Pros And Cons Of Data Mining

Data mining can be viewed as a result of the normal development of information technology Since 1960, database and information technology has been growing methodically from primitive file processing systems to complicated and prevailing database systems [11] [13].

Figure 1.1: History of data base system and data mining

Data mining drives its name for searching a important information from a large database to utilize this information in better way. It is, though, a misnomer, as mining for gold in rocks is usually called “gold mining” and not “rock mining”, therefore by analogy, data mining be supposed to called “knowledge mining” instead. But, data mining become the conventional customary term and very quickly a tendency that even overshadowed

In predictive modeling task, there are two types, Classification that is used for discrete target variable and Regression which is used for continues target variable. The goal of both tasks is to build a model that produce minimum error between predict and real value. An example of classification is the classification of an email as legitimate or spam, or in an online book store predict if a web-user will do a purchase or not. On the other hand, forecasting on the future of a price is a Regression task, because the price is a continues-value attribute. Association Rule Learning (Dependency modeling) is a method that describes associated features in data, searching for relationships between variables. As an example, Web pages that are accessed together can be identified by association analysis. Anomaly Detection (Outlier/change/deviation detection), this class identifies anomalies or outlier data records which cause errors, or might be of interest and requires further investigation. Another class is Clustering, which is the task to discover groups and structures in the data which in some aspect is “similar” or “dissimilar”, without using known structures in the data And the last class, Summarization, attempts to provide a more compact representation of the data set, including visualization and report

Biologists when they a long time ago created a taxonomy (hierarchical classification) made a form of clustering according to genus, family, species and so on But also recently they have applied clustering to analyze the myriad amount of genetic information, such as a group of genes that has similar functions.

• Information Retrieval. The World Wide Web consists of billions of web pages that are accessed with the help search engine queries. Clustering can be used to create small clusters of search results.

• Psychology and Medicine. Clustering techniques are used to analyze frequent conditions of an illness and identifying different subcategories. For example, clustering is used to identify different types of depression, and cluster analysis is used to detect patterns in the distribution/spread of a disease.

• Business. In this field there exists a large amount of information on current and potential customers. Clustering helps to group customer activities, as previously mentioned in detail.

Fig 1.2 shows one and two