Correlation and Causation

Correlation and causality itself does not imply causality, but news that can not resist the use of relevance is more meaningful. Today, we analyze three articles to properly use relevance and causality. Descriptions of test relevance and causal relationships include sleep and weight changes, guidance of audiences and obesity, and childbirth and unhealthyness. Discussions are often reported as causal relationships in the world related to news coverage. This is not the case, as further tests are required to show causality based on the unique relevance of the definition.

Relevance and Causal Relationship: In data science projects, causal relationship related errors may occur. As a result of the analysis, it turned out that books available at home were directly related to high test marks, and as a result of analyzing the devil's economics related to the mistake of cause, books are sent to all students in the state It was done. Further analysis shows that a small number of students have better academic results, even if they have never read a book. This helps make corrections on previous assumptions, including the fact that parents usually buy books in an exciting learning environment. If the two projects appear to be related to one another, it does not mean that one project leads to another project.

Some people try to avoid relevance and causality problems by asking about the importance of causality. If two factors are related, is not it enough? Former editor of The best selling book, The Long Tail (2006) and the "Connected" magazine Chris Anderson clearly think so. Anderson believes that in the 2008 article "End of theory: data flow has become outdated due to scientific methods becoming obsolete", causality can be avoided in the era of big data . Other tools that may be brought in. From the theory of human behavior, from linguistics to sociology. Forget taxonomy, ontology, psychology. Why do people know why they do what they do? The important thing is that they do this. We can track and measure it with unprecedented fidelity. There is enough data, the number speaks for itself.