Course

Objective:

Learning Outcomes:

• use data to detect intrusions, train classification models and evaluate them, and perform multiclass classification
• identify use cases for the application of AI for operations and supply chain management
• use machine learning techniques to detect machine failures, load and pre-process data using Azure Machine Learning, and build and evaluate a logistic regression model to predict machine failure
• improve the performance of a failure detection model, use SMOTE to generate records of minority classes, and illustrate how to perform hyperparameter tuning to improve model performance
• analyze academic papers from the domains of marketing and retail, focusing on predicting responses to social media posts, applying market basket analysis and customer segmentation, and predicting out-of-stock (OOS) status in retail locations
• predict retail sales and demand using regression models, customer lifetime value using regression and feature selection, and contrast the performance of different regression algorithms and feature selection techniques
• predict customer purchase intentions using classification, customer responses to marketing campaigns, and apply feature selection and contrast performance of evaluation metrics for classification models
• implement k-means, agglomerative, and DBSCAN clustering, tune hyperparameters of clustering algorithms and contrast their performance, and implement association rules mining for market basket analysis
• collate and analyze academic papers on machine learning (ML) in agriculture, summarize problem categories and solution constructs, and identify common recurring themes in research
• analyze agricultural data using scikit-learn and Microsoft Azure Machine Learning Studio, identify attribute relationships, and perform classification based on type using machine learning (ML) models
• implement clustering to soil data, recognize links between clusters identified by ML algorithms and crops cultivated in those clusters, and differentiate k-means and agglomerative clustering
• apply regression techniques to predict agricultural yields, comprehend real-world and statistical relationships in data, and differentiate linear (OLS), lasso, ridge, and other regression models
• use the right evaluation metrics for classification models, and study existing research for intrusion detection and malware detection

General Topics: