Fluent Essays

Choose three of these metrics from the attached file. and discuss them succinctly using your textbook as a reference and some other resources. Your references should not be less than four in total. db3_question.pdf Unformatted Attachment Preview Chapter 3 • Nature of Data, Statistical Modeling, and Visualization 123 to be numerically represented (even the nominal variables need to be converted into pseudo binary numeric variables), whereas decision tree algorithms do not require such numerical transformation—they can easily and natively handle a mix of nominal and numeric variables. Analytics projects that overlook data-related tasks (some of the most critical steps) often end up with the wrong answer for the right problem, and these unintentionally created, seemingly good answers could lead to inaccurate and untimely decisions. Following are some of the characteristics (metrics) that define the readiness level of data for an analytics study (Delen, 2015; Kock, McQueen, & Corner, 1997). • Data source reliability. This term refers to the originality and appropriateness of the storage medium where the data are obtained—answering the question of “Do we have the right confidence and belief in this data source?” If at all possible, one should always look for the original source/creator of the data to eliminate/mitigate the possibilities of data misrepresentation and data transformation caused by the mishandling of the data as they moved from the source to destination through one or more steps and stops along the way. Every move of the data creates a chance to unintentionally drop or reformat data items, which limits the integrity and perhaps true accuracy of the data set. • Data content accuracy. This means that data are correct and are a good match for the analytics problem—answering the question of “Do we have the right data for the job?” The data should represent what was intended or defined by the original source of the data. For example, the customer’s contact information recorded within a database should be the same as what the customer said it was. Data accuracy will be covered in more detail in the following subsection. • Data accessibility. This term means that the data are easily and readily obtainable— answering the question of “Can we easily get to the data when we need to?” Access to data can be tricky, especially if they are stored in more than one location and storage medium and need to be merged/transformed while accessing and obtaining them. As the traditional relational database management systems leave their place (or coexist with a new generation of data storage mediums such as data lakes and Hadoop infrastructure), the importance/criticality of data accessibility is also increasing. • Data security and data privacy. Data security means that the data are secured to allow only those people who have the authority and the need to access them and to prevent anyone else from reaching them. Increasing popularity in educational degrees and certificate programs for Information Assurance is evidence of the criticality and the increasing urgency of this data quality metric. Any organization that maintains health records for individual patients must have systems in place that not only safeguard the data from unauthorized access (which is mandated by federal laws such as the Health Insurance Portability and Accountability Act [HIPAA]) but also accurately identify each patient to allow proper and timely access to records by authorized users (Annas, 2003). • Data richness. This means that all required data elements are included in the data set. In essence, richness (or comprehensiveness) means that the available variables portray a rich enough dimensionality of the underlying subject matter for an accurate and worthy analytics study. It also means that the information content is complete (or near complete) to build a predictive and/or prescriptive analytics model. • Data consistency. This means that the data are accurately collected and combined/merged. Consistent data represent the dimensional information (variables of interest) coming from potentially disparate sources but pertaining to the same subject. If the data integration/merging is not done properly, some of the variables of different subjects could appear in the same record—having two different patient 124 Part I • Introduction to Analytics and AI records mixed up; for instance, this could happen while merging the demographic and clinical test result data records. • Data currency/data timeliness. This means that the data should be up-to-date (or as recent/new as they need to be) for a given analytics model. It also means that the data are recorded at or near the time of the event or observation so that the time delay–related misrepresentation (incorrectly remembering and encoding) of the data is prevented. Because accurate analytics relies on accurate and timely data, an essential characteristic of analytics-ready data is the timeliness of the creation and access to data elements. • Data granularity. This requires that the variables and data values be defined at the lowest (or as low as required) level of detail for the intended use of the data. If the data are aggregated, they might not contain the level of detail needed for an analytics algorithm to learn how to discern different records/cases from one another. For example, in a medical setting, numerical values for laboratory results should be recorded to the appropriate decimal place as required for the meaningful interpretation of test results and proper use of those values within an analytics algorithm. Similarly, in the collection of demographic data, data elements should be defined at a granular level to determine the differences in outcomes of care among various subpopulations. One thing to remember is that the data that are aggregated cannot be disaggregated (without access to the original source), but they can easily be aggregated from its granular representation. • Data validity. This is the term used to describe a match/mismatch between the actual and expected data values of a given variable. As part of data definition, the acceptable values or value ranges for each data element must be defined. For example, a valid data definition related to gender would include three values: male, female, and unknown. • Data relevancy. This means that the variables in the data set are all relevant to the study being conducted. Relevancy is not a dichotomous measure (whether a variable is relevant or not); rather, it has a spectrum of relevancy from least relevant to most relevant. Based on the analytics algorithms being used, one can choose to include only the most relevant information (i.e., variables) or, if the algorithm is capable enough to sort them out, can choose to include all the relevant ones regardless of their levels. One thing that analytics studies should avoid is including totally irrelevant data into the model building because this could contaminate the information for the algorithm, resulting in inaccurate and misleading results. The above-listed characteristics are perhaps the most prevailing metrics to keep up with; the true data quality and excellent analytics readiness for a specific application domain would require different levels of emphasis to be placed on these metric dimensions and perhaps add more specific ones to this collection. The following section will delve into the nature of data from a taxonomical perspective to list and define different data types as they relate to different analytics projects. u SECTION 3.2 REVIEW QUESTIONS 1. How do you describe the importance of data in analytics? Can we think of analytics without data? 2. Considering the new and broad definition of business analytics, what are the main inputs and outputs to the analytics continuum? 3. Where do the data for business analytics come from? 4. In your opinion, what are the top three data-related challenges for better analytics? 5. What are the most common metrics that make for analytics-ready data? ... Purchase answer to see full attachment