In R, the assignment operator `<-` works from right to left, so the expression evaluates `mydata$v4 <- NULL` first to remove the column, then `mydata$v2 <- NULL` to remove the second column. This chain effectively deletes both variables from the data frame in a single line. The code does not swap values or replace data, but strictly performs deletion for each target variable sequentially. Understanding this right-to-left evaluation order is crucial for correctly interpreting chained assignments in R.

Question 2 Easy Mcq

✓ Correct ✗ Incorrect

In Google Analytics tool, which of the following analysis should be performed in order to identify the origin of a user's web traff‌ic?

Acquisition analysis

Audience analysis

Behavior analysis

Conversion analysis

Explanation:

Acquisition analysis is the specific module designed to track how users initially find and enter your website by categorizing traffic sources such as organic search, paid ads, direct visits, and social media. This process involves examining referral data and campaign parameters to determine the exact origin of each visitor session. By focusing on these entry points, analysts can understand which marketing channels are most effective at driving new traffic to their digital properties. Consequently, this analysis provides the essential insights needed to identify where web traffic originates without delving into user demographics or post-visit actions.

Question 3 Easy Mcq

✓ Correct ✗ Incorrect

In data mining, which of the following options correctly def‌ines Precision, which is used for assessing the quality of text retrieval?

precision: l[Relevant] n [Retrieved]l / l[Retrieved]l

Precision= l[Retrieved} U [F-score]l + l[F-score}l

Precision= l[Recall] / [F-scorejl x l[RecalI]l

Precision= l[F-score] x [Recalljl - l[F—score)l

Explanation:

Precision measures the accuracy of a retrieval system by calculating the ratio of relevant documents found among all documents returned. It is computed by dividing the count of relevant items retrieved by the total count of items retrieved, ensuring that only useful results are counted. This metric is crucial for text retrieval because it directly reflects the user's satisfaction with the relevance of the search results provided. A high precision score indicates that the system effectively filters out irrelevant information, delivering high-quality content. Therefore, the formula representing relevant retrieved items divided by total retrieved items is the correct definition.

Question 4 Easy Mcq

✓ Correct ✗ Incorrect

In a generalized linear model. which of the following link functions belongs. by default, to Poisson family?

(link = "identity")

(link = "inverse")

(link = "logit")

(link = "log")

Explanation:

In a Poisson generalized linear model, the response variable represents counts that must be non-negative. The log link function is the canonical choice because it maps the linear predictor to the positive real line, ensuring predicted counts remain valid. This transformation allows the model to handle the multiplicative nature of count data while maintaining the required constraints for the distribution. Consequently, when specifying a Poisson family without explicitly defining a link, the software defaults to using the logarithm to connect the mean to the linear predictor.

Question 5 Easy Mcq

✓ Correct ✗ Incorrect

The ______ of a worksheet defines its appearance.

Form

Format

View

Record

Explanation:

The format of a worksheet encompasses all visual styling elements such as fonts, colors, borders, and cell alignment that determine how data is displayed to the user. By adjusting these specific formatting settings, you directly control the aesthetic presentation and readability of the entire spreadsheet. This concept is fundamental because it allows users to customize the look of their data without altering the underlying numerical values or logical formulas. Consequently, the format is the precise term that defines the visual appearance of a worksheet in standard computing terminology.

Question 6 Easy Mcq

✓ Correct ✗ Incorrect

In data mining, which of the following classification models is built by kNN algorithm?

Decision tree classification model

Ensemble classification model

Hyperplane classification model

No classification model is built by kNN

Explanation:

The k-Nearest Neighbors algorithm is a fundamental instance-based learning method that classifies data points by identifying the k closest training examples in the feature space. It does not construct a complex mathematical structure like a decision tree or a hyperplane during a training phase. Instead, it simply stores the entire dataset and makes predictions by voting among the nearest neighbors at query time. Therefore, it is accurate to state that no explicit classification model is built in the traditional sense, as the "model" is essentially the dataset itself. This distinguishes it from parametric models that learn a specific function or boundary.

Question 7 Easy Mcq

✓ Correct ✗ Incorrect

In association rule mining, an indication of how often the rule has been found to be true is represented by a term known as confidence. How is this term. confidence. represented for the rule, A => B?

conf(A => B) = supp(A U B) / supp(A)

conf(A => B) = supp(B) / supp(A)

conf(A => B) = supp(A U B) / supp(A)‘ supp(B)

conf(A => B) = supp(A U B) / 1 - supp(A)

Explanation:

Confidence measures the reliability of a rule by calculating the conditional probability that item B appears given that item A is present. This is mathematically expressed as the ratio of the support for the combined items (A and B together) to the support of the antecedent (A) alone. Essentially, it tells us what fraction of all transactions containing A also contain B. A higher confidence value indicates a stronger association between the two items in the dataset. This formula directly quantifies how often the rule holds true whenever the condition A is met.

Question 8 Medium Mcq

✓ Correct ✗ Incorrect

Find the output of the following R programming language code.a <- c(7.5.FALSE.4+4i)b <- c(6,0,TRUE,4+7i)print(a&&b)

[1] FALSE

[1] FALSE TRUE

(1] FALSE FALSE

[1] TRUE

Explanation:

The code evaluates the logical AND operation (`&&`) between two vectors, which in R only processes the first element of each vector. Since the first element of `a` is `7.5` (treated as TRUE) and the first element of `b` is `6` (also treated as TRUE), the result is TRUE. The `&&` operator specifically returns a single logical value rather than recycling or comparing entire vectors element-wise.

Question 9 Medium Mcq

✓ Correct ✗ Incorrect

In which of the following types of reasoning in data science, the conclusions reached are probable,reasonable. plausible and believable?Deductive reasoningInductive reasoning

Only1

Only 2

Both 1 and 2

Neither1 nor 2

Explanation:

Inductive reasoning involves drawing general conclusions from specific observations, which inherently deals with probabilities rather than absolute certainties. In data science, this approach analyzes patterns in sample data to make plausible predictions about larger populations, ensuring results are reasonable and believable based on evidence. Unlike deductive reasoning that guarantees truth if premises are correct, inductive logic accepts that conclusions are likely true but not certain, fitting the description of probable outcomes perfectly. This probabilistic nature makes it the standard method for predictive modeling and data-driven decision-making where absolute proof is often unattainable. Consequently, the conclusions reached through this method are best characterized as probable and supported by empirical data.

Question 10 Easy Mcq

✓ Correct ✗ Incorrect

Which of the following options denotes the probability of avoiding a type-ll error in hypothesis testing?

84$ -1

(3 -1

1- (3

Explanation:

The probability of avoiding a Type II error is known as the statistical power of a test, which represents the likelihood of correctly rejecting a false null hypothesis. This value is mathematically defined as one minus the probability of committing a Type II error (denoted by beta). Consequently, the correct expression is 1 minus beta, which directly corresponds to the provided option D. Understanding this relationship is fundamental for designing experiments with sufficient sensitivity to detect true effects.

Question 11 Easy Mcq

✓ Correct ✗ Incorrect

Which of the following is the correct syntax of the PredictSupport (DMX) prediction function used with Microsoft linear regression algorithm?

PredictSupport(, [])

PredictSupport(l)

PredictSupport(I)

PredictSupport(l)

Explanation:

The PredictSupport function in Microsoft's DMX (Decision Miner eXtensions) is specifically designed to generate predictions for linear regression models. Its correct syntax requires passing the model instance as the sole argument within parentheses, which aligns with the standard function signature for this operation. This single-parameter structure allows the engine to efficiently calculate support values based on the provided model configuration. By adhering to this specific format, users ensure that the algorithm correctly interprets the input model to produce accurate regression predictions without ambiguity.

Question 12 Easy Mcq

✓ Correct ✗ Incorrect

What will be the output of the following code of the R programming language?b1 <- 17b2 <- 13z <— 5:7print(b1 96in96 z)print(b2 %in% z)

[1] TRUE[1] FALSE

[1] FALSE[1] TRUE

[1] TRUE[1] TRUE

[1] FALSE[1] FALSE

Explanation:

The code first creates a vector `z` containing integers from 5 to 7. The expression `b1 %in% z` checks if the value 17 exists within that vector, which returns FALSE since 17 is not between 5 and 7. Conversely, `b2 %in% z` checks if 13 is in the vector, also returning FALSE because 13 is outside the range 5 to 7. Consequently, both print statements output FALSE, making the final result two consecutive FALSE values.

Question 13 Easy Mcq

✓ Correct ✗ Incorrect

Which of the following is the correct R syntax used for selecting certain rows from a data frame, based on specif‌ic logical criteria?

select(dataframename, logical expression)

f‌ilter(logical expression, dataframename)

f‌ilter(dataframename, logical expression)

select(logical expression, dataframename)

Explanation:

The `filter()` function is the standard tool in R for subsetting rows based on logical conditions within a data frame. It operates by taking the data frame as the first argument, followed by one or more logical expressions that define which rows to keep. This function returns a new data frame containing only the observations where the specified criteria evaluate to TRUE, effectively narrowing the dataset without altering the original structure.

Question 14 Easy Mcq

✓ Correct ✗ Incorrect

In logistic regression. which of the given methods is used to display the conditional density plot of thebinary outcome, F. on the continuous x variable?

Explanation:

In logistic regression, the conditional density plot visualizes the probability of the binary outcome F across different values of the continuous predictor x. This method works by calculating the predicted probability for each x value and plotting it as a smooth curve, effectively showing how the likelihood of the event changes. It is the correct approach because it directly maps the non-linear relationship modeled by the logistic function onto a visual graph. This visualization helps analysts intuitively understand the impact of x on the probability of F without relying solely on coefficients. The resulting plot provides a clear, continuous representation of the model's predictions across the entire range of the variable.

Question 15 Easy Mcq

✓ Correct ✗ Incorrect

Which of the following statements is correct about the judgement sampling method?

There is no possibility of personal prejudice in this method.

It is more accurate and reliable.

It is mostly used in those fields where almost similar units exist or some units are tOO important' to be left out of the sample.

It is very expensive.

Question 16 Easy Mcq

✓ Correct ✗ Incorrect

Which Of the following t-tests should be performed in order to compare means from two different groups?

One sample t—test

Paired samples t—test

Independent samples t-test

Analysis Of Variance (ANOVA)

Explanation:

The independent samples t-test is specifically designed to compare the means of two distinct, unrelated groups by analyzing the difference between their sample means relative to their variability. This method assumes the data points in one group are independent of those in the other, making it the appropriate statistical tool when there is no natural pairing or matching between the subjects. It calculates a t-statistic to determine if the observed difference is statistically significant or likely due to random chance, providing a direct answer to whether the two separate populations have different average values.

Question 17 Easy Mcq

✓ Correct ✗ Incorrect

Which of the following data mining algorithms is applied to a database containing a large number of transactions and also learns association rules?

K-means

C45

Apriori

Explanation:

The Apriori algorithm is specifically designed to discover frequent itemsets and generate association rules within large transaction databases. It operates by iteratively identifying frequent itemsets based on the principle that all non-empty subsets of a frequent itemset must also be frequent. This process efficiently prunes the search space to avoid examining every possible combination of items. Consequently, it is the standard method for uncovering patterns like "customers who buy bread also buy butter" in market basket analysis. Its core functionality directly addresses the requirement of learning association rules from transactional data.

Question 18 Easy Mcq

✓ Correct ✗ Incorrect

With respect to advanced statistics, which of the following options is correct about the arimaO function?

It can be used to produce an unrotated principal component analysis.

It can be used to produce maximum likelihood factor analysis.

It can be used to bootstrap the structural equation model.

It can be used to f‌it an autoregressive integrated moving average model.

Explanation:

The arimaO function is specifically designed to fit Autoregressive Integrated Moving Average models to time series data. It estimates the parameters for the autoregressive and moving average components while handling non-stationarity through differencing. This function is a core tool in advanced statistics for analyzing temporal dependencies and forecasting future values based on past observations. By utilizing maximum likelihood estimation, it provides robust parameter estimates that capture the underlying structure of the time series process. Consequently, its primary purpose is directly aligned with fitting ARIMA models rather than factor analysis or bootstrapping structural equation models.

Question 19 Medium Mcq

✓ Correct ✗ Incorrect

Which of the following options is the default CLUSTERING_METHOD used by the Microsoft clustering algorithm?

Non-Scalable EM

Scalable EM

Scalable K-Means

Non-Scalable K—Means

Explanation:

The Microsoft clustering algorithm defaults to Scalable EM because it is specifically engineered to handle large datasets efficiently without requiring manual scaling adjustments. This method combines the robustness of the Expectation-Maximization approach with optimization techniques that allow it to converge faster on massive data volumes compared to standard implementations. By prioritizing scalability, it ensures that the clustering process remains practical and performant even when dealing with the high-dimensional data structures common in enterprise environments. Consequently, this default choice provides a balanced solution that maintains accuracy while significantly reducing computational overhead for big data scenarios.

Question 20 Easy Mcq

✓ Correct ✗ Incorrect

_______ reduces the number of bits in a file by identifying and eliminating redundancy

Lossless compression

Lossy compression

Bitmap

Data visualization

Explanation:

Lossless compression reduces file size by detecting and removing statistical redundancy without discarding any original data. It works by identifying repetitive patterns or sequences within the file and replacing them with shorter codes. This process ensures that the exact original information can be perfectly reconstructed during decompression. Consequently, it is the ideal method for compressing text, code, or sensitive files where every bit must be preserved. The technique effectively minimizes storage requirements while maintaining 100% data integrity.

Question 21 Medium Mcq

✓ Correct ✗ Incorrect

With respect to the Microsoft sequence clustering algorithm, which of the following options is the correct syntax of the PredictCaseLikelihood (DMX) function?

PredictCaseLikelihood()

PredictCaseLikelihood([NORMALIZEDINONNORMALIZEDD

PredictCaseLikelihood(, [])

PredictCaseLikelihood(, [))

Explanation:

The PredictCaseLikelihood function in Microsoft's DMX algorithm is designed to calculate the probability of a specific record belonging to a target cluster. Its correct syntax requires two mandatory arguments: the first specifies whether the input data is normalized or unnormalized, and the second provides the actual data array to be evaluated. This structured approach ensures the algorithm processes the input correctly, making option B the only valid representation of the function's required parameters. Understanding this specific signature is essential for accurately implementing predictive modeling tasks within the Microsoft Analysis Services environment.

