Question 1 Actionable intelligence is the primary goal of modern-day Business Intelligence (BI) systems vs. historical reporting that characterized Management Information Systems (MIS). True False Question 2 Match the following scenarios with the correct type of analytics: A self-driving car making decisions about what to do based on analyzing data about possible future outcomes and consequences [ Choose ] Predictive analytics Descriptive analytics Prescriptive analytics Google Analytics reporting the number of posts, followers, page views, etc. [ Choose ] Predictive analytics Descriptive analytics analytics Prescriptive A financial institution using a credit score to help determine the likelihood of a consumer paying bills on time [ Choose ] Predictive analytics Descriptive analytics Prescriptive analytics Question 3 Visualization is generally considered a sub-area of prescriptive analytics. True False Question 4 Match the following tasks with their appropriate data preprocessing phase: Integrate/merge data from various sources [ Choose ] Data Consolidation Reduction Data Transformation Data Cleaning Data Deal with missing values and outliers [ Choose ] Data Consolidation Reduction Data Transformation Data Cleaning Data Normalize the data [ Choose ] Data Consolidation Reduction Data Transformation Data Cleaning Data Select an appropriate subset of the data [ Choose ] Data Consolidation Reduction Data Transformation Data Cleaning Data Question 5 A variable with possible values of blue/green/brown/other for eyes is an example of a _______________ variable binomial ordinal binomial nominal multinomial ordinal multinomial nominal Question 6 In your data collection phase, you have identified data sets of interest that refer to your university as VT, VaTech, Virginia Tech, and Virginia Polytechnic Institute and State University. _________________ seeks to recognize these apparent differences and reconcile them as the same information. Data reduction Data integration Data transformation Imputation Question 7 All of the following are true about a box-and-whiskers plot, EXCEPT The line across the box indicates the average value of the data Outliers are more than 1.5 times the upper/lower quartile The box represents the middle 50% of the data The plot can show whether the data is symmetrically distributed Question 8 Which of the following is true about skewness? The income distribution in the US is generally left-skewed. Skewed distributions are typically bi-modal. In a right-skewed distribution, both the mean and the median are to the right of the mode. A negatively-skewed distribution naturally has negative kurtosis. Question 9 All of the following are true about the "curse of dimensionality" EXCEPT Too many variables for the number of observations can result in models having trouble generalizing. Too many variables for the number of observations can result in a sparse data space that is hard for classification models to fit well. Too many observations for the number of variables can result in model overfitting. Too many variables for the number of observations unnecessarily increases storage space and processing time for the modelling algorithm. Question 10 For best results when deploying visual analytics environments, focus only on power users and management to get the best return on your investment. True False Question 11 To speed assimilation of numbers on a dashboard, the numbers need to be placed into context. This could include all of the following EXCEPT Indicating whether the numbers are good or bad Comparing the numbers of interest to other baseline or target numbers Including as much information as possible Using visual objects (e.g., traffic lights) or visual attributes (e.g., color coding) Question 12 Match the following visualization needs to the chart type that would BEST communicate the information: Bar chart [ Choose ] Stock price change over a 5 year period Car accidents by zip code Relationship between height and age in basketball injuries Amount of website traffic by origination site Breakdown of how Americans spend their leisure time Line chart [ Choose ] Stock price change over a 5 year period Car accidents by zip code Relationship between height and age in basketball injuries Amount of website traffic by origination site Breakdown of how Americans spend their leisure time Pie chart [ Choose ] Stock price change over a 5 year period Car accidents by zip code Relationship between height and age in basketball injuries Amount of website traffic by origination site Breakdown of how Americans spend their leisure time Map [ Choose ] Stock price change over a 5 year period Car accidents by zip code Relationship between height and age in basketball injuries Amount of website traffic by origination site Breakdown of how Americans spend their leisure time Scatter plot [ Choose ] Stock price change over a 5 year period Car accidents by zip code Relationship between height and age in basketball injuries Amount of website traffic by origination site Breakdown of how Americans spend their leisure time Question 13 During classification in data mining, a false positive is an occurrence classified as true by the algorithm while being false in reality. True False Question 14 When training a data mining model, the testing dataset is usually larger than the training dataset. True False Question 15 Which broad area of data mining applications uses rules to divide instances into defined groups? Associations Visualization Classification Clustering Question 16 Which common data mining subtask is also referred to as market-basket analysis? Clustering Regression Similarity matching Association rule mining Question 17 Regression is