Study Guide Midterm ExamDATA $3300$ The following is a list of topics that will be covered on the midterm exam. Please see the Canvas page Module $7$: Midterm Exam for details about the exam and study materials. Data Analytics & Data Science What data science is and how its used Common sources of data:a$.$ Social networksb. Traditional business systemsc. Internet of Things Different types of data analytics and their application$($s$)$:a$.$ Diagnosticb. Descriptivec. Predictived. Prescriptive The sequence of steps in the CRISP$-$DM process $($and the importance of each$)$:a$.$ Business Understandingb. Data Understandingc. Data Preparationd. Modelinge. Evaluationf. Deployment Data Quality & Preparation What ETL is and why it is important in data analytics Five data quality characteristics:a$.$ Accuracyb. Uniquenessc. Completenessd. Consistencye. Time$-$Appropriateness Common forms of dirty data $($and the threats they pose to data analysis$)$:a$.$ Errors $($typos$,$ misspellings$)$b$.$ Inconsistent Datac. Absence of Datad. Contradicting Datae. Reused Primary Keys Common steps in data cleansing $($and how long data cleansing takes as part of the overall data mining process$)$:a$.$ Parsingb. Correctingc. Standardizingd. Matchinge. Consolidating Data Understanding Familiarize yourself with the differences between qualitative and quantitative variable types:a$.$ Qualitative:i$.$ Nominalii. Ordinalb. Qualitativei. Ratioii. Interval Understand how each of the metrics below relate to data exploration $($data distribution, central tendency, and data dispersion$)$a$.$ Understand each of the following descriptive statistics and their purpose:i$.$ Meanii. Medianiii. Modeiv. Variancev. Standard deviationvi. Interquartile rangevii. Outliersb. Understand and identify:i$.$ Skewnessii. Kurtosis Basic principles of visualization $($and how to interpret visualizations$),$ including most common chart types and their purpose $($and how to create visualizations that clearly communicate the data, not just reinforce prior beliefs$)$:a$.$ Histogramsb. Line chartsc. Box plotsd. Pie Chartse. Stacked column chart Modeling Foundations What data mining is$,$ its appropriate applications, and common data mining tasks What is meant by the terms:a$.$ Data instance$/$Record$/$Case$/$Observationb$.$ Attributes$/$Variablesi$.$ Target attribute$/$Dependent variable The difference between supervised and unsupervised data mining The difference between classification and regression types of supervised data mininga. Classification $$ When the DV is categoricalb. Regression $$ When the DV is numerical Association Rules Analysis What association analysis is$,$ the type of data it requires, and the types of business questions it can answer What an association rule looks like:a$.$ Itemsets and their role in association rulesb. Antecedents and consequents Understand how to calculate and interpret:a$.$ Supportb. Confidencec. Lift Know the tradeoffs between adjusting the minimum support and confidence thresholds and the resulting association rules generated Clustering Analysis What clustering analysis is$,$ the type of data it requires, and the types of business questions it can answer What k$-$means clustering is:a$.$ Understand the steps involvedi. Why we sometimes normalize data in cluster analysis$1.$ z$-$score normalizationii. Impact of potentially over$-$weighting variables which are measuring the same thing in different waysb. Know what k stands for and how it is determinedi. Post$-$hoc evaluationii. Elbow ruleiii. Tractabilityc. Understand the basics of how the algorithm worksi. How are clusters determined?ii$.$ How is a centroid value determined?iii. What is the relationship between a cluster and a centroid?$1.$ How do you interpret results with centroid values? Ways to calculate similarity$/$dissimilarity between cases and how to interpret the distancea. Euclidian Distance $($know and interpret formula$)$b$.$ Intra$-$class similarityc. Inter$-$class similarity Understand how to interpret a cluster analysis through a centroid table and plot. Statistical Correlation What correlation analysis is$,$ the type of data it requires, and the types of business questions it can answer Know how to identify and interpret:a$.$ Correlation coefficientb. Correlation analysis resultsc. Convergent validity $($and when to use it$)$d$.$ Coefficient of determination $($know how to calculate it$)$ What scatter plots look like for strong versus no relationship between two variables Assumptions and limitations of correlation analysis:a$.$ Homoscedasticityb. Normal distribution of datac. Impact of outliers Good Luck!