Data Scientist with Python

Data Scientist with Python

Ziel

1. Basics of data analytics with Python

• Working with the Data Lab
• Basics and concepts in Python
• Presentation of the tools pandas, matplotlib and Seaborn
• Database queries with SQL Alchemy

2. Linear algebra

• Mathematical background
• Basic concepts of linear algebra
• Calculation with vectors and matrices
• Use of the Python library numpy

3. Probability distribution

• Statistics in data science algorithms
• Discrete and continuous distributions
• Versioning code in Git

4. Supervised learning (regression)

• Concepts of supervised learning
• Using linear regression
• Using the Python package sklearn
• Understanding regression models
• Evaluation of predictions
• Bias variance trade-off and regularization
• Measuring the quality of the model

5. Supervised learning (classification)

• Introduction to classification algorithms
• The k-Nearest Neighbors algorithm
• Assessment of classification performance
• Optimization of the parameters
• Splitting the data into training and evaluation sets

6. Unsupervised learning (clustering)

• Concepts of unsupervised learning
• The k-Means algorithm
• Evaluation of the performance metrics
• Alternatives to k-means clustering

7. Unsupervised learning (dimension reduction)

• Reducing dimensions in the data view
• Principal Component Analysis (PCA)
• Creating uncorrelated features from original data
• Introduction to feature engineering

8. Identification and excluding outliers

• Methods for detecting outliers
• Criteria for unusual data points
• Robust measures and reducing the influence of outliers

9. Collecting and merging data

• Reading data from web pages and PDF documents
• Use of regular expressions
• Structuring text data before processing

10. Logistic regression

• Concepts of logistic regression
• Performance metrics for evaluation
• Using non-numerical data in models

11. Decision trees and random forests

• The concept of decision trees
• Combining multiple models into ensembles
• Methods for improving the quality of predictions

12. Support Vector Machines

• Use of Support Vector Machines (SVM)
• Introduction to Natural Language Processing (NLP)
• Text classification with bag-of-words models

13. Neural networks

• Basics of artificial neural networks
• Basics of deep learning
• Deeper understanding of neural network layers

14. Visualization and model interpretation

• Derive and visualize functionalities of models
• Methods for interpretation and visualization
• Apply model-agnostic methods

15. Using distributed databases

• Using the Python package PySpark
• Reading data from distributed databases
• Basics of big data analysis
• Using machine learning algorithms in distributed systems

16. Exercise project

• Work on a comprehensive exercise project independently
• Solve a prediction problem using a larger data set
• Preparation for the final project

17. Final project

• Independent analysis of the data project
• Presentation of results and 1:1 feedback session with mentoring team
• Certificate for Data Analyst with Python