“Introduction to Data Analysis” - course RUB 20,000. from MSU, training 13 weeks. (1.5 months), Date: May 12, 2023.
Miscellaneous / / December 02, 2023
The program is aimed at managers, analysts, business analysts, team leaders, those in need of a brief and accessible presentation of data analysis methods - machine learning methods and neural networks.
Admission Requirements
The program is intended for students who have higher education or are receiving higher education (in the penultimate and final year of study)
Dates: May 12,16,17,19,23,24, 2023
Classes from 17.00 to 20.00
Lecture 1 Entry requirements. Introduction to the program
Statements of objectives
Program overview
Linear algebra terms
Examples of object representation
Rules for working with matrices and vectors at the 1st-2nd year level of a technical university.
Lecture 2 Basic types of models for finding patterns in data
Regression analysis
Data clustering
Simple and generalized decision trees
Data reduction - principal component analysis
Evolutionary algorithms
Neural networks
Lecture 3 Introduction to Data Analysis
Introduction to Data Analysis and Pattern Recognition
Primary data transformation, search for outliers
Regression analysis, rolling control
Decision trees, simple and generalized forms
Lecture 4 Proximity (similarity) of objects. Clusters and their search
Cluster as a connected component of a graph.
Constructing a minimum spanning tree.
Method of K means, simple and generalized versions.
Hierarchical cluster analysis, dendrograms
Lecture 5 Principal component method
Factors and their search, SVD matrix decomposition
Geometric meaning of factors
Regression on factors
Multidimensional scaling
Lecture 6 Advanced Analysis Methods
Evolutionary algorithms – GMDH, genetic
Kernel functions – “signless” data analysis
SVM and support vectors
“When there is little data” – Bootstrap Method
Families of predictive algorithms
"Fuzzy" signs (Fuzzy)
"Fuzzy" classifiers
Lecture 7 Neural networks. Part 1
Perceptron model and its limitations
Classic neural networks, layer of neurons, two types of neurons
Problems solved by neural networks, “Deep learning”
Lecture 8 Neural networks. Part 2
Image Analysis and Convolutional Neural Networks
Neural networks and Feature Engineering
Overfitting problem
Prospects for the development of neural networks
Graphics Processing Units (GPUs).
Lecture 9 Consolidation of knowledge
Repetition of basic material using a practical example
Summarizing
Cumulative credit
The course covers the basics of Cassandra 4-x architecture, development of conceptual, logical and physical data models. Covers all the necessary technical details for using Cassandra for scalable storing data in Java projects, as well as for monitoring, configuration and configuration productivity.
4
51 500 ₽