Quiz Problems

Misclassification would happen

What would happen when you use very small C (C~0) for Support Vector Machine? Misclassification would happen Data will be correctly classified Cannot comment on classification accuracy None of the above

True

While simple linear regression has more than one dependent variables, it has one independent variable.

True

Standard residual can be used to determine outliers.

True

Artificial neural networks are sensitive to input; hence, data needs to be scaled before it can be used for training and validation

Training data includes both inputs and desired outputs

Back propagation is a learning technique that adjusts weights in the neural network by propagating weight changes. Forward from input to output Backward from output to input Forward from input to hidden nodes Training data includes both inputs and desired outputs

True

Categorical data bears no quantitative relationships among categories.

False

Descriptive analytics uses data to understand past and present and predicts the risk.

dendrogram

A _____ or tree graph is a graphical device for displaying clustering results. Vertical lines represent clusters that are joined together. The position of the line on the scale indicates the distances at which clusters were joined. scattergram dendrogram scree plot icicle diagram

.067

A distributor has two branches in TX. First branch has 70% of the customers, and the second one has the remaining 30%. The customer satisfaction at the first branch is 88% and at the second branch is 98%. The distributor has a centralized call center serving both branches. If an unsatisfied customer makes a call to the call center. What is the probability that it is a customer of the second branch? Please enter your answer as a DECIMAL number in the form of x.yzd (3 decimal point accuracy).

True

A fifth-order polynomial regression model is a better fit to a data than third-order model.

True

An artificial neural network with three hidden layers and one output layer is called 5 layers network.

True

ANOVA uses F-test to determine if independent variables are significant for the regression.

training data

Data used to build a classification model is called ______. validation data training data test data production data

False

If a person says an event is likely to happen, that means the event has less than 0.5 probability

4/6

If you roll a six-sided die numbered from 1 to 6 once and note the number, what is the probability that you will get either an even number or 5? 1/6 2/6 3/6 4/6

True

Logistic regression can be used to classify binary outcomes.

False

Multiple R value is between [0, 1].

True

Predictive analytics analyzes the past data and extrapolates to future.

True

R-Squared value can be less than zero in some situations.

Prescriptive

Solving an equation to determine the total demand given in the form of Demand=40000-100*Price-20*Price2 is an example of ______________ analytics. Descriptive Diagnostics Predictive Prescriptive

Categorical

Supplier ID on a data set is an example of __________ data type. Interval Categorical Ratio Ordinal

agglomerative clustering

This clustering algorithm initially assumes that each data instance represents a single cluster. conceptual clustering agglomerative clustering K-Means clustering expectation maximization

Cluster Analysis

Which method of analysis does not classify variables as dependent or independent? Multivariate regression analysis Logistic regression Cluster Analysis Support vector machines

Identify best alternatives

Which of the following is NOT associated with predictive analytics? Analyze past performance Extrapolate to future Identify best alternatives Predict future

Invented by Hebb

Which of the following is NOT true about the perceptron? Invented by Hebb Is a simplified model of the biological neuron Can be used to make multi-layer neural networks Weights can be trained by adjusting them by an amount proportional to the difference between the desired output and the actual output

Have a natural zero point

Which of the following is a characteristic of ratio data type? Course evaluation results at the end of a semester is an example of it Have a natural zero point Ratios are not meaningful Temperature readings are an example of ratio data type

Uses kernel functions to do a higher dimensionality mapping

Which of the following is correct about the Support Vector Machines (SVM)? Inputs are mapped to lower dimensional space where data becomes likely to be linearly separable Is a method useful for clustering Is very useful since overfitting is never a problem Uses kernel functions to do a higher dimensionality mapping

Impossible, Unlikely, Equal Chances, Likely, Certain

Which of the following is correct order of statements for an event with zero to one probability? Certain, Likely, Equal Chances, Unlikely, Impossible Impossible, Unlikely, Certain, Likely, Equal Chances, Certain, Likely, Unlikely, Impossible, Equal Chances, Impossible, Unlikely, Equal Chances, Likely, Certain

Is an alternative to machine learning

Which of the following is true about the supervised learning? Is an alternative to machine learning Training data includes both inputs and desired outputs To achieve generalization, the actual outputs of the system being trained should be as close as possible to the target outputs (training data outputs) Clustering is an example of supervised learning

As the training data size increases, the model accuracy on the training data increases for the same parameter set

Which of the followings about the decision trees is NOT always correct? As the maximum number of leaf nodes increases, the model accuracy on the training data increases As the training data size increases, the model accuracy on the training data increases for the same parameter set No need to scale input attributes It is possible to get 100% classification accuracy on the training data by choosing a large value leaf nodes.

True

While R-Squared value always increases as the number of independent variables increases, adjusted R-Squared may increase or decrease when a new independent variable included in the model.

ratio, interval

While _______ data has true zero point, ________ data does not have a true zero point. ordinal, interval ratio, interval integer, ratio categorical, ratio