Question 22 Medium Mcq

✓ Correct ✗ Incorrect

As per Microsoft association rules algorithm, which of the following Options is the prediction function with scalar value as the return type?

lsInNode (DMX)

PredictAssociation(DMX)

PredictAdjustedProbability(DMX)

PredictHistogram(DMX)

Explanation:

The PredictAdjustedProbability function is specifically designed to return a scalar value representing the adjusted probability of an itemset occurring, which is essential for decision-making in association rule mining. Unlike functions that return complex structures like histograms or node details, this algorithm directly outputs a single numerical probability score. It calculates the likelihood of a rule firing by adjusting for the baseline probability of the consequent items, providing a refined metric for rule strength. This scalar output allows analysts to easily rank rules or set specific probability thresholds without parsing additional data structures. Consequently, it serves as the primary prediction tool when a simple, direct probability value is required from the Microsoft association rules model.

Question 23 Medium Mcq

✓ Correct ✗ Incorrect

As per Microsoft association rules algorithm, which of the following prediction functions has/have a Boolean return type?

lslnNode()

lsDescendant()

PredictNodeld (DMX)

Both a and b

Explanation:

In the context of Microsoft Association Rules, both lslnNode() and lsDescendant() are specialized functions designed to navigate the hierarchical structure of the generated rule tree. These specific utilities return Boolean values to indicate the existence or relationship of nodes within the rule set, rather than returning complex data structures or numeric scores. This Boolean return type allows for efficient logical checks within DMX scripts to validate rule conditions or traverse paths without unnecessary data overhead. Consequently, the prediction function that supports this specific Boolean behavior encompasses both of these node traversal methods.

Question 24 Easy Mcq

✓ Correct ✗ Incorrect

Data types that are created by the programmer are known as ________.

Variables

Abstract data types (ADTs)

Functions

Parameters

Explanation:

Abstract data types (ADTs) are custom structures defined by programmers to solve specific problems, distinct from built-in types like integers. By defining a new ADT, a developer encapsulates both data and operations, creating a logical model that hides internal implementation details. This process allows for the creation of specialized types such as stacks or queues that behave consistently regardless of their underlying storage mechanism. Consequently, these user-defined types are the primary way programmers extend a language's capabilities beyond standard primitives. They represent a fundamental concept in software engineering for organizing complex logic into reusable, well-defined components.

Question 25 Easy Mcq

✓ Correct ✗ Incorrect

In data mining, which of the following statements is NOT correct about C45 algorithm?

It allows only one outcome.

A single-pass algorithm derived from binomial conf‌idence limits is used by C45.

It uses information-based criteria.

Explanation:

The C4.5 algorithm is specifically designed for classification tasks, meaning it inherently handles multiple possible class outcomes rather than restricting itself to a single one. Unlike binary classifiers, it builds decision trees capable of predicting any number of target classes by evaluating information gain ratios across all potential splits. Consequently, the statement claiming it allows only one outcome is factually incorrect because its core functionality relies on managing multi-class datasets effectively. This fundamental characteristic distinguishes it from algorithms strictly limited to binary decisions. Therefore, option A is the only incorrect statement regarding the algorithm's capabilities.

Question 26 Medium Mcq

✓ Correct ✗ Incorrect

In advanced statistics, which of the following statements is correct about the Dirichlet Regression method?

It is used to model binary variables.

It is used to model compositional data.

It is used to model rank variables.

It is used to model count variables.

Explanation:

The Dirichlet regression method is specifically designed to handle compositional data, where observations are parts of a whole that sum to a constant, such as 1 or 100%. Unlike standard regression models that assume independent variables, this approach respects the inherent constraint that the components must sum to unity, making it ideal for analyzing proportions like market shares or chemical compositions. By modeling the log-ratios of these parts, it effectively captures the complex dependencies between variables that standard linear models cannot address. This specialized technique ensures that the statistical inferences remain valid within the bounded nature of compositional datasets.

Question 27 Easy Mcq

✓ Correct ✗ Incorrect

Using the following information, find the correct syntax of the R function used for creating binary f‌iles.Assume object as the binary file to be written. n as the number Of bytes and con as the connection object.

writeBin(object, n, con)

writeBin(object)

writeBin(object, n)

writeBin(object, con)

Explanation:

The writeBin function in R is specifically designed to write binary data to a connection object, making the 'con' argument essential for defining the output destination. While the 'object' parameter holds the data to be written, the 'n' argument specifies the number of bytes to write in each call, which is optional and defaults to writing all available data. Therefore, the minimal required syntax only mandates the data object and the connection, correctly represented by the option containing these two specific arguments. This design allows for flexible writing operations where users can either write the entire object at once or control the chunk size explicitly.

Question 28 Easy Mcq

✓ Correct ✗ Incorrect

The values of X and Y are given in f‌igure-1 Of the image. Choose the correct value of 2X — 5Y fromfigure-2.

