Data Mining for Internet Search Engine - Essay Example

Data Mining Introduction Data mining is notably the simplest and most powerful tool entailing new technology that presents the potential for companies and individuals to focus on the most significant information in the data they have about their customers and potential customer behaviors. Data mining has the capacity to discover information within the data that no other tool, including queries and reports can identify. The course of this dissertation is to reflect the importance of data mining through giving its use in knowledge mining, and its benefits to organisations. Additionally, it reflects a case application of data mining in internet search engine as well as, the ethical context entailed in the use of data mining tool. Thus, in this essay, I argue the case for the importance of data mining. What is data mining? The data mining is a process of knowledge discovery, entailing computer assisted activities that help in digging and analyzing vast compositions of data set and extracting meaningful information from that data (Han & Kamber, p 23, 2012). Benefits of data mining to organisations Among the significant benefits of data mining, is the realisation that through its procedures in sifting through the warehoused information, the tool has the capacity to facilitate the analysts in recognizing significant factor, relations, patters, anomalies and trends that would otherwise remain unnoticed in usual circumstances. For the business operating organisations, the tool has a lot to offer, particularly in the understanding of the business entity and discovering the relations and characteristics of the business entities to facilitate in better decision-making basis (Han & Kamber, p 87, 2012). Moreover, the tool to the organisations also facilitates the spotting of trends particularly on sales and development that factor the creation of smarter and accurate marketing campaigns reaching the customers loyalty prediction. In facilitating the business data, to specific applications of data mining also include, market segmentation in which the tool studies the existent characteristics of the consumers that bus same products from the company (Han & Kamber, p 156, 2006). Secondly, it facilitates customer churn, which entails predicting the customer most likely to leave and join the competitor in the business. Moreover, the use of data mining helps in the procedures of fraud detection, identifying the transactions likely to turn fraudulent hence, protects the organisations accordingly. It also gives the organisations prediction tool for interactive marketing and trend analysis, all facilitating organisation growth, and mastery of trends within its industry. Further, data mining is a significant tool that can notably help in generating new opportunities for the organisations (Linoff & Berry, p 44, 2011). Another concept of significance is the automated discovery of patterns previously unknown. Such discovery actions can spell the distinction line between the business decline or success, particularly when it entails discovering some fraudulent activity before it happens. Further, it also significantly is a time saving technology, splitting the time spend on traditional means to conduct such activities of data analysis (Olson & Delen, p 56, 2008). Thus, the automated technology brings the significant benefit that it is time saving, hence, the course for engaging the technology accordingly. Steps of data mining process in knowledge discovery The traditional procedures entailed in data handing to establish knowledge are notably tiring, lengthy and costly in some aspects. However, the advent of the data mining technology significantly lessened the approach to managing the processes of analysis to synthesise new findings of knowledge from existing data constituents (Linoff & Berry, p 102, 2011). The massive context of data and data sources corresponding to web access, system files and all other physical content data, can prove tiresome and bulky to synthesise new knowledge discovery, even in a learning concept. The steps entailed include the following in the listed order. The first step is to establish the source of information. In this initial engagement, the course is to establish the source that one is analysing and studying to retrieve data. In such, tools and techniques such as clustering are evidently most applicable as they aid in determining areas where need to concentrate the efforts of information synthesis, hence, discovering the new knowledge accordingly. The second step entails picking data points, in which depending on the complexity of data, the extraction of the knowledge can prove straightforward or complex. However, techniques such as Bayesian approaches for building of corpus data to work with probabilities can prove useful (Olson & Delen, p 82, 2008). Clustering also puts the information in groups that align to given criteria of selection, which aids in the faster description of the data for knowledge identification (Olson & Delen, p 86, 2008). This step is essentially notable as it marks the establishment of the essential information groups that constitute the classification of the knowledge in discovery. Further, in projecting the course for the information discovery, the third and fourth steps entail extracting and identifying values, a learning technique that is rather complex (Suh, p 141, 2012). This learning step is extensively reliant on current and past data in the establishment of valid experiences and ultimately lead to the compilation of new information (Suh, p 142, 2012). The step entails identifying valid values and information, as well as, how to spot and eliminate data that is not useful to the context of new knowledge discovery. In interpreting the knowledge, validation and qualification of the information is essential to prove the authenticity