Data Mining and Data Warehousing - Assignment Example

? Data mining Data Mining Data mining is the latest and the most powerful technology, and that have great potential in helping companies to focus only on the most vital information in the collected data on the behavior, of their potential customers and their current customers (Olson & Delen, 2008). This method gets the data that can’t be obtained through reports and queries with a high level of effectiveness. This can also be referred to as the computer assistance in digging for and analyzing data and finally analyzing the contents meaning. These tools predict and analyze the future business trends which allow businesses to apply knowledge based decisions. Raw data that is stored in the business corporate database increase day by day and as time passes. This is from various numerous credit and cash transactions in the company which is measured in gigabytes or terabytes. This data is usually stored in the centralized database; the raw data does not provide much information. Data warehousing Companies have decided to store their data and invest in a tremendous resource. The information and data on their potential and current customers is stored in this data houses as they are becoming part of the technology. These warehouses are used in consolidation of data which is located in the desperate databases. This data houses store stores usually stores large quantities of data on categories for easier and faster retrieval and interpretation by users. They also enable business managers and executives to store and retrieve large amounts of transactions, and the data required in responding to markets and make more informed business ideas and decisions. Better decision making When the best and available data are collected, data analysis are performed and the most appropriate predictive model is created which results in better understanding on the customers reactions and behaviors towards the marketing programs and reasons for leaving the business. To add on this, various models may results in increased funds success, late payments and reduced bad loans. The good predictive analytics aids businesses in the use of information of previous events to project on new future projects and a good outcome (Olson & Delen, 2008). These are pattern based predictions which are based, on the interrelations between elements of data that cannot be seen on a spreadsheet analysis which leads to a good decision and accurate information. Data mining is a powerful tool which makes it good for business analytics, and the models utilizing procedures to bring about deserving results in customer service. It is also easy to determine which good have been sold and the resulting reactions from customers with increased abundance of data and information, and the growing interrelationship in departmental functions. The processing of customers response can also be time consuming and demanding, labor intensive and expensive in terms of the company staff and this makes its predictive analytic activity to enhance the discovery of products sold to customers. Web mining This technique involves data mining processes such clustering, prediction and the modeling of the differences that analyzes the results of intermediate action, in addition to this, apart from data mining, web mining is a dependent of a real time system that invokes targeted offers on behalf of a process which can be up selling and customer retention and requirement analysis from the customers. This also supports individual marketing of customers based on horizontally collected data in numerous data sources as various transactions occurs. In web mining, real time data processes are identified across all transactions with customers and hence an instance feedback is obtained and hence is the best tool to prevent anomalies and fraud. Clustering This is the method of which data items are grouped in data mining according to their logical relationships on consumer preferences this data is mined to prove market segments or consumer affinities (Han & Kamber, 2006). The most important data mining mode is customer clustering which is, used to make customer relationships management (CRM) and in marketing of goods and services to consumers. This method uses the customer-purchase transaction data in tracking of the buying behaviors and in creating strategic business initiatives. When clustering is properly done, a company does lose customers and business market. Prediction algorithms Prediction Algorithms focuses on the future business values which are based on the business present and current records. The common tools used in this prediction include regression, support vector machines, neural networks and the discriminant analysis. Techniques such as neural networks, genetic algorithms, rough set theory and fuzzy logic are used in failure detection and control prediction in a, business task (Linoff & Berry, 2011). The advantage of these algorithms is the ability to forecast on the probability in a specific data situation, i.e. if the data is normal, this means that the predicted probability is equal to 1 and on otherwise the probability is equal to 0 and the data is referred to as non-conventional. Oh to the higher extent, these algorithms are reliable in giving correct probability results. Privacy concerns The collection of data from customers such as all the customers data e.g. every credit card transaction, every clinic visit, every employment application made can expose the customers information making him/her show every record had. Much of this information is usually stored on the databases of a data mining system, and hence not a lot of people will be comfortable when all this information is exposed (Olson & Delen, 2008). Customers should be privileged on the rights to give the information they want into business databases, but large competitive companies avoid giving this privileges to customers since they require every aspect of information form the customer. Predictive analysis Predictive analysis as a data mining method in business can be used to give out solutions and advantages to business in real life situations through the following ways. Market segmentation – this is whereby the business identifies the common characteristics of their customers who always buy a particular product from your company. Interactive marketing – this is the prediction of everybody who is accessing the various web marketing sites in a business. Market basket analysis – this is done through the identification and understanding the mostly purchased goods from the company. References Han, J., & Kamber, M. (2006). Data mining concepts and techniques (2nd ed.). Amsterdam: Elsevier ;. Linoff, G., & Berry, M. J. (2011). Data mining techniques for marketing, sales, and customer relationship management (3rd ed.). Indianapolis, Ind.: Wiley Pub.. Olson, D. L., & Delen, D. (2008). Advanced data mining techniques. Berlin: Springer. Read More