What is Data Mining and how it brings benefits to the Business - Term Paper Example

What is Data Mining and how it brings benefits to the Business? Report [Type the 10 Table of Contents What is Data Mining? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 2-4 Data Mining Outlook in the Past ~~~~~~~~~~~~~~~~~~~~~~~~~ 4 Data Mining Tools ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 5-6 Data Mining Application ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 6-7 Examples and Benefits of Data Mining ~~~~~~~~~~~~~~~~~~~~~~~ 7 Integrative Structure and Outcome Model of Relationship Benefits ~~~~~ 7 Mining Consumer data in to Intelligence ~~~~~~~~~~~~~~~~~~~~~~~~~~8 Bibliography ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 9 What is Data Mining? Data mining is a knowledge discovery process that is also known as Knowledge Discovery in Databases (KDD). The primary function of data mining or KKD is to analyze and search large number of data patterns in a database. Likewise, it utilizes computerized data analysis techniques to expose relationships of data items that were previously hidden or undetected. However, the data that is analyzed via different techniques is fetched from data warehouses, where many databases are interconnected with each other. Major techniques that are involved in the process of data miming are regression, classification and clustering (Data Mining. 2007).Data mining is incorporated for gaining in depth patterns for market intelligence from data warehouses containing massive amount of data. However, the issue that arises is not the quantity of data, as we already have massive amount of data to work with, it is the methodology that is required to learn data. Likewise, data provides all the attributes but how to utilize them for gaining benefit is another question. This is the area that is addressed by data mining, as it is used for extracting valuable information from large amount of data saved on periodic basis. Likewise, information that can be extracted may contain relationships and different patterns. For instance, a retail store may indicate that some products are more in demand in one channel of distribution, there may be two different products that are sold at the same time in a specific geographic location, some specific products are more in demand in some geographic locations and similarly, some products are more demanding in certain events may be associated with religious events. If we take an example of Wal-Mart, the store has found that if there is a probability of a hurricane, the demand of beet increases in that specific geographical area, therefore, stores have to stock more beers that usual in this sort of situation (Keating 2008). Employee associated with utilizing patterns of customer behavior from data mining, i.e. a financial analyst would seek facets of the store or organization that may become bankrupt, similarly, human resource managers would seek information of a successful potential employee, employees working in a credit card department would like to get information associated with credit card debts payments from potential customers and also to analyze the legitimate credit card transactions against the falsified ones, marketing department executives would like to extract information associated with product purchases (Keating 2008). For instance, any online store would like to know what kind of brands are more popular for shoes; Adidas, Nike or Reebok. If all the mentioned information is available to any organization’s different departments, than efficient decisions will be made accordingly. Credit card department may focus and target potential customers to prioritize for minimizing fraudulent and fake transactions. Similarly, human resource department can short list based on background checks and hires the most appropriate candidate. Likewise, the marketing department will focus on demands of customer for different products. Accordingly, after making all the departments capable of taking efficient and informed decisions based on insights from data miming, businesses may reconstruct their objectives, strategies, goals and offerings on different products. However, data mining techniques are not limited to the mentioned benefits that it injects in to business processes and decisions, law enforcement agencies also use data mining for detecting and analyzing probability of crime scenarios and environment of these crimes. Moreover, stock exchange can utilize data mining techniques for detecting fake activities, so do pharmaceutical organizations by mining data for predicting effectiveness in compounds and to reveal innovative chemical objects that may directly contribute for a specific disease. Similarly, data miming is utilized by the airline industry to forecast delayed flights, so do weather experts for predicting weather by analyzing patterns. They can forecast the weather by informing the time and day for weather types such as sunny, warm, cloudy, bright skies, snow, cold, heavy or natural disaster. Furthermore, trust organization and nonprofit organizations also use data mining techniques for accessing the prospect of donations from donors for any sort of charity work or cause. Data Mining Outlook in the Past As mentioned earlier, data mining specialist abstracts value data from the data warehouse, however, the strategy for pulling off valuable information is to identify relationships and patterns that are located within the stored data archives. Certainly, in the past, humans have always searched for patterns of activity i.e. astronomers looking at starts, weather forecasts. Likewise, in this current information age, humans are still eager to identify and understand patterns of activity. For instance, predictions for election results, new launched products, popular products, temperature variances due to global warming. We can roughly say that over the past 30 years, there is an evolution in a slow pace from data processing to data mining. In the late 1960’s data was organized and collected by organizations by database management methods that facilitated data organization along with queries. Moreover, Online Transaction Processing technique was one of the routine methods for retrieving and storing data at a rapid pace. It was possible because OLTP which is “a class of program that facilitates and manages transaction-oriented applications, typically for data entry and retrieval transaction processing” (Online Transaction Processing (OLTP). 2007) and “also refers to computer processing in which the computer responds immediately to users’ requests” (Online Transaction Processing (OLTP). 