Freie Universität Berlin (FU BERLIN)

This document outlines foundational and advanced machine learning concepts tailored for bioinformatics. It begins by contrasting supervised methods, such as linear regression and Naive Bayes, with unsupervised techniques like K-means clustering and expectation maximization. The text then progresses to complex probabilistic structures, including Hidden Markov Models and Bayesian Networks. A significant portion explores deep learning, detailing artificial neural networks, convolutional networks, a...

his comprehensive document provides an extensive overview of fundamental machine learning and modern data science concepts. It clearly contrasts modern data-driven approaches with classical hypothesis-driven experimentation. Key topics include essential data preprocessing and visualization techniques, emphasizing dimensionality reduction methods like PCA, MDS, and t-SNE. The text deeply explores anomaly detection and clustering algorithms, such as K-means and Spectral Clustering. It also covers ...

This document provides a comprehensive overview of Database Management Systems for data science. It contrasts file systems with databases, introducing schemas, instances, and architecture. It covers relational design, SQL, functional dependencies, and normalization to prevent anomalies. The text explores analytical processing via OLAP, data warehousing, and multidimensional cubes. Additionally, it details physical storage, B+-tree indexing, and transaction management, emphasizing ACID properties...

structured and comprehensive overview of machine learning, covering foundational concepts, data handling, preprocessing, visualization, dimensionality reduction, anomaly detection, clustering, supervised learning, model evaluation, and advanced techniques like kernel methods and regularization.