Conceptinmachinelearning

• Dataedaanalysis
  • Tabular Data - Concepts in Machine Learning - Data EDA and Analysis
  • Table of Content
  • 1. Machine Learning Workflow
  • 2. Decisions in analytics are increasingly driven by data and models, and key aspects of our Machine Learning Workflow are getting depend on cleaning data
  • 3. How data can be messy:
  • 4- Dealing with missing data:
  • 5- Dealing with outliers
  • 6. Exploratory Data Analysis (EDA):
  • 7- EDA techniques:
  • 8- Feature Engineering and Variable Transformation:
  • 9- Transformations
  • 10- Variable Selection:
  • 11- Notes:
  • 12- Estimation vs Inference:
  • 13- Machine Learning and Statistical Inference:
  • 14- Parametric vs. Non-parametric;
  • 15- Common Distributions:
  • 16- Frequentist vs. Bayesian Statistics:
  • 17- Recap:
  • 18- Hypothesis Testing:
  • 19- Type1 and Type2 Errors:
  • 20- Hypothesis Terminologies:
  • 21- Business Examples:
  • 22- Significance Level and p-value:
  • 23- F-Statistic:
  • 24- Power and Sample size:
  • 25- Correlation vs Causation:
  • Credits:
• Mlsupervisedclassification
  • Tabular Data - Concepts in Machine Learning - Supervisied Learning - Classification
  • Table of Content
  • One-vs-All technique
  • Choosing Error Metric
  • Confusion Matrix
  • KNN Models:
  • Support Vector Machines (SVM):
  • Decision Trees:
  • MultiClass Classifier
  • Unbalanced Classes
  • Model Interpretability:
  • Credits
• Mlsupervisedregression
  • Tabular Data - Concepts in Machine Learning - Supervisied Learning - Regression
  • Table of Content
  • Traditional Statistical Modeling vs Machine Learning
  • Machine Learning in Context with AI
  • Model: A Learning Algorithm
  • Interpretation and Prediction:
  • R2 metric:
  • Features & Target Transformation
  • Cross Validation
  • Bias-Variance Trade off:
  • Sources of Model Error
  • Regularization
  • Feature Selection
  • Ridge Regression
  • Lasso Regression
  • Elastic Net
  • Recursive Feature Elimination (RFE)
  • Regularization under the hood
  • Credits:
• Mlunsupervised
  • Tabular Data - Concepts in Machine Learning - Unsupervisied Learning
  • Table of Content
  • Description of Unspervised ML
  • Two cases of unsupervised learning
  • The Curse of dimensionality
  • K-means
  • Choosing the right Clustering ALgorithm
  • Clustering Algorithms Comparison
  • Determining The Number of Clusters
  • Dimentionality Reduction Techniques
  • Principal Component Analysis (PCA)
  • Credits