2007) was considered to be one of the advancements in computing. Data Mining Tools Tools for data miming can be categorized in four segments (Keating 2008): Prediction Tools Classification Tools Clustering Analysis Tools Association Rules Discovery Tools Prediction Tools are considered as a method of prediction. Likewise, the sole purpose of this tool is to utilize traditional statistical forecasting in order to calculate a value for a variable. The classification tool is the most common tool that is used for data mining. The sole purpose of this tool is to differentiate unlike classes of different objects and activities. If we take an example of a credit card industry, the transactions will be only of two natures i.e. legitimate and fraudulent. Accordingly, by deploying data mining classification tools, organizations can classify whether the transactions are legitimate that may lead to minimize risk and saving money. Another example will be of an advertising company that may want to forecast consumer interest so that the promotion can be tailored accordingly. However, there are many other aspects in this case but data mining provides in-depth knowledge of targeted audience i.e. customers, so that the campaign can be designed effectively. Clustering Analysis tool is used for clustering. Likewise, this tool is considered to be most effective for clustering group of different products that will be consolidated naturally. Likewise, data mining was most beneficial for identifying groups that needs to be consolidated, which are identified by this tool. However, data mining eliminates unnecessary groups that are not considered for any contribution in the decision making process for businesses. Whereas, some groups that are identified, becomes vital, and can be utilized by the organization to generate business. The clustering analysis tools is most widely adopted because it is used for an economic term that is most popular called as ‘Market Segmentation’. This term refers to a division of customers, who are dispersed anywhere, in to segments. Although, there may be a limitation factor consisting location, economic conditions, culture, lifestyle, income of people etc. likewise, after the division of customers in to segments, different strategies are set for each market segment. Association Rules Discovery tool is used for linking characteristics. For example, what kind of household products are used in certain high society areas?, What sort of household items are used area wise, What kind of fast food chains are most popular in certain areas, what kind of books are popular among certain type of people etc. Organizations or marketing departments use these kinds of customer behavior strategies to target customer according to their needs. Data Mining Application Apart from the tools, there are certain applications that are required to effectively deploy data mining techniques and methods. Likewise, the two major applications that are used currently are SAS enterprise Miner and SPSS. Both of these application contains a wide range of tools supporting all data mining techniques i.e. prediction tool, classification tool, clustering analysis tool and association rule discovery tool. These two most widely used data mining software are stand alone and have the capability to import and process data, on wired and wireless networks, in almost any format. As these two tools are stand-alone fully functional data mining software, there are other various data mining tools that do not support statistical support, as compare to these two, but they can be integrated with spreadsheets such as Microsoft Excel. Likewise, these small data mining tools can provide an interface for examining effectiveness and familiarity with data mining. As, Microsoft Excel is a small program, it can support data mining characteristics by integrating with these small tools and can handle it effectively. Moreover, it provides a broader picture and demonstrates the usefulness of data miming for business (Keating 2008). Furthermore, these small tools also illustrate statistics associated with diagnostic that are utilized for analyzing the information retrieved from these tools incorporated with Microsoft Excel. Examples and Benefits of Data Mining A massive amount of research is conducted on data mining in order to derive business benefits by making informed decisions based on extracted data from large amount of data. Few of the many methods, approaches, models, frameworks are illustrated below: Integrative Structure and Outcome Model of Relationship Benefits A model for relationship benefits was developed by (Kong, Zhang 2011). The purpose of this model was to analyze the influence mechanism of profits associated with relationships that were impacted on customer satisfaction and loyalty. The study for this model demonstrated data mining techniques to investigate the composition of relationships created on relational bonds in order to establish integrative structure model. Likewise, the research on this model investigated the impact of relationship benefits associated with satisfaction of the customers and the impact on customer satisfaction and loyalty for constructing an integrative outcome model. Mining Consumer data in to Intelligence Implementation of data mining techniques for a telecommunication company in China embraces translation of customer data in to intelligent and purposeful data. The name of the company is China mobile and is the world’s largest telecom organization. The company used data mining as a strategic tool to maintain its competitive standing by focusing shifts for gaining insight of customer preferences and their habits (Chau 2007). Threse Cory, who is an analyst with ‘Analysys Research’, says that the telecommunication sector should not emphasize on internal data but take a holistic approach i.e. also collect external data for getting insights to the upcoming expectation of market trends. China mobile and its subsidiaries are developing their own data warehouses, as the major benefits businesses can extract from incorporating business intelligence in data mining are: Focus on Customer Centricity, Enhanced competition, liberalization and customer usage analysis (Chau 2007). However, in order to achieve these major benefits major applications that must be used in this process are churn management, revenue assurance, new product launch, measurement of service success rates, measuring partnership success rates, detecting active and inactive consumers of mobile service (Chau 2007). However, the approach should be from reactive to proactive. Likewise, reactive approach includes the usage of applications for data capture that will be analyzed by implementing business intelligence with data mining to derive results and benefits (Chau 2007). Bibliography Book 1. Data Mining. 2007. Network Dictionary, , pp. 134-134. Journal 1. KEATING, B., 2008. Data Mining: what is it and how is it Used? Journal of Business Forecasting, 27(3), pp. 33-35. 2. Online Transaction Processing (OLTP). 2007. Network Dictionary, , pp. 351-351. 3. KONG, Q. and ZHANG, M., 2011. The Integrative Structure and Outcome Model of Relationship Benefits: Using Data Mining. Journal of Software (1796217X), 6(1), pp. 48-55. Article 1. CHAU, F., 2007. Mining customer data into intelligence. (cover story). Telecom Asia, 18(3), pp. 24. Read More