Data mining is processing data in order to identify patterns and trends that can be used to make decisions or judgments. The process of data analysis, discovery, and mining-model building is often iterative. Different sources and tools provide flexibility in techniques and a variety of results. Although terminology is inconsistent throughout the industry, this article considers eight different mining techniques as examples.
 
Association is a simple correlation between two or more items. Classification describes a type of data source using multiple attributes. Clustering uses one or more attributes for identifying correlating results. Prediction, used in combination with the other data mining techniques, involves analyzing trends, classification, pattern matching, and relation to determine the likelihood of an event. Sequential patterns, normally analyzed over a long term, are used to help identify trends, or regular occurrences of similar events. Decision trees use questions and answers along with other techniques to help classify or identify data. Combinations of techniques is more common than the application of a single technique. Long term (memory) processing is keeping a record of the information and learning from it.
 
Preparing a suitable data model and structure allows for efficient data mining. SQL is commonly used to communicate with a database. According to ANSI (American National Standards Institute), it is the standard programming language for relational database management systems. Concepts used to design in SQL translate to any data query. Examples include structuring and formatting the information so that it can identify sequences and relationships, as well as using source data to build new inquiry points. Finally, reduction can be used to summarize or quantify the information and then output that information in a standardized structure based upon the totals, sums, statistics or other analysis selected for output.
 
REFERENCE

​Brown, M. (2012). Data Mining Techniques. IBM. Retrieved from https://www.ibm.com/developerworks/library/ba-data-mining-techniques

INSIGHT
 
Applying the concepts in this article starts by changing the words “business” and “customer” to “education” and “student,” and then considering the ideas in a general fashion. Begin by realizing that even though big data refers to extremely large groups of people, these mining concepts can be applied when analyzing any data. Then consider the Common Core curriculum where standards do not predicate consistent content types or delivery of information. Because students have different learning styles, these variations can have significant impacts on an individual’s ability to learn the topic. These inconsistencies may be even more apparent when considering emotional intelligence in relation to the material. Gaining insight into these dynamics through systematic assessment of students, teaching styles, and curriculum choices; and mining the data, could reveal trends that can be followed or adaptations that could be implemented which may help increase assimilation of material.

​Ethics are principles of right and wrong above and beyond legal parameters. These are especially important because technology today allows for easy access to information, as well as readily available tools that enable misleading manipulation of data. This publication is segmented into canons, each providing a real-life example along with recommended behavior presented in a comprehensible format that encourages supplemental discussion and promotes use as a template for the adoption of new policies; it does not undermine the need for formal and systematic training in this area. The nine canons are divided into three sections: Integrity, Quality and Security.
 
A person with integrity demonstrates honesty and professionalism. They will not cheat, steal or lie, and they will comply with the standards of behavior that apply to their role. Among these are an ability to perceive data as providing only partial information about the whole person. Especially important is current knowledge of “responsibilities regarding the protection of student data under the Family Education Rights and Privacy Act (FERPA), the Individuals with Disabilities Education Act (IDEA), and the Health Insurance Portability and Accountability Act (HIPAA)” (Forum Guide to Data Ethics, 2010, p. 13). Following unbiased standardized data reporting protocols and governing operations minimizes the possibility that others may misinterpret information. Finally, be willing to hold others accountable for ethical use of data, understanding that they should be doing the same with you.
 
Quality data has practical utility, accuracy, validity, and reliability. It is also timely and cost-effective. Take care to include relevant data, definitions, and documentation that promotes understanding and analysis because value judgements depend greatly on information that provides context.
 
Good security is valuable as it preserves resources, protects private information, and promotes functional operation of the academic system. Establishing policies and enforcing procedures to protect sensitive data helps guarantee privacy and confidentiality.

REFERENCE

National Forum on Education Statistics. (2010). Forum guide to data ethics (N FES 2010–801). U.S. Department of Education. Washington, DC: National Center for Education Statistics. 

INSIGHT

Because non-cognitive data describes personal traits of an individual, it should be treated as private information and treated with the utmost integrity and security for the protection of both the individual and the quality of the data.