Literature Review

In this report the topic that is selected for the business intelligence is supervised data mining. Supervised is a learning mechanism of the data mining task. Several examples of training are included in the training data. It is called supervised data mining because for each of the input that is provided as an object a desired output is generated from that. The main report will be based on the business intelligence. Business intelligence mainly refers to the applications, technologies, collected practices, analysis, integration and presentation of business information. The main objective of business intelligence is to provide a better support to the decision making system of the business by applying new and innovative technologies. The decision support system of the business intelligence is data driven. In this report importance of the business intelligence will be given for further discussion. In the report further will be elaborated on the topic that is supervised data mining. The working procedure of supervised data mining will be described in the report. The business intelligence will be also defined in the report. The report will show that how the supervised data mining will be able to use in the business intelligence (Witten et al., 2016). It has been seen in the several surveys that the organization those are using the solutions of the business intelligence get more success and they can handle the projects of data mining more effectively and efficiently.

This literature review substantially discussed about the business intelligence and supervised data mining technology.The business intelligence is very useful for the business organizations to make decisions. In the following part of the literature review the author described the business intelligence and how the business intelligence works. The review also stated that the business intelligence provides the predictive views of the current business. In the recent time the data mining in the business is used widely. The amount of the data in the field of the business is getting large day by day and to handle those data and then used the data for further use in the business field. The large amount of data is handled by the business intelligence in order to provide new strategies for the business and provide new business opportunities (Provost and Fawcett, 2013).Structured and unstructured data can be handled easily by the business intelligence to provide business solutions. As the data mining is helping the business intelligence a lot and for that the predictive data mining procedure is appropriate. The business intelligence has a fixed value to achieve as the goal needed to be cleared form the business perspective. In this paper the theoretical approach will be taken to discuss the survey on the topic supervised data mining. The paper will discuss about the data mining more specifically the directed form of data mining that is the supervised data mining.

Supervised Data Mining

It is the most appropriate way of data mining when the user or the organization has a particular target value which will be able to predict about the data. There may be several outcome of the target or the target can be of numeric values that is continuous. Supervised data mining can only be used when the user have the ideal subset of the data point (Cortez and Embrechts, 2013). Then the data are used to build the model like the typical data point when it gets any of the data points of the different target values.

Classification of the supervised data mining is started as the mentioned method above. The classification can be of many types when applied to the business. Same as the classification there is another method that is called regression that target the value as per the numerical value rather than the category of the data. Regression model is used by the data mining when any numerical data needed to be taken care of by the. Target value must be assessed by the organization in order to get the desired outcome. The supervised data mining also known as the predictive data mining as it has the capability of predict the user data that consists the numeric data.

Supervised learning in the data mining is the most directed way of doing the task. The working procedure in supervised data mining is very simple. In the supervised data mining a input is given and with the input a desired output. Both of the input and output are part of the training data in the supervised data mining. During the process of learning in the supervised data mining correct targets are known for some and they are given as the input to the model. In the supervised data mining to work accurately it is necessary to construct a proper test set, validation and a construction of proper training. Data mining basically searches the pattern in the data. It is basically used when the data amount is large and need some data to be extract from that large amount of data. Data mining has the fix goal which is to extract the particular data whenever required (Tan, Steinbach and Kumar, 2013). Data mining is works as the knowledge discovery in the database. The supervised data mining is from of machine learning which enables the machine to learn several thing by the provided data set. An algorithm of supervised data mining first analyses the data of the training and then produces a function that is inferred, that can be later used for the new mapping of the data. There are several steps that needed to be followed while a problem is solved using the supervised learning of data mining. In the first step the training examples needed to be determined. It is required to be done before doing any further steps, where the business organization or the user will be able to identify that what kind of data set is used in the training. Then in the second step, training set it collected from the data. In this step the input object and the output object is collected. In the third step, the input feature is converted on to vector feature, that are consists of many of the features which are object descriptive (Torgo, 2016). It has to be make sure that there will be less number of features. In the fourth step, the structure of the learned function is determined for using the supervised data mining algorithm. Decision trees are often used in this step. In the fifth step all the design is completed and then the learning algorithm is run on the training set that previously collected. There are several parameter are controlled by the algorithms. In the sixth and the final step, accuracy of the functions that are learned is evaluated. After these all step the measured set and the training set gets separated.

Working Procedure

Data mining is use to generate the Business Intelligence. It is most important to use the data that is collected rather that only to collect the data. 4To use this information the appropriate thing that need to have is business intelligence. The business intelligence has the ability to convert a data on to information and a normal information to a knowledge. This is the best possible way where the decision making in the business is optimized. Business intelligence can be called as the set of different methodologies, technologies, and application to collect, transform and the refine the data from the transactional system and information that are unstructured (Lison, 2015). Different type of data mining tools are used in the business intelligence to make the decisions more accurately and efficiently. The supervised data mining helps the business to provide better customer management relationship, by mining the data of the customer habits and the patter of the customer data. Then the strategy of the business intelligence used in the knowledge to increase the benefit of the particular business.

