What is Data Mining?
Data mining is a subfield of data science that use sophisticated analytical methods to extract actionable insights from data (Jackson, 2002). The goal of data mining (DM) is to sift through large amounts of data in search of patterns or correlations that will be of value to the data’s intended audience. Data mining is a subprocess of knowledge discovery in databases (KDD), a methodology for collecting, cleaning, and analyzing data in data science. Historically, data mining (DM) has been viewed as a useful Business Intelligence (BI) tool for gaining new insights (Wang & Wang, 2008).
The Data Mining Process
There are four main steps in data mining: collecting data; processing and transforming it; analyzing it to draw conclusions and make predictions; and finally, writing up and presenting the results in reports (Stedman, 2021). In the data gathering phase, relevant data, such as structured, unstructured, and semi-structured data, are gathered from internal and external sources and stored in various places, such as separate source systems, a data warehouse, or a data lake. At this point, the data goes through a series of preparation steps. First, errors and other data quality issues are found and documented; then, the data is profiled and prepared for cleansing. Also, data transformation is used to ensure data sets are comparable. The ETL level addresses data extraction, transformation, and loading (Ong et al., 2011). Data analysis seeks to extract useful information and meaningful trends from large data sets. At these stages, intelligent patterns are used to extract data patterns using classification and clustering methods. Writing and presenting reports is the end-user layer, which encompasses querying, reporting, and virtualization tools to communicate findings to the appropriate audience.
Advantages of Data Mining
In general, the increased ability to uncover hidden patterns, trends, correlations, and anomalies in data sets leads to increased business benefits from data mining. That information can improve business decision-making and strategic planning by combining traditional data analysis and predictive analytics (Stedman, 2021). Thus, using data mining insights, executives and risk managers can better evaluate a company’s financial, legal, and cybersecurity threats and devise mitigation strategies.
Furthermore, using the insights from data mining, businesses can better cater their advertising and marketing efforts to their target demographic (Stedman, 2021). Data mining can also help sales teams increase the percentage of qualified leads that become customers and upsell existing customers on related products and services. Companies can now anticipate customer service difficulties and equip call center representatives with the most current information for use during conversations with clients, thanks to data mining.
For instance, data mining is utilized in the marketing industry to categorize markets and study constantly shifting customer preferences. Analyzing the connections between characteristics such as age, gender, taste, and location make it feasible to predict customer behavior through tailored loyalty marketing campaigns (Iberdrola, 2022). Marketers can also use data mining to predict which users would abandon a service, which keywords will generate the most significant interest, and which addresses should be included on a mailing list for optimal results.
Disadvantages of Data Mining
While data mining is a highly effective marketing strategy, it is not without its limitations. Data mining has many drawbacks, one of which is that it functions best with huge amounts of data (Prasanna, 2022). If, for instance, there are just 100 customers in a shopping database, data mining will not yield valuable results from the information of these 100 customers. Instead, data mining will be more fruitful if the list has 500,000 customers because there will be more data to mine.
Secondly, privacy and data security worries are significant drawbacks of data mining (Prasanna, 2022). Previously, corporations would only exchange customers’ personal information with one another if doing so was a necessary part of offering a service. Many consumers fret that their private information is being sold to unscrupulous parties.
References
Iberdrola. (2022). Discover how data mining will predict our behaviour. https://www.iberdrola.com/innovation/data-mining-definition-examples-and-applications
Jackson, J. (2002). Data Mining; A Conceptual Overview. Communications of the Association for Information Systems, 8, 19. https://doi.org/https://doi.org/10.17705/1CAIS.00819
Ong, I. L., Siew, P. H., & Wong, S. F. (2011). A five-layered business intelligence architecture. Communications of the IBIMA, 2011. https://doi.org/10.5171/2011.695619
Prasanna. (2022). Data Mining Advantages And Disadvantages. AplusTopper. https://www.aplustopper.com/data-mining-advantages-and-disadvantages/
Stedman, C. (2021). Data mining. TechTarget. https://www.techtarget.com/searchbusinessanalytics/definition/data-mining
Wang, H., & Wang, S. (2008). A knowledge management approach to data mining process for business intelligence. Industrial Management & Data Systems, 108(5), 622-634. https://doi.org/https://doi.org/10.1108/02635570810876750