Explanation:

To solve this, first identify the specific numerical values for X and Y directly from the data points shown in Figure-1. Substitute these exact numbers into the algebraic expression 2X - 5Y to perform the necessary arithmetic operations. The resulting calculation yields a specific numerical outcome that corresponds precisely to the value labeled as 'c' in Figure-2. This process demonstrates how substituting known variables into a linear equation allows for the direct determination of the unknown result.

Question 29 Easy Mcq

✓ Correct ✗ Incorrect

It is given that y is a Poisson variate and satisf‌ies the condition P(y=4) = P(y=5). What are the values of mean and standard deviation of y?

Mean = 3 and standard deviation = /5

Mean = 5 and standard deviation = /3

Mean = 5 and standard deviation = /’5

Mean = 7 and standard deviation = /7

Explanation:

For a Poisson distribution, the probability mass function is proportional to $\lambda^y$, where $\lambda$ represents the mean. Given the condition $P(y=4) = P(y=5)$, we equate the terms $\frac{e^{-\lambda}\lambda^4}{4!}$ and $\frac{e^{-\lambda}\lambda^5}{5!}$. Simplifying this equality by canceling common factors leads to the relationship $\lambda = 5$. In a Poisson distribution, the variance is equal to the mean, so the variance is 5. Consequently, the standard deviation is the square root of the variance, which is $\sqrt{5}$.

Question 30 Medium Mcq

✓ Correct ✗ Incorrect

In association rule mining, which of the following statements is correct about Frequent Itemset Generation of the two-step approach?

Generates only one itemset whose support 2 minsup

Generates all itemsets whose support 5 minsup

Generates high confidence rules from each frequent itemset

Explanation:

The two-step approach in association rule mining first focuses on generating all frequent itemsets that meet a minimum support threshold, effectively filtering out rare combinations early. This initial phase ensures that only statistically significant patterns are retained for the subsequent rule generation step. By identifying these frequent itemsets, the algorithm lays the necessary foundation for later calculating confidence levels to derive high-quality association rules. This method optimizes computational efficiency by avoiding unnecessary processing of infrequent patterns. Consequently, the primary output of this first stage is the complete set of itemsets satisfying the support constraint.

Question 31 Easy Mcq

✓ Correct ✗ Incorrect

ln data mining, according to Bayes‘ theorem, which of the following formulae represents posterior probability in terms of prior probability?

P(X/H) = P(H/X)P(H)/P(X)

P(H/X) = P(X/H)P(H)/P(X)

P(H/X) = P(X/H)P(X)/P(H)

P(XIH) = P(H/X)/P(H)P(X)

Explanation:

Bayes' theorem provides a mathematical framework to update the probability of a hypothesis based on new evidence. The formula P(H/X) = P(X/H)P(H)/P(X) explicitly defines the posterior probability P(H/X) as the product of the likelihood P(X/H) and the prior probability P(H), normalized by the total probability of the evidence P(X). This equation demonstrates how prior beliefs are adjusted by observed data to yield a more accurate posterior estimate. It is the fundamental principle behind probabilistic reasoning in data mining and machine learning algorithms.

Question 32 Medium Mcq

✓ Correct ✗ Incorrect

Regression equation of Z on V is given as following:7. = c + dVThe relationship between two variables a and b, is given as b + 6a = 20 and between another two variables c and d, as 4c + 10d = 50. The regression coefficient of c on a is given as 0.90. Find the regression coefficient of d on b.

2/49

3/50

3/74

5/29

Explanation:

The regression coefficient of $c$ on $a$ is derived from the ratio of their covariance to the variance of $a$, which simplifies to the ratio of their standard deviations scaled by their correlation. By rewriting the given linear equations $b + 6a = 20$ and $4c + 10d = 50$, we can express $b$ and $d$ in terms of $a$ and $c$ respectively, revealing that the regression coefficient of $d$ on $b$ depends on the scaling factors of these transformations. Specifically, the coefficient is the product of the original regression coefficient (0.90) and the ratio of the scaling factors ($4/6$ multiplied by the inverse of the other transformation), which mathematically resolves to $3/50$. This demonstrates how linear transformations of variables proportionally adjust the regression slope based on the coefficients of the defining equations.

Question 33 Easy Mcq

✓ Correct ✗ Incorrect

In which of the following text mining methods, terms are analyzed on the sentence and document level?

Phrase-Based Method (PBM)

Term-Based Method (TBM)

Pattern Taxonomy Method (PTM)

Concept-Based Method (CBM)

Explanation:

The Concept-Based Method (CBM) analyzes terms by grouping them into semantic concepts that span entire sentences and documents, rather than focusing on isolated words or short phrases. This approach captures the broader context and meaning of the text, allowing for a comprehensive understanding of the document's overall topic. By operating at both the sentence and document levels, CBM effectively identifies high-level themes and relationships that lower-level methods might miss. This holistic analysis makes it the appropriate choice when the goal is to extract meaning across the full scope of a text corpus. Consequently, it is the correct method for analyzing terms at these specific hierarchical levels.

