Training Plan

Introduction to Data Science

• Introduction to data science
• Discovery-based data science
• Basic data types: nominal values, binary values, ordinal values, numeric values, interval values, vb.
• Other data types: time series, text, graph, multimedia, geographical
• General overview of end-to-end data science problem solving process:
  • Identifying the main purpose
  • Turning business problem into data problem
  • Identifying data sources, collecting and storing data
  • Controlling data and checking data quality
  • Data preprocessing steps
  • Building model from data, training machine learning models
  • Evaluating and visualizing model outcomes
  • Putting the model into production and identifying the effects on company performance

Statistics Review for Data Science

• How can we summarize and visualize data sets?
• How can we detect trends and outliers in data sets?
• How can we model relationships between variables?
• How can we deal with missing data?
• How can we perform single and multiple variable regression? Why is it useful?
• How can we select meaningful variables?

Introduction to Data Processing in Python 1

• Quick introduction to Python programming language
  • Basic data structures
  • Statements, conditions, loops
  • File input/output (I/O)
  • Programming exercises

Introduction to Data Processing in Python 2

• Introduction to numpy library, “array” data structure and its basic functions
• Visualization using matplotlib and seaborn libraries
• Introduction to pandas library, “DataFrame” data structure and time series
• Data loading, data analysis ve visualization exercises

Introduction to Machine Learning 1

• What is machine learning?
• What are supervised, semi-supervised and unsupervised learning?
• What are classification, regression, clustering, dimensionality reduction and anomaly detection?
• What are generalization and model capacity?
• How should we divide the data for model training? What are training, validation, and test sets?
• Regression methods
  • Linear regression
• Classification methods 
  • Logistic regression
  • Decision trees

Introduction to Machine Learning 2

• Support vector machines
• Feature extraction methods
• Feature selection methods
• Model combination methods
• Dimensionality reduction methods
  • Principal component analysis
  • Linear discriminant analysis
• Clustering methods
  • K-means clustering
  • Hierarchical clustering

Introduction to Machine Learning 3

• Introduction to deep learning
• What is the reason that deep learning is so successful? Under which circumstances, it is inadequate?
• How do we train deep learning models? (back-propagation algorithm, etc.)
• What are special network architectures?
  • MLP, CNN, RNN, LSTM and GRU models
• Parameters in deep learning models
  • Activation functions
  • Neuron counts in layers
  • Regularization methods

Machine Learning Applications in Python 1

• Introduction to scikit-learn library
• Linear regression
• Logistic regression
• Decision trees
• Model and hyper-parameter selection methods
• Support vector machines
• Feature extraction methods
• Feature selection methods

Machine Learning Applications in Python 2

• Model combination methods
• Tree-based models
• Dimensionality reduction methods
  • Principal component analysis
  • Linear discriminant analysis
• Clustering methods
  • K-means clustering
  • Hierarchical clustering

Machine Learning Applications in Python 3

• Introduction to tensorflow and keras libraries
• MLP models
• CNN models
• RNN models
• LSTM and GRU models