Databases And Data Mining

Nowadays, the World Wide Web is the widely used database type. Here information is stored in tables and can be easily stored and queried. The main purpose of a database is to store and retrieve information in a way that is accurate and effective. Also, it is designed to store, retrieval, modification, and deletion of data in conjunction with various data-processing methods. Recent databases are managed using a database management system.

The purpose of a database management system is to provide a better system to manage the different databases it contains which includes performance, security, availability, and many more (Kotu & Bala, 2015).

Types of Database Management Systems
 Hierarchical databases.
 Network databases.
 Relational databases.
 Object-oriented databases.
 Graph databases.
 ER model databases.
 Document databases.
 NoSQL databases.

Few DBMS examples are MySQL, PostgreSQL, Microsoft Access, SQL Server, FileMaker, Oracle, RDBMS, dBASE, Clipper, and FoxPro. Even there are so many database management systems available, it is essential here to be a way for them to communicate with each other.

Databases are used everywhere in this modern world; Banks, retail, websites, and warehouses, etc. Banks use databases to keep track of their customer accounts, balances, and deposits. Retail stores can use databases to store pricing details, customer information, sales information and quantity on hand, etc. Websites use databases to store content, customer login information, and preferences and may also store saved user input data. Warehouses use databases to manage inventory levels and storage location also. Databases are used anywhere where data needs to be stored and easily retrieved. The filing tasks have been replaced by databases. Simply databases act as our memory power. It stores all the information and can be retrieved at any time (Viloria et al., 2019).

Recently obtaining data from large databases has been grabbing attention of many researchers while as the primary area of getting revenue by a majority of the industries, organizations etc. On the other hand in this digital era there has been observed an escalation in the ability to generate as well as gather data from different sources. Some sources of data gathering and collection are the information retrieved from bar codes of different products, computerization of transactions etc. These sources have been providing us with a large amount of data. Consequently, there have been formed many databases that are employed for different applications such as business management, scientific and engineering data management etc (Maimon & Last, 2013) .

The number of such databases grows continuously due to the availability of robust and powerful database systems (Maimon & Last, 2013). The growth in amount of data and number of databases has eventually raised a requirement for new techniques and tools that can enable smart and automatic retrieval of data and transform it into useful information. Thus, data mining has become a centre of attraction and gained significant importance (Chu, 2013) .

Data mining can be explained as a technology that can be employed for exploring hidden information or data by examining or evaluating large volumes of data that are stored in databases (Tan, 2018). This data is retrieved by employing certain data mining techniques such as machine learning Margaret Tan, (2018), artificial intelligence and statistical approaches (ZenTut, 2019) . Since the data mining process is used by several industries such as chemical, aerospace, marketing etc thus it should be ensured that the process is reliable and it can be easily used by people having little or no knowledge at all about the data mining.

Need for Data Mining
As stated in the aforementioned discussion, data mining can be described as a process of obtaining large volume of data so as to know about the insights and visions of the data. Due to the escalation in demand of data by data industries there has been consequently observed an increase in the demand for data analysts and data scientists. It is by employing these techniques that large volumes of data retrieved from different databases can be transformed into meaningful information that can be used by the business organizations in order to make precise and better decisions for their business. Thus using data mining the organizations can successfully predict the behavior and insights of their customers which leads to great success and data-driven business.

Data mining or knowledge discovery has also been described as a process of examining the data from different perspectives and then compiling it to get some useful information. The technique enables users to explore the data from different dimensions, classify it and then summarize the relationships. Thus, data mining can be explained as the process that finds correlations or pattern among bunch of fields in high relational databases.

Data Mining Techniques
In order to enable a better data mining system, there have been proposed and implemented several approaches. These techniques emphasize that data mining has been constantly evolving (Cooley, Mobasher, & Srivastava, 1997). These innovative techniques improve the performance of data mining concepts such that the companies employing these techniques get a more comprehensive insight in their own data and even predict some better future trends. Some of the key types of data mining techniques are listed below:
Classification: This type of technique helps to classify or categorize data into different classes. It enables classifying a data mining system as per the kind of database on which data mining in performed.
Clustering: The clustering techniques helps to know about the data exhibit similarity in trends etc or that are similar to each other.
Regression: This kind of analysis is helpful in knowing and examining the relationship between variables.
Association Rules: Such techniques aim at finding association between different items in a database. It also helps to find a hidden pattern in the data set.
Outer detection: This type of technique aims at finding that items in a data set which do not exhibit any similarity. It extracts such items and is thus sometimes also known as outlier mining.
Sequential Patterns: Such techniques aim at identifying similar patterns or trends among the data in the dataset.

Challenges in data mining techniques
In order the data mining to be effective, it is necessary to enquire the challenges that are encountered. Some of the key challenges are:
• The approach proposed should be able to handle or manage different types of data
• The data mining algorithms applied should be scalable and efficient.
• The data mining approach should be able to provide useful and certain data mining results.
• The approach enabled should be able to extract information from different sources of data.
• There should also be constraint about privacy and security of data.