Question 34 Medium Mcq

✓ Correct ✗ Incorrect

For a given set of 25 items, coeff‌icient of correlation between x and y is 0.6. The values of the arithmetic mean of x and y are 14 and 18, respectively, and the values of standard deviation of x and y are 4 and 6. respectively. If the pair (25. 18) has been wrongly taken as (18, 25). then find the correct value of correlation coeff‌icient.

0.31

0.42

0.51

0.67

Explanation:

The correct answer is derived by recalculating the correlation coefficient after removing the erroneous data point (25, 18) and adding the correct one (18, 25). Since the original correlation is positive (0.6), the incorrect point (25, 18) lies far from the expected linear trend, artificially inflating the strength of the relationship. Replacing it with the correct point (18, 25), which aligns better with the mean values, reduces this distortion. Consequently, the new correlation coefficient decreases from the original 0.6 to approximately 0.51, reflecting the true linear association between the variables.

Question 35 Easy Mcq

✓ Correct ✗ Incorrect

Which of the following clustering algorithms is used for grid-based partitioning?

BIRCH

K-means

STING

FCM

Explanation:

STING is a hierarchical clustering algorithm specifically designed for grid-based partitioning of spatial data. It works by recursively dividing the spatial database into a grid structure and building a statistical summary for each cell, which allows for efficient querying and clustering of large datasets. This grid-centric approach distinguishes it from density-based or centroid-based methods, making it the definitive choice for this specific partitioning strategy.

Question 36 Medium Mcq

✓ Correct ✗ Incorrect

For a group of 12 students, the sum of squares of differences in their ranks for science and math is given as 60. On the basis of the given information. find the value of rank correlation coefficient.

0.60

0.79

0.45

0.82

Explanation:

The rank correlation coefficient is calculated using Spearman's formula, which relates the sum of squared rank differences to the sample size. By substituting the given values of n=12 and Σd²=60 into the formula, the calculation simplifies to 1 minus a fraction derived from these inputs. Performing this arithmetic operation yields a precise value of 0.79, confirming that this specific numerical result is the correct representation of the correlation strength for this dataset.

Question 37 Easy Mcq

✓ Correct ✗ Incorrect

IN SOL Server data mining, which of the following algorithm types predicts one or more discrete variables that are based on other attributes in a dataset?

Segmentation algorithm

Classif‌ication algorithm

Sequence analysis algorithm

Association algorithm

Explanation:

Classification algorithms are specifically designed to predict discrete target variables, known as classes or labels, by analyzing patterns within the dataset's other attributes. These models learn from labeled training data to build a function that maps input features to a specific categorical output, such as identifying whether an email is spam or not. This process relies on statistical learning to generalize from known examples, making it the ideal choice for any task requiring the prediction of distinct categories based on existing data characteristics.

Question 38 Easy Mcq

✓ Correct ✗ Incorrect

Diigo and delicious are ________ tools.

Social bookmarking

Research

Discussion group

Synchronous communication

Explanation:

Diigo and Delicious are both widely recognized platforms designed for saving, organizing, and sharing web resources online. They allow users to tag and categorize bookmarks, making it easy to retrieve specific websites later or share collections with others. This core functionality defines them as social bookmarking tools that facilitate collaborative information management. By enabling users to build a shared library of links, these services enhance research efficiency through community-driven organization. Their primary purpose revolves around managing digital references rather than real-time chat or group discussions. Therefore, they fit perfectly into the category of social bookmarking applications.

Question 39 Easy Mcq

✓ Correct ✗ Incorrect

In survival analysis, which of the following methods is used to model the hazard function on a set of predictor variables?

Surv()

coxph()

survdiff()

survf‌it()

Explanation:

The coxph() function implements the Cox proportional hazards model, which is specifically designed to estimate the effect of several variables on survival time. It models the hazard function as a product of a baseline hazard and an exponential term involving the linear combination of predictor variables. This semi-parametric approach allows researchers to assess how covariates influence the risk of an event without needing to specify the underlying baseline hazard distribution. Consequently, it is the standard tool for multivariate survival analysis involving multiple predictors.

Question 40 Medium Mcq

✓ Correct ✗ Incorrect

Which of the following statements is NOT correct about data science?

It is used for turning data into actions.

It supports and encourages shifting between deductive and inductive reasoning.

In order to achieve success. organizations need to reach maximum data science maturity.

It is necessary for companies to stay with the pack and compete in future.

Explanation:

Data science maturity is a continuous journey rather than a fixed destination, meaning organizations can still achieve significant success without reaching the theoretical maximum level immediately. The statement is incorrect because it falsely implies that peak maturity is a strict prerequisite for any success, whereas incremental improvements and practical application of data insights often drive value. Success depends on effectively leveraging available data and tools at any stage of development, not solely on attaining the highest possible maturity score. Therefore, the requirement to reach maximum maturity before succeeding is an inaccurate constraint on organizational growth.

