Data mining is a logical combination of data knowledge and statistical analysis developed in business knowledge or a process that uses statistical, mathematical, artificial, artificial and machine-learning techniques to extract and identify useful information for related knowledge from multiple databases big. Data mining includes tasks known as knowledge extraction, data archeology, exploration in data pattern processing and information harvesting
The data mining component of the KDD process is often a recurrent iterative application of certain data mining methodologies. In this discussion using the term pattern and model. The data pattern is defined as the instantiation of the model. For example f (x) = 3x + x is the pattern of model f (x) = ax + x. Data mining performs model matching or determines patterns from or to observed data.In the development of data mining technology, there is a model used to make the process of information feeding of existing data. According to IBM, data mining model can be divided into 2 parts:
Verification model
This model uses the hypothesis of the user, and tests the previous estimates using the available data. The emphasis on this model lies in the user responsible for the preparation of estimates (hypothesis) and problems in the data to negate or confirm the expected results (hypothesis) taken.
For example, in the field of marketing, before a company releases a new product of the market, the company must have information about the customer's propensity to buy the product to be issued. Estimates (hypothesis) can be structured to identify potential customers and characteristics of existing customers. Data about previous customer purchases and data about customer circumstances can be used to compare purchases and customer characteristics to establish and test predicted targets. From the overall operation thereafter can be screened carefully so that the number of estimates (hypothesys) that previously many will be reduced in accordance with the actual situation. The main problem with this model is that no new information can be created, but only proof or weaken the forecast (hypothesys) with existing data. The data present in this model is only used to prove support for a previously taken hypothesis. So this model is entirely dependent on the ability of the user to analyze the problems to be extracted and obtained information.
Discovery model
This model is different from the verification model, wherein this model the system directly finds important information hidden in a large data. The existing data are then sorted-to-find a pattern, existing trends, and common circumstances at that time without user intervention and guidance. These findings state the facts in the data found in the shortest possible time.
For example, suppose a bank wants to find customer groups that can be targeted a product to be output. In the existing data then held the search process without any process of estimation (hypothesis) before. Until finally all the customers
grouped by the same characteristics.
The data mining component of the KDD process is often a recurrent iterative application of certain data mining methodologies. In this discussion using the term pattern and model. The data pattern is defined as the instantiation of the model. For example f (x) = 3x + x is the pattern of model f (x) = ax + x. Data mining performs model matching or determines patterns from or to observed data.In the development of data mining technology, there is a model used to make the process of information feeding of existing data. According to IBM, data mining model can be divided into 2 parts:
Verification model
This model uses the hypothesis of the user, and tests the previous estimates using the available data. The emphasis on this model lies in the user responsible for the preparation of estimates (hypothesis) and problems in the data to negate or confirm the expected results (hypothesis) taken.
For example, in the field of marketing, before a company releases a new product of the market, the company must have information about the customer's propensity to buy the product to be issued. Estimates (hypothesis) can be structured to identify potential customers and characteristics of existing customers. Data about previous customer purchases and data about customer circumstances can be used to compare purchases and customer characteristics to establish and test predicted targets. From the overall operation thereafter can be screened carefully so that the number of estimates (hypothesys) that previously many will be reduced in accordance with the actual situation. The main problem with this model is that no new information can be created, but only proof or weaken the forecast (hypothesys) with existing data. The data present in this model is only used to prove support for a previously taken hypothesis. So this model is entirely dependent on the ability of the user to analyze the problems to be extracted and obtained information.
Discovery model
This model is different from the verification model, wherein this model the system directly finds important information hidden in a large data. The existing data are then sorted-to-find a pattern, existing trends, and common circumstances at that time without user intervention and guidance. These findings state the facts in the data found in the shortest possible time.
For example, suppose a bank wants to find customer groups that can be targeted a product to be output. In the existing data then held the search process without any process of estimation (hypothesis) before. Until finally all the customers
grouped by the same characteristics.
Advertisement
No comments