Benefits of Data Mining to the Businesses When Employing - Assignment Example

? Data Mining Data Mining Determine the benefits of data mining to the businesses when employing: When predictive analytics are employed in any business, they help such an entity come up with mathematical models that will assist the business to better understand the variables behind the driving success. Predictive analytics depend on formulas which make comparisons between the failures and successes of the business in the past thus putting the business in a position to predict its outcomes in future. This data mining process adds value to a business in that it enables it to understand both the motivations and behaviors of the customers making it more effective in its marketing. The more the business understands why some of its customers are loyal and how it can continue to retain and attract customers from different segments, the more the business will be able to come up with compelling offers and messages that are relevant. Predictive analytics results in an analytical framework which helps in the prediction of product preferences and customer buying habits required in the discovery of meaningful relationships and patterns in the customer data so as to accomplish better market targets and drive customer loyalty and value (Turkey, 1997). Association discovery can be employed in a business to determine the affinity patterns of its products. This information is important in optimizing the manner in which the business orders are picked and accumulated from their centers of distribution. The same information obtained through association discovery can be based on to generate dynamic rules which would permit business orders to be picked or accumulated taking into consideration the chances of another order that is identical occurring in a span of few days. The outcome of such an optimization is that decisions take less time thus substantially saving on the business costs (Agresti, 2002). Web mining allows the business to sift through information regarding the market situation so as to identify where the market value is. Such discoveries will help the business come up with new opportunities as the business with such information will have the ability to implement parallel processing systems and high performances as it is in a position to analyze large data in a very short time. The business can also make use of data mining to try out different business models to best understand the market situation and adapt to it. Such information also helps the business in making better predictions (Witten and Eibe, 2011). The clustering of information using data mining related to customers have served as important networks to business in the process of handling numerical market data. The market is normally the main target for any business when it comes to information clustering. This data mining technique helps the business segment potential customers on the basis of given sets of attributes. The technique also helps businesses to adapt the best operation methods through discovering market facts that are hidden giving the business an added advantage when it comes to its competitive position. The business will also be in a position to understand better what their customers really need and want. 2. Assess the reliability of the data mining algorithms. Decide if they can be trusted and predict the errors they are likely to produce Data mining algorithms is one of the main methods used in the exploration of linkages and relationships among sets of data. Such algorithms have been relied on by many businesses as they are believed to have the ability of reducing computational efforts and are in a position to deal with structural system’s failure mode. Despite data mining algorithms being viewed as being highly accurate and efficient in their operation, the algorithm has been associated with a number of identification, exploration and exploitation errors. Reliability is seen in the way any given data mining model operates on different sets of data. A model that is reliable is one which comes up with the same predictions type or generates the same types of patterns without considering the supplied test data (Witten and Eibe, 2011). 3. Analyze privacy concerns raised by the collection of personal data for mining purposes. As data mining has continued to evolve, its impact on personal privacy has become more complex and has also raised controversies. Some of the privacy concerns include; 1. Profile creation: Data mining technologies normally contain both non identification and identification information about individuals. Such information includes; individual names, there security number, driver’s license number, gender, physical attributes, education background and age. The valid concern is that such information readily available in data mining techniques allow for the aggregation in that individual profiles can be easily created using such information and the profile taken to disparate systems. This is a valid concern which has to be addressed. To deal with such incidents, several computerized database networkings have come up that allow for “de-identify” of the profiled individuals. This is achieved by having the database contain information that can only be aggregated by preserving many of the valuable information about individuals (Turkey, 1997). 2. Identification of criminals and Terrorists: there are also concerns on privacy issues in relation to the use of data mining in identifying people who have been involved in criminal or terrorist activities. Data mining has been used in predicting which person is most likely to be involved in such crimes but concerns arise when it comes to those using such information and technology. The people have to be careful while putting in place all the requirements that give details on when these kinds of actions may be staged against a suspect following the data mining activities. In some occasion, the process fails to materialize thus innocent victims are held responsible for crimes they didn’t commit. Such concerns are only valid when it comes to the mined information which is afterwards realized not to be useful in the investigation. To deal with such situations, measures such as the Total Information Awareness program which integrates any information about an individual in such a manner that is better classifies, detects and identify potential criminals more effectively (Agresti, 2002). 3. The other concern is the government access to individual information. The government has been accused for collecting personal information and going ahead to engage in mining the same information. The concern is valid only if the government takes such a move for the wrong reasons. The government may reach out for information on an individual using data mining techniques with a good motive, for example for job assignment. This concern can be dealt with using PPDM which is used a privacy-preserving determinant. The program puts together sets of data of multiple individuals without revealing each individual data. 4. Provide at least three (3) examples where businesses have used predictive analysis to gain a competitive advantage and evaluate the effectiveness of each business’s strategy. a. The analytical customer relationship management (CRM) applies predictive analysis in its data on customers while pursuing the objectives of the CRM. The objective of the CRM is to achieve a view that is holistic on the customers regardless of where the information on the customer is in the business. The CRM makes use of the predictive analysis for its sales, marketing campaigns and rendering services to customers. This tool is effective as in enables the CRM to remain focus on its efforts effectively with regards to its customer base breadth (Agresti, 2002). b. Most businesses are faced with various types of fraud and this has remained a major problem to business for many years. Some of the fraud include, identity theft, inaccurate credit applications, insurance claims that are false and transactions that are fraudulent. A predictive analytic model can be used to help check on such kinds of fraud and reduce the exposure of a business to fraud. Such a model will work effectively in detecting fraud in financial statement through allowing the auditors to estimate the relative risks of the business and increase its audit procedures substantively. c. Predictive analysis can also be effectively used in cross-selling in corporate organizations. Such organizations gather and keep abundant data such as sale transactions and customer records. Exploiting relationships that are hidden in such data can result in a competitive advantage to the business organization. In order for such corporation to offer multiple products, customer behavior analysis can result in efficient products cross selling. This will in turn transform to higher profitability for every customer and also strengthen the relationship between the customers and the organization (Witten and Eibe, 2011). References Agresti, A. (2002). Categorical Data Analysis. Hoboken: John Wiley and Sons Turkey, J. (1997). Exploratory Data Analysis. New York: Addison-Wesley Witten, I. & Eibe, F. (2011). Data Mining: Practical Machine Learning Tools and Techniques. New Jersey. Morgan Kaufmann Read More