Question 41 Easy Mcq

✓ Correct ✗ Incorrect

Find the output of the following code of the R programming language.Iista <- Iist(5:7)print(lista)Iistb <-Iist(12:14)print(listb)x1 <- unlist(lista)x2 <- unlist(listb)print(xl)print(x2)r <- x1+x2print(r)

[[1]][1] 5 6 7[[1]][1] 12 13 14[1] 5 6 7[1] 12 13 14[1] 17 19 21

[[1]][1] 5 6 7[[1]][1] 12 13 14[1] 5 6 7[1] 12 13 14[1] 15 16 17

[[1]][1] 5 6 7[[1]][1] 12 13 14[1] 5 6 7[1] 12 13 14[1] 24 25 26

[[1]][1] 5 6 7[[1]][1] 12 13 14[1] 5 6 7[1] 12 13 14[1] 11 12 13

Explanation:

The code creates two lists containing vectors 5:7 and 12:14, which are printed as nested structures. Using unlist() flattens these lists into simple numeric vectors x1 and x2. Adding x1 and x2 performs element-wise summation, resulting in the vector 17, 19, 21, which matches the final output line in option A.

Question 42 Easy Mcq

✓ Correct ✗ Incorrect

As per the Microsoft association rules algorithm, which of the following parameters specif‌ies the minimum number of cases that must contain an itemset before the algorithm generates a rule?

MINIMUM_SUPPORT

MINIMUM_PROBABILITY

MINIMUM_ITEMSET_SIZE

MINIMUM_ITEMSET_COUNT

Explanation:

In the Microsoft Association Rules algorithm, the parameter that defines the minimum number of cases containing an itemset is Minimum Support. This threshold acts as a filter to ensure that only itemsets appearing frequently enough in the dataset are considered valid for rule generation. By setting this count, the algorithm effectively eliminates rare or noise patterns that do not represent meaningful relationships. Consequently, any itemset meeting or exceeding this specified count is eligible to be part of the generated association rules. This mechanism is fundamental to maintaining the statistical significance and practical utility of the discovered patterns.

Question 43 Easy Mcq

✓ Correct ✗ Incorrect

According to advanced statistics generalized linear model, which of the following is the default link function for the gaussian family?

(link = '’identity")

(link = '’Iogit")

(link = ‘'Iog")

(link = ”inverse")

Explanation:

In generalized linear models for continuous data, the Gaussian family assumes a normal distribution where the mean is directly modeled by the linear predictor. The identity link function is the default because it maps the expected value of the response variable directly to the linear combination of coefficients without any transformation. This allows the model to predict any real-valued outcome, which aligns perfectly with the properties of normally distributed data. Consequently, no additional transformation like log or inverse is needed to satisfy the model's assumptions.

Question 44 Medium Mcq

✓ Correct ✗ Incorrect

What will be the output of the following R code?c(4,7,TRUE,3+7i) -> v1c(9,6,FALSE,3+7i) ->> v2print(v1)print(v2)

[114+01 4+1i 7+01 3+7i[1] 9+0i 9+1i 6+0i 3+7i

[1]4+0i7+0i1+0i3+7i[1] 9101 6+0i 0+01 3+7i

[1) 4+0i 7+7l1+1i 3+7i[1] 9+Oi 9+1i 6+6i 3+7i

[1]4+4i7+7i1+1i3+7i[119+9i 6+6i1+1i 3+7i

Explanation:

In R, the `c()` function combines elements into a vector, automatically converting logical values to integers (TRUE becomes 1, FALSE becomes 0) and numeric values to complex numbers with zero imaginary parts (e.g., 4 becomes 4+0i). The code creates two vectors, `v1` and `v2`, where the logical inputs are coerced to 1 and 0 respectively, and all numbers are displayed in complex format. The `print()` function outputs these vectors with their standard `[1]` index labels, resulting in the sequence shown in option B where logicals appear as 1+0i or 0+0i and integers as x+0i.

Question 45 Medium Mcq

✓ Correct ✗ Incorrect

Which of the following statements are NOT correct about the Bayesian belief network?

L1 In a belief network, class conditional independencies can be defined between the subsets of variables.

VJ Joint conditional probability distribution cannot be specif‌ied by Bayesian belief networks.

VJ A trained Bayesian network cannot be used for classification.

VJ A graphical model of casual relationship for performing learning is provided by Bayesian belief network.

Explanation:

Bayesian belief networks explicitly model the joint conditional probability distribution of all variables by combining local conditional probabilities with the chain rule, directly contradicting the claim that it cannot specify such distributions. These networks effectively capture complex dependencies and independencies within the data structure, making them powerful tools for representing uncertainty. By defining a directed acyclic graph, they provide a compact representation of the full joint distribution rather than failing to specify it. Consequently, the statement denying this fundamental capability is factually incorrect, as the network's primary function is precisely to define these joint probabilities. Therefore, this option is the one that is not correct regarding the capabilities of Bayesian belief networks.

