The present era has been popularly called the era of knowledge revolution. This knowledge revolution is driven by what we call the process of knowledge discovery. When we look at the contours of knowledge discovery, we observe that this phenomenon is commanded by an information superhighway. The information superhighway transforms data and sources it from various techniques of data mining. Data mining has numerous applications in the present times.

In this article, we explore the terminology and techniques of data mining. 


In simple terms, data mining refers to the process of extracting information out of large data sets. The data that is available to us is usually in unstructured format. If this data is properly mined, it can generate effective insights and analytics. This means that data mining is a technique which operates upon data sets to fetch useful information and discover immersive knowledge. This can later be used for data forecasting mechanisms.


In this section, we discuss the five most important techniques that are extensively used in data mining. 

  • The first and foremost technique of data mining involves recognition of various types of patterns within the available data sets. The technique of pattern recognition allows us to classify data sets into different types. Once we are able to detect different patterns within data sets, we can give it a graphical interpretation. This can be used to track various changes in data sets as new data types are added to the existing repository.
  • The second technique of data mining is related to the classification of data into various types on the basis of different attributes. The technique of classification can be performed on the basis of a single parameter or a set of parameters which are deemed together as a feature set. For instance, we may classify the population of an urban area on the basis of the levels of income. The population that is taken into account is the data set and the perimeter selected is the level of income. Consequently, the divisions that are obtained include upper class, middle class and lower class on the basis of thresholds of income already set for the purpose.
  • The third technique of data mining is related to the association of data with various elements and parameters. For instance, let us assume that a number of customers buy a painting after buying a wall clock. While the primary data set is the number of purchases of wall clocks, the secondary data set is the number of purchases of a wall painting. We may mine the transaction history of different customers and find that the number of consumers who buy a wall clock has a strong association with the number of consumers who buy a painting. 

This correlation can help us in forming association rules with various products and fine tune our recommendation system. This can in turn increase revenue and growth of a company in the long run. Hence, association rules can prove to be a game changer when it comes to data mining in the ecommerce sector. 

  • The fourth technique that is followed in data mining is called the process of clustering. As the name signifies, clustering is a process related to unsupervised machine learning and establishes relations between various products or data sets. After establishing relationships between various data sets, this process is used to segregate such data sets on the basis of parameters that are found to be common. This means that no specific parameters are considered at the beginning of clustering. As the process goes on, clusters are formed on the basis of feature and attribute detection.
  • The fifth important technique related to data mining is called predictive methodology. This methodology is related to forecasting of future data sets by extrapolating the present ones in real time. Needless to mention, this process involves the mining of historical data sets to sketch a forecast of upcoming events based on those datasets. This technique finds application in numerous financial domains like the stock market analysis. This is also a useful technique for determining the boom and bust cycles of a business based on data sets about slowdown and recession. It also allows us to identify the prospective areas of investment based on previous sectoral performance. 

Concluding remarks 

Data mining presents new and prospective opportunities to analyze and assess new areas and sectors that are dependent upon data either directly or indirectly. 


Please enter your comment!
Please enter your name here