• Classification Analysis: Major purpose of classification analysis regarding data mining is that the information regarding data that are present resulting to the fact that the data and metadata are both termed in the processing of data management. Data recording is also one of the main prospect that will ensure the fact that the data are segmented in clusters. Certain clustering of outlook email deals with the fact that the data will be done with utmost efficiency.
• Advantage: In case of implementing classification analysis, the major advantage is that the data that are collected are divided in departments in order to store the data in proper compartments which helps the researchers to collect information in a very efficient way.
• Association Rule Learning

In case of Associate rule learning the major aspect that deals with the processing of gaining information regarding the dependency of the data on other data that are present in the global data base. In case of data mining for ding the source of the data and their dependency, leads to the processing of the expression of the data in a much lenient way. This acts as the catalogue designing of program building (Baker and Inventado, 2014).

Advantages: This method helps the researchers to understand the dependency of the data that are required for the formation of the business management. In case the data manage the major advantage it provide is that it helps in understanding the source of the data that are required for the completion of the data management.

• Anomaly: Anomaly of the data management will ensure the fact that the data that are not processed can be done with the help of this methodology. In this case the data that are processed can be done with the help of the anomaly process. This also helps in detecting the odd one out of the entre data set. This is the reason that the data set that is being distinguished from gets the data management done in a very efficient way.

Advantage: the major factor of this process is that the data management that is done with the help of this method ensures the fact that the data management that is done helps in processing of the fraud cases. This is the major reason that the data management which might get affected negatively ad the security for the data base might be at stake. This is the major reason that the implementation of Anomaly in the data mining system has been increasing in such a fast rate.

Clustering Analysis: Cluster analysis deals with the facts that the data base of the system will get profited as this terminology ensures the fact that the data that are having a similar framework are attached with each other leading to the fact that categorization can be done in this process with utmost efficiency. Data that have equivalent data structure tends to stay together and the data that stay together are supposed to be stay together and form a cluster. This is the major reason that the data management gets easier in the course of the business processing (Santos et al., 2013).

Data Mining in BI

Advantage: The major advantage of the cluster methodology is that the data management can be diversified in clusters and the data that are similar in nature gets compartmented together. This is the major reason that the usage of this methodology has been occurring. This leads to the fact that the data management can be done with utmost proficiency.

Regression analysis: Regression analysis is one of the major fact that leads to the understanding of the connectivity of the data that are present in data structure. In case the interrelation between the data are understood, the data management gets easier. This methodology helps in understanding the relation that is present in a proper directional way. Regression analysis helps in understanding the fact that the relation between the business projections is highly dependent on the dependency matrix of the products that are interrelated with each other (Patil and Sherekar, 2013).

Advantage: major advantage of this methodology is that the data management can be done with utmost efficiency. This leads to the fact that the data managing that is done with the help of the Regression analysis is very precise in nature and the data analysis will include the fact that the source of data can be identified with this methodology.

Conclusion

 This report concludes data mining for the business intelligence. The report discussed the supervised data mining method.  This is called as supervised data mining as there is an expected output for every input provided. The report discusses the business intelligence briefly. The main aim is to support the business by applying innovations. The report has discussed the way data is driven for business intelligence. Th working procedure of supervisor is being discussed in the above report. The business intelligence and the need is also described in the above part of the report. Thus, we can conclude that, data mining is required in the business intelligence. Many organization are their which became successful after implementing business intelligence in their organization. So this will increase the efficiency of the workers and company will earn huge profits.

References:

Baker, R.S. and Inventado, P.S., 2014. Educational data mining and learning analytics. In Learning analytics (pp. 61-75). Springer, New York, NY.

Chitra, K. and Subashini, B., 2013. Data mining techniques and its applications in banking sector. International Journal of Emerging Technology and Advanced Engineering, 3(8), pp.219-226.

Cortez, P. and Embrechts, M.J., 2013. Using sensitivity analysis and visualization techniques to open black box data mining models. Information Sciences, 225, pp.1-17.

Lison, P., 2015. “An introduction to machine learning.

Patil, T.R. and Sherekar, S.S., 2013. Performance analysis of Naive Bayes and J48 classification algorithm for data classification. International journal of computer science and applications, 6(2), pp.256-261.

Provost, F. and Fawcett, T., 2013. Data Science for Business: What you need to know about data mining and data-analytic thinking. ” O’Reilly Media, Inc.”.

Santos, I., Brezo, F., Ugarte-Pedrero, X. and Bringas, P.G., 2013. Opcode sequences as representation of executables for data-mining-based unknown malware detection. Information Sciences, 231, pp.64-82.

Tan, P.N., Steinbach, M. and Kumar, V., 2013. Data mining cluster analysis: basic concepts and algorithms. Introduction to data mining.

Torgo, L., 2016. Data mining with R: learning with case studies. Chapman and Hall/CRC.

Witten, I.H., Frank, E., Hall, M.A. and Pal, C.J., 2016. Data Mining: Practical machine learning tools and techniques. Morgan Kaufmann.