TO: Prof. Jason Ellis
FROM: Tasnuba Anika
DATE: 10/21/20
SUBJECT: Expanded Definition of Data Mining
Introduction
Table of Contents
The purpose of this document is to discuss about the term “Data Mining” in detail. The term I am defining is how data mining is used in healthcare system. The article talks about the importance usage of data mining in healthcare system which I am going to elaborate and talk more in details. In this document I will first discuss the definitions, context and then the working definitions.
Definitions
Data mining can be interpreted as searching relevant information from large amount of data. In this technique huge sets of data are analyzed, and similar patterns are recognized. The data mining concept became popular in 1990 as software companies started using this methodology to track customer needs. After inspecting the data, they created software that would meet user expectations and market their product.
Computer systems used in data mining can be immensely functional to regulate human constraints for instance inaccuracy caused by tiredness and to give advice in decision making. In the article “Data mining in healthcare: decision making and precision” the author talks about how “Computer systems used in data mining can be very useful to control human limitations such as subjectivity and error due to fatigue and to provide guidance to decision-making processes” (Ionuț ȚĂRANU, 2015, p.1). Getting information utilizing the computers can enhance productivity, saves time, and helps to solve problems efficiently. In this way doctors can easily diagnose patients’ complications and treat them accordingly.
There is a system in data mining called predictive model which is discussed in the article “The Hazards of Data Mining in Healthcare” where author talks about how “data mining techniques used for predictive modelling are based on common hospital practices that may not necessarily follow best practice models, where such a model can recommend inappropriate medication orders that will be used and reinforce poor decision making within the healthcare organization and impact the delivery of patient care” (Mowafa Househ et al., 2017,p. 82). It predicts diseases, update doctors with the new treatments and provides many details regarding healthcare. Moreover, it assists the practitioners to enhance their diagnosis and surgery planning strategy. When there is huge amount of data but the resources are limited it is a matter to concern but only cloud computing have the solution as in the article “A Comprehensive Survey on Cloud Data Mining (CDM) Frameworks and Algorithms” the authors discussed how “Cloud data mining fuses the applicability of classical data mining with the promises of cloud computing” (Hrishav Bakul Barua et al.,2019, p. 1). Which will allow to use huge amount of data efficiently.
Context
In the article “The Hazards of Data Mining in Healthcare” the author talks about “One major challenge that relates to building data mining models is the quality and the relevance of the data used in the healthcare data mining project [4]. Healthcare data is complex and collected from a variety of sources that include structured and unstructured healthcare data and “without quality data there is no useful results”” (Mowafa Househ and Bakheet Aldosari, 2017, p.2). Mining data is not enough it has to be qualityful and relevant to what we need without the quality the data is of no use. “Data sharing over the healthcare organization is another challenge for data mining sector.” (Mowafa Househ and Bakheet Aldosari, 2017, p.2). Privacy concerns include complexity while collecting data from one hospital to another healthcare institution. In past years, many health centers faced security threats and it is creating barriers in data mining. That is why Hospitals are not willing to share their data because they want to keep their information safe. If hospitals shared each other’s data, then data mining outcome would have been similar in all healthcare institutions.
Data mining needs proper technology and logical strategies, and methods for communicating and tracking which can allow computing of outcomes. There are many unorganized raw data are available in the hospitals. Those data are different and voluminous by style. These types of data can cause problems in data mining and eventually generates incorrect results. That is why in the article “Data mining in healthcare: decision making and precision” the author mentioned “The ability to use a data in databases in order to extract useful information for quality health care is a key of success of healthcare institutions” (Ionuț ȚĂRANU, 2015,p.3,).
Working Definition
Overall, we can say that data mining has great significance in the medical field, and it illustrates inclusive operation that requires rigorous comprehension of necessity of healthcare institutions. Data warehouse is a storage where all data are saved and can be retrieved. In this way they can be incorporated to format health center information system. To address the challenges of data mining we must make sure the data mining experiments are done properly. Necessary research should be conducted, and doctors should not only depend on the data mining results to treat patients. They should analyze the outcomes to determine whether the result is correct or not. Otherwise, patients’ health will be in danger.
References
BARUA, H. B., & MONDAL, K. C. (2019). A Comprehensive Survey on Cloud Data Mining (CDM) Frameworks and Algorithms. ACM Computing Surveys, 52(5), 1–62. https://doi-org/10.1145/3349265
Househ, M., & Aldosari, B. (2017). The Hazards of Data Mining in Healthcare. Studies In Health Technology And Informatics, 238, 80–83. Retrieved from http://search.ebscohost.com.citytech.ezproxy.cuny.edu/login.aspx?direct=true&db=mdc&AN=28679892&site=ehost-live&scope=site
ȚĂRANU, I. ionut. co. (2015). Data mining in healthcare: decision making and precision. Database Systems Journal, 6(4), 33–40. Retrieved from http://search.ebscohost.com.citytech.ezproxy.cuny.edu/login.aspx?direct=true&db=aci&AN=115441821&site=ehost-live&scope=site
