Lectures of the Technosphere. 1 semester Introduction to Data Analysis (Spring 2016)
Listen and watch the new compilation of lectures of Technosphere Mail.Ru. This time we present the open course spring course “Introduction to Data Analysis”, where students are introduced to the field of data analysis, the main tools, tasks and methods that any data researcher faces in his work. The course is taught by Evgeny Zavyalov (Mail.Ru Search project analyst, who extracts useful business knowledge from data generated by the search engine and desktop applications), Mikhail Grishin (research programmer from the data analysis department) and Sergey Rybalkin (senior programmer from the Allods Team studio) .
From the first lecture you will learn what data analysis is, what tools are used for data analysis, and how Python works.
A more detailed study of the syntax of the Python language and its use.
')
The lecture covers both standard Python libraries and the libraries most commonly used for data analysis. There is a story about the properties, descriptors, common tasks needed to process data in Python. Evgeny Zavyalov deals with the topic of working with web, mail and websites.
We will talk about the main approaches to data visualization and explanatory data analysis. Examples of the application of previously acquired knowledge using parsing open dataset will be considered. Work will continue with the Numpy, Pandas libraries. In the same lecture, the acquaintance with the R language, as a possible alternative to a bunch of Python and libraries, will begin.
Talk about the advantages of the R language (about disadvantages, see the fourth lecture), which came out of the academic environment, but came close to the possibilities with Python (and inspired the latter for some borrowing). In the West, the free language R is a de facto standard that is not yet so widely known in Russia.
Recall the main theorems, the probability, the laws of the distribution of random variables, the estimation problem. Not only let's touch on fundamental knowledge, but also look at the ways of their practical application.
At the second lesson on statistics, methods for obtaining estimates, interval estimates, statistical hypothesis testing and the very concept of a “statistical hypothesis” will be considered.
Mikhail Grishin continues the theme of the previous lecture: talks about parametric tests and summarizes the material studied.
The concept of “non-parametric statistics” is given, the difference in the choice of parametric and non-parametric tests (arguments for and against) is given, the story of non-parametric estimates (bootstrap and non-parametric density estimates) is presented.
In addition to the multiple testing of hypotheses, you will find in the lecture the method of principal components, ANOVA (analysis of variance) and, in part, linear regression.
The topic of linear regression, linear algebra, robust regression will continue, the autoregression model will be considered.
Sergey Rybalkin provides the most basic principles of the Java language: what this language is for, what are its advantages, how does the language work with what you write on it, basic syntax constructs, comparison with C ++, classes, interfaces, inheritance and much more.
The second lecture on the basics of Java. The hierarchy of exceptions, the collection framework, work with collections, generics, generalization of the knowledge gained and the way for further research.
Actual lectures and workshops on programming for mobile and web development are laid out on the Tekhnostrim channel. If you are interested, study at the university, want to get and apply knowledge in the field of development, pay attention to our educational projects: Technopark at MSTU. Bauman, Technosphere at MSU. Lomonosov, Technotrak at MIPT, Technoat at MEPI or come to our online courses .
Lecture 1. Introduction to Python
From the first lecture you will learn what data analysis is, what tools are used for data analysis, and how Python works.
Lecture 2. Advanced Python
A more detailed study of the syntax of the Python language and its use.
')
Lecture 3. Python libraries for data analysis. Numpy, PyTable, Pandas
The lecture covers both standard Python libraries and the libraries most commonly used for data analysis. There is a story about the properties, descriptors, common tasks needed to process data in Python. Evgeny Zavyalov deals with the topic of working with web, mail and websites.
Lecture 4. Visualization, analysis dataset. EDA
We will talk about the main approaches to data visualization and explanatory data analysis. Examples of the application of previously acquired knowledge using parsing open dataset will be considered. Work will continue with the Numpy, Pandas libraries. In the same lecture, the acquaintance with the R language, as a possible alternative to a bunch of Python and libraries, will begin.
Lecture 5. R and libraries
Talk about the advantages of the R language (about disadvantages, see the fourth lecture), which came out of the academic environment, but came close to the possibilities with Python (and inspired the latter for some borrowing). In the West, the free language R is a de facto standard that is not yet so widely known in Russia.
Lecture 6. Introduction to Statistics
Recall the main theorems, the probability, the laws of the distribution of random variables, the estimation problem. Not only let's touch on fundamental knowledge, but also look at the ways of their practical application.
Lecture 7. Introduction to statistical estimation
At the second lesson on statistics, methods for obtaining estimates, interval estimates, statistical hypothesis testing and the very concept of a “statistical hypothesis” will be considered.
Lecture 8. Parametric statistical tests
Mikhail Grishin continues the theme of the previous lecture: talks about parametric tests and summarizes the material studied.
Lecture 9. Non-parametric tests
The concept of “non-parametric statistics” is given, the difference in the choice of parametric and non-parametric tests (arguments for and against) is given, the story of non-parametric estimates (bootstrap and non-parametric density estimates) is presented.
Lecture 10. Multiple hypothesis testing
In addition to the multiple testing of hypotheses, you will find in the lecture the method of principal components, ANOVA (analysis of variance) and, in part, linear regression.
Lecture 11. Time series analysis
The topic of linear regression, linear algebra, robust regression will continue, the autoregression model will be considered.
Lecture 12. Java: the basics of the language. Part 1
Sergey Rybalkin provides the most basic principles of the Java language: what this language is for, what are its advantages, how does the language work with what you write on it, basic syntax constructs, comparison with C ++, classes, interfaces, inheritance and much more.
Lecture 13. Java: the basics of the language. Part 2
The second lecture on the basics of Java. The hierarchy of exceptions, the collection framework, work with collections, generics, generalization of the knowledge gained and the way for further research.
Actual lectures and workshops on programming for mobile and web development are laid out on the Tekhnostrim channel. If you are interested, study at the university, want to get and apply knowledge in the field of development, pay attention to our educational projects: Technopark at MSTU. Bauman, Technosphere at MSU. Lomonosov, Technotrak at MIPT, Technoat at MEPI or come to our online courses .
Source: https://habr.com/ru/post/311694/
All Articles