Top Online Data Science Courses

In the past few years, courses in Data Science have become, perhaps, the most popular area of ​​online education: dozens of offers can be found at Coursera, edX, and for mastering data analysis at a basic level, even special projects dedicated only to this discipline (for example, DataCamp) have appeared . In this compilation, we have collected the most interesting courses on Data Science on various platforms. For your convenience, we have carefully reviewed the reviews - both on the websites of the educational providers themselves, and on third-party portals, where the advantages and disadvantages of various courses and specializations are assessed. Since the number of courses is huge, we focused on those that offer to teach the listener a whole range of skills - for example, in the case of Coursera, we will not focus on individual courses (even in Data Science, their number approaches one hundred, and the content of many duplicates each other) but about specializations.

Coursera

Recently, a significant part of new courses on Coursera appears in the format of "specializations" - programs of 5-10 subjects, after which you can get a certificate, for which, however, you have to pay. Access to course materials without payment and obtaining a certificate depends on the university organizer (you can often watch lectures for free, but it is impossible to pass estimated tasks).

Statistics with R Specialization

A good course for beginners: if you suddenly decide to do Data Science, but completely forgot the statistics (or never studied it), then specialization can serve as a good introduction to the subject, as well as give a basic idea of ​​R. No prior knowledge is required, but if If you understand the subject, you are unlikely to find something new for yourself.

1. Introduction to Probability and Data
2. Inferential Statistics
3. Linear Regression and Modelings
4. Bayesian Statistics
5. Statistics with R Capstone

Data Science Specialization

Not the most successful of the Coursera specializations: beginners criticize it for not very clear instructions, and those who already have some experience - for the lack of truly new and interesting material.

Courses:

1. The Data Scientist's Toolbox
2. R Programming
3. Getting and Cleaning Data
4. Exploratory Data Analysis
5. Reproducible Research
6. Statistical Inference
7. Regression models
8. Practical machine learning
9. Developing data products
10. Data Science Capstone

Machine learning specialization

This is a mid-level specialization: students are expected to have a basic knowledge of university mathematics, as well as programming experience in Python. The specialization includes only four courses, but each of them will require 5-7 weeks of study.
Courses:

1. Machine Learning Foundations: A Case Study Approach
2. Regression
3. Classification
4. Clustering & Retrieval

Big Data Specialization

A significant part of those who have completed this course do not recommend it to beginners or to those who already have experience in Data Science: the material is not very well presented, and the feedback from the teachers almost does not work. You can view the course materials, but it’s better to refrain from paying for the certificate.
Courses:

1. Introduction to Big Data
2. Big Data Modeling and Management Systems
3. Big Data Integration and Processing
4. Machine Learning with Big Data
5. Graph Analytics for Big Data
6. Big data capstone

Data Mining Specialization

Not all courses of this specialization are equally successful, however, in general, this is a fairly good quality specialization of the average level of complexity. It is assumed that students already know how to program and are familiar with statistics at a basic level.
Courses:

1. Data visualization
2. Text Retrieval and Search Engines
3. Text Mining and Analytics
4. Pattern Discovery in Data Mining
5. Cluster Analysis in Data Mining
6. Data mining capstone

Data Analysis and Interpretation Specialization

Another entry level specialization. You do not need any basic knowledge, so the course will seem too easy for most readers. However, in the quality of the introductory course, this specialization is quite a good choice.

1. Data Management and Visualization
2. Data Analysis Tools
3. Regression Modeling in Practice
4. Machine Learning for Data Analysis
5. Data Analysis and Interpretation Capstone

Udacity

The key idea of ​​the large courses (nanodegree) Udacity is the link with the labor market. Courses are created together with leading representatives of the industry, and if you pay for a special subscription, Udacity even guarantees you a job - or a refund. It should be noted that the nanodegree from Udacity is quite expensive from $ 200 per month.

Machine learning engineer nanodegree

A comprehensive course in machine learning, which will last about 12 months. Training is based on projects where you are performing some real (or close to reality) task. Courses are rather an addition to these projects where you can get the missing knowledge. It is assumed that before the start of the course you already know Python, as well as statistics, linear algebra and mathematical analysis at the level of the university’s first courses.

Data Analyst Nanodegree

The structure of nanodegree data analysis is similar to what we described above - there are also several projects on which you will work for a year (for example, data visualization or work on A / B tests). Despite the fact that students need programming skills and basic knowledge of statistics, the course can be called rather introductory - not so many topics are considered, although students note that the material as a whole is presented quite qualitatively.

edX

Another platform, similar in format to a mixture of Coursera and Udacity - here the courses are created by universities in corporate partnership. Micromasters - analogue of specializations - include 4-5 courses.

Data science

Micromaster Data Analytics from the University of San Diego offers a fairly standard program of four courses:

• Python for Data Science
• Statistics and Probability in Data Science using Python
• Machine Learning for Data Science
• Big Data Analytics Using Spark

One of the significant differences of the program is the opportunity to continue education in a real university. In case of successful completion of the course, you can apply for the Master of Predictive Analytics program at Curtin University: courses passed on edX will be credited as a quarter of the credits required to receive a diploma.