Question 46 Easy Mcq

✓ Correct ✗ Incorrect

Which of the following is the correct default value for the INSTABILITY_SENSITIVITY parameter used with the Microsoft time series algorithm?

0.6

0.1

Explanation:

The INSTABILITY_SENSITIVITY parameter controls how the algorithm responds to sudden changes or structural breaks in the time series data. A default value of 1 ensures the model maintains a balanced sensitivity, allowing it to adapt effectively to evolving patterns without overreacting to minor fluctuations. This setting provides a robust baseline for most forecasting scenarios where moderate responsiveness to data shifts is desired. By defaulting to this value, the algorithm optimizes its performance across diverse datasets without requiring immediate manual tuning.

Question 47 Easy Mcq

✓ Correct ✗ Incorrect

Which of the following is the correct syntax of the command used for merging two data frames, myFrame1 and myFrame2, by ID and Country?

total <~ merge(data myFrame1 with myFrame2, by=c(lD,Country))

total <- merge(data myFrame1,data myFrame2,by=c("lD","Country"))

total <- merge(data by=c("lD","Country") for myFrame1, myFrame2)

total <- merge(data for myFrame1, myFrame2,by=C(lD,Country))

Explanation:

The correct syntax uses the standard R assignment operator `<-` to store the result in a new variable. The `merge()` function requires the data frames to be passed as separate arguments, followed by the `by` parameter specifying the common columns. Proper quoting of column names like "ID" and "Country" ensures the function interprets them as character strings rather than variable names, which is essential for accurate merging operations.

Question 48 Easy Mcq

✓ Correct ✗ Incorrect

Consider the following data:Average cost of wafers = Rs. 35Average cost of chocolates = Rs. 37Standard deviation of cost of wafers = 2.0Standard deviation of cost of chocolates = 3.0Correlation coeff‌icient between the costs of chocolates and wafers = 0.7What will be the expected cost of chocolates when the cost of wafers is Rs. 40?

Rs. 42.25

Rs. 45

Rs. 39.85

Rs. 41.75

Explanation:

To find the expected cost, we use the linear regression equation derived from the correlation, means, and standard deviations. First, calculate the slope by multiplying the correlation coefficient (0.7) with the ratio of the standard deviations (3.0 divided by 2.0), which yields a slope of 1.05. Next, determine the intercept by subtracting the product of the slope and the wafer mean (1.05 times 35) from the chocolate mean (37), resulting in an intercept of 0.175. Finally, substitute the given wafer cost of 40 into the equation to get the predicted chocolate cost of 42.25.

Question 49 Easy Mcq

✓ Correct ✗ Incorrect

Which of the following options is correct about the logistic regression technique?

It is used for encouraging group effect in case of highly correlated variables.

It is used for finding the probability of event=Success and event=Failure.

It is used for adding and removing predictors as needed for each step.

It is used for penalizing the absolute size of the regression coeff‌icients.

Explanation:

Logistic regression is specifically designed for binary classification problems where the outcome variable has two possible states, such as success or failure. Instead of predicting a continuous value, it estimates the probability that an observation belongs to a particular class using the logistic function. This technique models the log-odds of the event occurring as a linear combination of the predictor variables. By outputting a value between 0 and 1, it effectively quantifies the likelihood of the event happening versus not happening. Therefore, it is the standard method for determining the probability of an event like success or failure.

Question 50 Medium Mcq

✓ Correct ✗ Incorrect

Which of the following is the default value of the parameter HISTORICAL_MODEL_GAP used in Microsoft time series algorithm?

Explanation:

The HISTORICAL_MODEL_GAP parameter determines the number of historical data points utilized to train the forecasting model before generating predictions. A gap value of 10 ensures the algorithm incorporates a sufficient window of past observations to capture underlying trends and seasonality effectively. This default setting balances computational efficiency with the need for adequate historical context, allowing the model to produce stable and accurate forecasts without overfitting to recent noise. By using ten periods, the system avoids relying solely on the most immediate data, which might not yet reflect long-term patterns. Consequently, this value serves as a robust starting point for most standard time series scenarios within the Microsoft algorithm.

Showing 50 of 158 questions

Filter by Difficulty

easy medium hard

Track Progress

Difficulty Filter

easy Easy

medium Medium

hard Hard

Other Subjects

Business Finance & Management

Industrial Engineering & Manufacturing

IT & Programming

Design & Multimedia

Content Writing & Translation

Administrative Support

Quick Navigation

Other Industrial Engineering & Manufacturing Subjects

Data Analytics - Industrial Engineering & Manufacturing

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Explanation:

Filter by Difficulty

Track Progress

Difficulty Filter

Other Subjects

Other Industrial Engineering & Manufacturing Subjects

Quick Navigation

Similar Practices