Shulou(Shulou.com)06/03 Report--

This paper mainly describes the four most commonly used data analysis methods in the field of data mining analysis: descriptive analysis, diagnostic analysis, predictive analysis and instruction analysis.

When analysts who are new to the field of data mining analysis are asked what are the most important abilities of data mining analysts, they give a variety of answers.

In fact, what I want to tell them is that the most important ability in the field of data mining analysis is the ability to translate data into meaningful insights that can be clearly understood by non-professionals.

It is necessary to use some tools to help you better understand the importance of data analysis in mining data value. One of these tools is called four-dimensional analysis.

In a nutshell, analysis can be divided into four key methods.

These four methods are described in detail below.

1. Descriptive analysis: what happened?

This is the most common method of analysis. In the business, this approach provides data analysts with important indicators and methods of business measurement.

For example, monthly revenue and loss bills. Data analysts can get a lot of customer data from these bills. Understanding the geographic information of customers is one of the methods of "descriptive analysis". The use of visualization tools can effectively enhance the information provided by descriptive analysis.

two。 Diagnostic analysis: why did it happen?

The next step of descriptive data analysis is diagnostic data analysis. By evaluating descriptive data, diagnostic analysis tools can enable data analysts to analyze the data in depth and get to the core of the data.

The well-designed BI dashboard can integrate the functions of data reading, feature filtering and data drilling according to the time series, in order to better analyze the data.

3. Predictive analysis: what might happen?

Predictive analysis is mainly used for forecasting. The possibility that an event will occur in the future, predicting a quantifiable value, or predicting the time at which something will happen can be done through a predictive model.

Prediction models usually use a variety of variable data to achieve prediction. The diversity of data members is closely related to the predicted results.

In an environment full of uncertainty, forecasting can help make better decisions. Prediction model is also an important method being used in many fields.

4. Imperative analysis: what needs to be done?

The next step in data value and complexity analysis is instruction analysis. The instruction model helps users decide what action should be taken based on the analysis of "what happened", "why it happened" and "what might happen". In general, instruction analysis is not a separate method, but an analysis method that needs to be completed after all the previous methods have been completed.

For example, traffic planning analysis takes into account factors such as the distance of each route, the speed of each route, and current traffic control to help choose the best route home.

Conclusion

Finally, it should be noted that each analysis method is of great help to business analysis, and it is also applied to all aspects of data analysis.

Original link: http://www.kdnuggets.com/2017/07/4-types-data-analytics.html

Reprint, please indicate from: grape city control

About Grape City

Grape City is a global control industry leader and the world's leading provider of enterprise application customization tools, enterprise reports and business intelligence solutions, serving more than 75% of the world's Fortune 500 companies.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.