What is data mining?


A large amount of data is generated every second and it is necessary to have knowledge of the different data mining tools that can be used to handle this huge data and apply interesting data mining algorithms and visualizations in a short time.

Data mining It is the set of methodologies used in the analysis of data from various dimensions and perspectives, finding previously unknown hidden patterns, classifying and grouping the data and summarizing the relationships identified.

For example, data mining can help companies identify their best customers. Organizations can use data mining techniques to analyze the previous purchase of a particular customer and predict what a customer could buy in the future. You can also highlight purchases that are out of the ordinary for a customer and may indicate fraud.

Companies can use to find inefficiencies in manufacturing processes, potential product defects or weaknesses in the supply chain.

History of data mining

One of the first articles to use the phrase "data mining" was published by Michael C. Lovell in 1983. At that time, Lovell and many other economists had a fairly negative view of the practice, believing that statistics could lead to conclusions incorrect when not informed by the knowledge of the subject.

But in the 1990s, the idea of ​​extracting value from data by identifying patterns had become much more popular. Database and data warehouse providers began using the buzzword to market their software. And companies began to become aware of the potential benefits of the practice.

In 1996, a group of companies that included Teradata and NCR led a project to standardize and formalize data mining methodologies. His work resulted in the Industry Standard Process for Data Mining (CRISP-DM). This open standard divides the data mining process into six phases:

  1. Business understanding
  2. Data comprehension
  3. Data preparation
  4. Modeling
  5. Evaluation
  6. Deployment

Companies such as IBM continue to promote the CRISP-DM model to this day, and in 2015, IBM released an updated version that expanded the basic model.

In the early 2000s, web companies began to see the power of data mining, and the practice really took off. While the phrase "data mining" has been eclipsed by other buzzwords such as "data analysis," "big data" and "machine learning," the process remains an integral part of business practices. In fact, it is fair to say that data mining has become a de facto part of the management of a modern business.

Types of data mining

Scientists and data analysts use many different data mining techniques to achieve their goals. Some of the most common include the following:

  • The grouping It involves finding groups with similar characteristics. For example, marketers often use clustering to identify groups and subgroups within their target markets. Clustering is useful when you don't know what similarities may exist within your data.
  • The classification classify the elements (or individuals) into categories based on a previously learned model. Classification often comes after grouping (although you can also train a system to classify data based on the categories defined by the scientist or data analyst). The grouping identifies the potential groups in an existing data set, and the classification places the new data in the appropriate group. Computer vision systems also use classification systems to identify objects in images.
  • The Asociation Identify data that is commonly found near each other. This is the technique that drives most recommendation engines, such as when Amazon suggests that if you bought an item, you might also like another item.
  • Anomaly detection Look for data that does not fit the usual pattern. These techniques are very useful for fraud detection.
  • Regression It is a more advanced statistical tool that is common in predictive analysis. It can help social network and mobile application developers increase participation, and it can also help forecast future sales and minimize risk. Regression and classification can also be used together in a tree model that is useful in many different situations.
  • Text mining analyze how often people use certain words. It can be useful for the analysis of feelings or personality, as well as to analyze publications in social networks for marketing purposes or to detect possible leaks of employee data.
  • The summary puts a group of data in a more compact and easy to understand way. For example, you can use the summary to create graphs or calculate averages from a given data set. This is one of the best known and accessible forms of data mining.

Data mining tools

Organizations have a wide variety of proprietary and open source data mining tools at their disposal. These tools include data warehouses, ELT tools, data cleaning tools, dashboards, analytical tools, text analysis tools, business intelligence tools and others.

Leave a Reply

Your email address will not be published. Required fields are marked *