of new knowledge gathered. The final step entails interpreting and reporting the results. In this conceptualisation, the step helps in resolving the knowledge discovered into quantifiable values (Pujari, p 64, 2002). For instance, it includes putting the information in values easy to understand such as numerical counts, or grouping depending on specific elements. Aspects such as ranking are applicable in this context. Finally, combine the knowledge discovered in the final compilation or results. Thus, data mining presents a stepwise approach to data for analysis accordingly to derive new knowledge accordingly. Data mining for Internet Search Engine Web data is notably challenging in mining as the nature is highly dynamic, changing from time to time. Such operations entailing the search engine rely extensively on the concept of data mining to extract meaningful information depending on the search command given. When an individual is searching information on the internet, the search engine uses data mining techniques accordingly, including techniques such as classification for the category of search in place. The classifications tool gives the results that fall under the search prompt (Wang, p 42, 2003). In illustration, the search engine will provide a list of results with the keyword from the search, using the classification rules or decision tree algorithms on a given set of data. Further, the search engines also employ the association rule, in which it gives the search results related to the given prompt of search and that occur frequently together (Wang, p 44, 2003). For example, when searching the internet for an item say Television, the results occur with a home theatre since the user may have interest in combining the two, a technique that keeps the search engine ahead of the user. Further, anomaly detection is evident, particularly when the search engine does not identify the command given and results expected, since the object are not conforming to the overall data behavior. For instance, the search engine, when user might search for ‘heart attack’, the anomaly detection may give ‘attack n china’. This is irrelevant to the search keyword and is lies outside the context expected for display. Thus, through these concepts, the internet search engine encompasses the techniques of data mining accordingly to facilitate faster and accurate result of information. Data mining ethics The potential threat in data mining activity leads to the question who should mine data and why (Rawassizadeh, p 1059, 2012). The argument against data mining is that it invades the privacy of the people, as such; no person or unit has the right to investigate data about others (Rawassizadeh, p 1060, 2012). It is only with a better understanding of the use of the data mined that we can entirely discuss the ethical implications of data mining. Organisations such as Electronic Privacy Information Center have the concern regarding the ethical aspects entailed in unauthorised access to data about another entity (Cory, p 23, 2015). For instance, in business, industrial espionage is evident crime advancing accordingly from implications of the use of data mining. It is ethical for companies to share data, but not ethical for a company to explore another company’s data without authorisation (Richards & King, p 398, 2014). Similarly, the government is another entity caught in this ethics battle, as it pursues the noble course to manage terrorism (Cory, p 23, 2015). The government is using data mining to exploit the data of what people engage privately, a factor that is invading the privacy of the individuals. Therefore, the entire central issue in data mining in summation entails the debate on privacy. Conclusion The importance of data mining as a tool is evident from the vast application of the tool in different continuums. From the course of this essay, the very definition of data mining gives the concept entailed in applying the tool. Additionally, the benefits of using data mining in organisations show the vast application of the tool in the field. Further, the application of data mining in knowledge synthesis, as well as in search engines are illustrative of the importance of data mining as a tool. The ethical context reflects the intensity entailed in using data application tool, a factor reflecting its importance in business field as well as the in government responsibilities including security among other sectors. Thus, the importance of data mining is not doubtable in reflection of its vast application. References Cory Robinson, S 2015, The Good, the Bad, and the Ugly: Applying Rawlsian Ethics in Data Mining Marketing, Journal Of Mass Media Ethics, 30, 1, pp. 19-30, Business Source Complete, EBSCOhost, viewed 15 April 2015. Han, J., & Kamber, M. (2012). Data mining: concepts and techniques. Haryana, India ; Burlington, MA, Elsevier. Han, J., & Kamber, M. (2006). Data mining concepts and techniques. Amsterdam, Elsevier. http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&db=nlabk&AN=193647. Linoff, G., & Berry, M. J. A. (2011). Data mining techniques: for marketing, sales, and customer relationship management. Indianapolis, IN, Wiley Pub. Olson, D. L., & Delen, D. (2008). Advanced data mining techniques. Berlin, Springer. Rawassizadeh, R 2012, Towards sharing life-log information with society, Behaviour & Information Technology, 31, 11, pp. 1057-1067, Business Source Complete, EBSCOhost, viewed 15 April 2015. Pujari, A. K. (2002). Data mining techniques. Hyderabad, Universities Press. Richards, N, & King, J 2014, BIG DATA ETHICS, Wake Forest Law Review, 49, 2, pp. 393-432, Business Source Complete, EBSCOhost, viewed 15 April 2015. Suh, S. C. (2012). Practical applications of data mining. Sudbury, Mass, Jones & Bartlett Learning. Wang, J. (2003). Data mining: opportunities and challenges. Hershey, PA, Idea Group Pub. Read More