Types of Data Sources Used in Data Mining (Essay Sample)

Fname Lname
Professor XYZ
Subject ABC
21 January, 2015
Data Sources
One of the most perplexing tasks in data mining involves selecting the most appropriate data sources that answer comparative questions in the course of data mining process. Arguably, lots of data are available that can answer such questions. The implication is that the experts involved in data mining must carefully select the data sources to ensure that such data can wholesomely and efficiently respond to the study questions.
The choice for data sources revolves around two main classifications: primary and secondary. Primary data sources comprise of those sources of data predominantly meant for research. On the other hand, secondary data sources consist of those data sources for which the data had applications in other study areas but may also have a significant contribution towards answering the current study's questions. All in all, regardless of the data sources used, data mining experts must carefully evaluate all data sources depending on how present and credible they are. Conventionally, data sources must address the following key areas for data mining experts to consider them reliable:
* Main purpose of the study
* Experts who collect the data
* Nature of data collected
* Period of data collection
* Methodology used to obtain the data
* Consistency of the data collected with other sources
Primary Data
Primary data comprises of the data collected by experts directly from participants with the main aim of providing credible answers to the study questions. Main forms of primary data sources include; interviews, surveys, and questionnaires. While primary data specifically addresses the study questions, they have the shortcoming of being expensive and time-consuming. The data collection experts must ensure that the process entails an adequate number of observations, availability of key variables, proper control, and an appropriate follow-up duration. The research question is the sole determinant of the type of data required. As such, the data mining experts must match the data to the questions and make decisions about whether to use primary or secondary data sources.
Secondary Data
Secondary data comprises of data initially used for previous studies but can still play a part in answering the study question. The most common forms of secondary data sources include; electronic records, paper-based records, administrative, and regulatory data. One important factor to consider while using data sources is that such sources are not equal as they may contain some aspects of bias. As such, it is recommendable for data mining experts to remain cautious and evaluate all secondary sources of information accordingly. They must critically evaluate evidence in support of conclusions rather than merely accepting the face values in print.
Presentation
Data sources usually have an impact on the presentation proposal. For instance, missing data and changes of data over time may neg