Data Science & AI in Business Analytics
This course offers a fully self-paced learning experience through pre-recorded video lectures that can be accessed and replayed indefinitely, with no live online sessions required. Enrollees receive free access to the Python environment setup, downloadable Excel exercise files, and complete Python scripts for every practical project. The program also includes one attempt at the final course assessment exam upon completion of the video modules.
Course syllabus
- 01
Introduction to Python and Python libraries, exploratory data analysis, technical analysis of share prices using Relative Strength Index, Rate of Change indicators, and Head and Shoulders patterns, risk and return calculations using average return, standard deviation, and cumulative return, loan approval analysis with machine learning, and sentiment analysis of financial news.
- 02
Apply Data Science and AI to employee management using K-means clustering, Gradient Boosting, Logistic Regression, Decision Tree, and Random Forest models. Evaluate model performance using accuracy, ROC AUC, precision, recall, and F1-score, then compare models and select the best performer.
- 03
Measure employee satisfaction using machine learning models, with special focus on SVM methods including Recursive Feature Elimination and Permutation Feature Importance. Compare models, select the best performer, study employee turnover through survival analysis, and apply sentiment analysis to the VAK training method.
- 04
Manage mall customers using Hierarchical clustering, predict e-tailer sales revenue using XGBoost, CatBoost, HGBoost, and NGBoost, compare boosting methods using error metrics, analyze seasonal revenue with Prophet, forecast weekly and yearly e-tailer revenue, and identify good and bad credit using K-means clustering.
- 05
Measure hospital and bank efficiency using Data Envelopment Analysis (DEA). Study GARCH volatility across stock markets, gold, and oil, then measure relationships between these markets using Wavelet Coherence, Cross Spectrum Coherence, and Cross Power Spectrum Coherence models.