Welcome to Knowledge, Discovery and Mining. This lecture will be given by Assoc.Prof.Dr.Tugba Taskaya Temizel. The assistant who will be doing the applied portion of the course is Mehmet Ali Akyol.
This page will contain the lab codes for the class as RMarkdown documents. You can go to the Github repository for the full Rmarkdown files to use in your local environment.The entirety of materials will be available through ODTUClass. This website only aims to provide a collection of rendered RMarkdown files.
The syllabus for this year’s lecture is as follows:
- Week 1: Introduction: Review of the Basics (Background) - Discussion about data pre-processing, basics of data analysis techniques.
- Week 2: Data Warehousing
- Week 3: Mining Frequent Patterns, Associations, and Correlations - Efficient and scalable frequent itemset mining methods, Mining various kinds of association rules, Interestingness measures
- Week 4: Classification - Logistic regression, Neural Networks: ADALINE, backpropagation algorithm, radial basis functions, associative networks, recurrent networks
- Week 5: Classification cont. - Support vector machines, Combining models: Ensembles (boosting, bagging), mixture of experts, random forest, XGBoost
- Week 6: Clustering - Similarity and distance measures, self-organizing maps, clustering validation techniques, singular value decomposition
- Week 7: Clustering cont. - Topic Modelling: Latent Dirichlet Allocation (LDA)
- Week 8: Evaluation and Credibility - Error measures, creating baselines, comparison of classifiers
- Week 9: Recommendation Systems - User-based and item-based collaborative systems, matrix factorization techniques
- Week 10: Anomaly Detection Algorithms
- Week 11: Introduction to Information Retrieval and Information Extraction Techniques
Grading
- Assignments (%30)
- Midterm (%25)
- Final (%25)
- Final Project (%20)