simply the degree of association between two or more numeric variables, while correlation implies that there is a causal relationship between explanatory variables and a response variable. True False Question 18 What does the robustness of a data mining method refer to? Its ability to predict the outcome of a previously unknown data set accurately Its speed of computation and computational costs in using the model Its ability to construct a prediction model efficiently given a large amount of data Its ability to overcome noisy data to make somewhat accurate predictions Question 19 The basic idea behind a decision tree is that it recursively divides a training data set until each division consists entirely or primarily of examples from one class. True False Question 20 In k-fold cross-validation, the complete data set is randomly split into mutually exclusive subsets of approximately equal size and _______ using each individual subset in turn, while using the rest of the data in each iteration as a _______ data set. tested, training trained, test trained, training tested, test Question 21 Match the following definitions to the common measures of central tendency. Numerical average of data values [ Choose ] Mode Mean Median Middle value of an ordered list of numbers [ Choose ] Mode Mean Median Value that appears most often in a set of numbers [ Choose ] Mode Mean Median Question 22 The talent of the data science team is as important as the data they have to work with. True False Question 23 A critical skill in data science is the ability to decompose a data-analytics problem into pieces such that each piece matches a known task. This is important for all the following reasons EXCEPT this avoids wasting time and resources “reinventing the wheel” there are tools available for known tasks people can focus on the more interesting parts of the process that require human creativity and intelligence, such as story-telling there are many, many different types of tasks Question 24 Which of the following statements about supervised and unsupervised learning tasks is true? Supervised tasks can be accomplished using the same techniques as unsupervised tasks. Classification is an unsupervised task. If a specific target can be provided, the problem can be addressed as a supervised task. All of the above Question 25 What determines how the weights are chosen for the objective function in a parametric model? The data itself The business goal The preference/experience of the analyst The modeling tool Question 26 Which of the following is NOT a supervised data mining algorithm? k-Means Clustering Neural Networks k-Nearest Neighbor Support Vector Machines Question 27 A simple split partitions data into two mutually exclusive subsets in order to estimate the accuracy of a classification model. True False Question 28 A key part of the Data Understanding phase of the CRISP-DM model is Estimating the costs and benefits of each data source Understanding the use scenario of the problem to be solved Normalizing or scaling the data Applying data modeling techniques to the data Question 29 Business Understanding Data Understanding Data Preparation Model Building Question 30 The Evaluation phase of the CRISP-DM model includes all of the following EXCEPT Ensuring that the model satisfies the original business goals Deploying the data mining techniques Using both quantitative and qualitative assessments of the data mining results Thinking about the comprehensibility of the model to stakeholders Question 31 A _________ is an undesirable situation where a variable collected in historical data gives information on the target variable. leak missing value link fraud Question 32 Selecting informative attributes can reduce uncertainty help subset data so that the dataset is not so large increase the accuracy of modeling all of the above Question 33 The creation of models from data, in which we generalize from specific cases to general rules, is known as model deduction. True False Question 34 What is the entropy of the dataset in the attached picture with respect to the Yes/No target variable (rounded to two decimals)? Supporting Materials • • 0.65 0.95 0.98 0.29 Question 35 For the dataset shown in the attached picture and the target variable Yes/No, what would the information gain be if the dataset were split on body shape (oval/rectangle)? • • approximately 0.33 approximately 0.65 approximately 0.78 approximately 0.20 Question 36 Entropy... is a measure of information gain. is a measure of uncertainty in a data set. can be either positive or negative. will ideally be 1 at a leaf node in a decision tree. Question 37 Which of the following does NOT describe support vector machines (SVMs)? An SVM can estimate class membership probability. SVMs are based on supervised learning. An SVM chooses the line to minimize the margin between two classes. An SVM can be applied when the data are not linearly separable. Question 38 A fitting graph plots... True positive rate vs. false positive rate True positive rate vs. false negative rate Accuracy vs. size of training set Accuracy vs. model complexity Question 39 You will always need all the attributes in a data set to find the best fitting model. True False Question 40 Information gain... can be used with supervised segmentation for regression problems. requires absolute purity to be of any value. can be applied to any number of child subsets. is independent of the size of the child subsets. Question 41 Data visualization can assist with the data mining process in all the ways below EXCEPT impute missing values add historical context point out missing data communicate relationships in data
Purchase answer to see full attachment