Data mining is a field of several disciplines, which deals mainly with research that leans towards applied mathematics and computer science. Data mining takes into consideration legislations that are geared towards preventing unwanted exceptions and backdoors. Secondly, data mining opts to consider legislations that will be applicable in the future. Data mining considers three perspectives in defining what type of data it uses. The first perspective is that the data can be stored in several databases in any format. Secondly, analysis of data is through several ways by either comparing the data or by performing complex mathematical operations on the data. Lastly, data mining output can be anything (Kamber, 2006).
Cancelled data access applications
These applications are used to prevent access to data or information by making the data private and only accessible to certain individuals. The data in this context are only visible to specific access levels.
Use of data mining in law enforcement agencies
Data mining is applicable in law enforcement to discover new information. This enables addition of value to crime information, which is a deviation from the traditional way of counting crimes and generating of summary reports. This way of data analysis allows new information with good operational value. Second use of data mining is through anticipating or predicting future events. Another use of data mining in law enforcement is to characterize crime trends and patterns. This helps the law enforcers to develop specific ways of reducing crime rate (McCue, 2007).
Examples of threats to personal privacy through data mining
An example of threat posed by data mining is disclosure of personal information without consent. This enables linking of records in the databases, which eventually allows identity of the specific persons that the data represents. The systems in this context are potentially abusive. Another threat is open provision of information. Government agencies disseminate data publicly to make them more efficient on consumer behavior. This type of data, which is for evaluation of needs, provision of critical analysis, that enable making of independent conclusions is vulnerable to unwanted access. Another potential threat is combination of electronic data files with data mining techniques, which then raise the level of breach in privacy. This has enabled the unlocking of potentiality of data hence creating major dilemma for electronically disseminated data. Lastly, a potential threat is that data mining methods can break disclosure methods that protect individuals (Ian, 2007).
Steps in minimizing these threats
A way to control the threats posed is through offering controlled access and use of data centers where there is controlled access to the data. Another way to minimize these threats is with privacy-preserving data publishing. This process uses techniques such as randomizing of the data, k-anonymity and l-diversity. Second step to minimizing these threats is through changing of results of data mining applications to preserve privacy. An example to this is the use of association rule hiding techniques where some association rules are suppressed in order to preserve privacy of the data. Lastly, another method used is through cryptographic methods for distributed privacy. This will involve computation of a common function that is applicable across the multiple sites (Ian, 2007).
Is the science of data mining a business necessity or a luxury?
Data mining is a business necessity rather than a luxury because it considers discovery of patterns form large groups of data. This is eventually important since in the business world it is used to predict pattern in which the customers purchase what type of goods or services (Kamber, 2006).
How the processes of data mining/web mining/text mining are wrongfully used to discriminate among clients/customers.
In fighting crime and terrorism, analysis of large amounts of data may enable access to information, which contains errors. Due to the information that is incorrect, risk profiles may contain false positive information. These people do not share the group’s characteristics but are part of the group. The other group is the false negatives, which contain people who constitute the risk profile described, and they do not belong to the group. Data mining may lead to stigmatization of particular groups when the information that is contained in risk profiles of company data become public (Zantinge, 1996).
Ways to prevent discrimination among clients
The best method that is used to prevent this issue is through removal of sensitive attributes that are contained in the databases. The attributes can be ranging from gender, ethnic variations, and religion, criminal or medical records of a group or an individual. Another method in preventing this issue of discrimination is through transparency and accountability. This method revolves around the use of data instead of access to the data. Lastly, it can be noted that making the data anonymous and private may prevent in discrimination of the clients (McCue, 2007).
Successful performance of goal-oriented investigations and global-oriented investigations enable establishment and maintaining of good information position. The availability of large quantities of data makes it possible for law enforcements agencies to apply data mining in finding new information. Data mining plays an important role in extraction of useful information from the data through provision of technologies that are necessary for the process. Data mining is proven to be important in fighting crime since it gives a trend that gives an insight of how the crime rate is in a given area hence helping in decision making in allocation of police forces. In another view, data mining can be controversial since it may negatively affect personal and civil rights of an individual, which lies on privacy and discrimination (Zantinge, 1996).
Ian, D. (2007). Knowledge Discovery and Data Mining: Challenges and Realities. Hershey, New York: Prentice Hall. Web.
Kamber, M. (2006). Data Mining: Concepts and Techniques. 2nd ed. Oxford: Morgan Kaufmann. Web.
McCue, C. (2007). Law enforcement data mining and predictive analysis: Techniques and tools. Web.
Zantinge, D. (1996). Data mining. Harlow, England: Addison Wesley Longman. Web.