arena — 1
- Computer Labs at GT instead of downloading Arena locally
Week 10 — Missing Data & Optimization
Covered concepts of data imputations and optimization problem types (convex, non-convex, etc.).
- as.numeric in R: How to Convert to Numeric Value (r-lang.com)
- Mode Imputation (How to Impute Categorical Variables Using R) (statisticsglobe.com)
- What is stepAIC in R?. In R, stepAIC is one of the most… | by Ashutosh Tripathi | Medium
Week 11 — Optimization & Advanced Models
Covered concepts of Stochastic Optimization and other advanced models such as Natural Language Process (NLP), Survival Models, Gradient Boosting etc.
- Solving Balanced Diet Problem in Python using PuLP — Machine Learning Geek
- pulp: Pulp classes — PuLP v1.4.6 documentation (coin-or.org)
Remaining Weeks
Week 12–14 were Case Study analysis which required you to pull your knowledge from all the prior weeks, and Week 15 was a course summary.
Other related blogs:
- Overall course review and tips: ISYE 6501: Intro to Analytics Modeling — Georgia Tech’s OMSCS Review & Tips
- Helpful exam strategy to secure an A in the course: Exam Strategy for ISYE 6501: Intro to Analytics Modeling — Georgia Tech’s OMSCS
Thank you for reading this article. If you enjoyed my content, please consider following me to show your support for my work.
