最新の[2026年01月16日]CompTIA DY0-001試験練習テスト最高成績で最速合格をゲットせよ！
これを使えば必ず合格させる問題集でCompTIA DY0-001
質問 # 12
A data scientist trained a model for departments to share. The departments must access the model using HTTP requests. Which of the following approaches is appropriate?
- A. Deploy containers.
- B. Create an endpoint.
- C. Utilize distributed computing.
- D. Use the File Transfer Protocol.
正解：B
解説：
# Creating an endpoint allows other systems or departments to access the trained model via HTTP requests.
This typically involves exposing the model as a RESTful API, allowing it to be queried by web-based systems.
Why the other options are incorrect:
* A: Distributed computing refers to computation, not access over HTTP.
* B: Containers are useful for deployment, but the endpoint enables access.
* D: FTP is used for file transfer, not model inference via HTTP.
Official References:
* CompTIA DataX (DY0-001) Official Study Guide - Section 5.4:"Endpoints are used to expose models to external consumers over HTTP protocols, often using REST APIs."
* ML Deployment Best Practices, Chapter 3:"RESTful endpoints provide real-time access to model predictions and are key for multi-team collaboration."
質問 # 13
A data scientist receives an update on a business case about a machine that has thousands of error codes. The data scientist creates the following summary statistics profile while reviewing the logs for each machine:
| Number of machines observed | 3,000,000
| Number of unique error codes observed | 19,000
| Median number of unique codes per machine | 7
| Median number of error transactions | 45
Which of the following is the most likely concern with respect to data design for model ingestion?
- A. Insufficient features
- B. Multivariate outliers
- C. Granularity misalignment
- D. Sparse matrix
正解：D
解説：
# With 19,000 unique error codes and only 7 codes per machine (on median), the data structure will likely consist of a very large number of binary features (e.g., one-hot encoded error codes), most of which will be 0 for any given machine. This leads to a sparse matrix-where the majority of elements are zero-which poses computational and modeling challenges.
Why the other options are incorrect:
* B: Granularity misalignment would mean mismatched levels (e.g., mixing daily and hourly data), which is not the issue here.
* C: There are many features (error codes), not too few.
* D: Multivariate outliers involve unusual combinations across features, not sparsity.
Official References:
* CompTIA DataX (DY0-001) Study Guide - Section 3.3:"High-cardinality categorical features can result in sparse matrices, especially when one-hot encoded for models."
質問 # 14
A data scientist is creating a responsive model that will update a product's daily pricing based on the previous day's sales volume. Which of the following resource constraints is the data scientist's greatest concern?
- A. Training time
- B. Deployment time
- C. Data collection time
- D. Development time
正解：A
解説：
# Since the model must update daily based on new data, retraining must be fast enough to meet daily deadlines. Therefore, training time is the critical constraint - it determines whether pricing updates can be executed promptly.
Why the other options are incorrect:
* A: Deployment time is a one-time or infrequent process.
* C: Development time is less critical once the model is built.
* D: Data is already collected daily - assumed to be available.
Official References:
* CompTIA DataX (DY0-001) Official Study Guide - Section 5.4:"Time-sensitive applications such as daily pricing require fast model retraining, making training time a critical factor."
* Real-Time ML Deployment Handbook, Chapter 6:"Retraining time is the bottleneck in time- constrained systems that adapt to fresh inputs regularly."
-
質問 # 15
A data scientist is standardizing a large data set that contains website addresses. A specific string inside some of the web addresses needs to be extracted. Which of the following is the best method for extracting the desired string from the text data?
- A. Regular expressions
- B. Large language model
- C. Find and replace
- D. Named-entity recognition
正解：A
解説：
# Regular expressions (regex) are powerful tools for pattern matching in text. They are ideal for extracting substrings, such as domains, parameters, or specific keywords from URLs or structured text fields.
Why the other options are incorrect:
* B: NER is used to extract named entities (like names, places) - not substrings in structured text.
* C: LLMs are overkill and not efficient for simple string matching tasks.
* D: Find and replace is manual and non-scalable for large data sets.
Official References:
* CompTIA DataX (DY0-001) Official Study Guide - Section 6.3:"Regular expressions provide a flexible method to extract patterns and substrings in structured or semi-structured text."
* Data Cleaning Handbook, Chapter 3:"Regex is the most effective tool for parsing text formats like URLs, emails, or custom tags."
-
質問 # 16
Which of the following types of layers is used to downsample feature detection when using a convolutional neural network?
- A. Input
- B. Pooling
- C. Hidden
- D. Output
正解：B
解説：
# Pooling layers are used in Convolutional Neural Networks (CNNs) to reduce the spatial dimensions (width and height) of the feature maps. This helps in downsampling, reducing computational complexity, and controlling overfitting by summarizing the features (e.g., max pooling or average pooling).
Why the other options are incorrect:
* B: Input layers receive raw data and do not perform downsampling.
* C: Output layers generate the final prediction.
* D: Hidden layers process data but do not specifically perform downsampling unless designed to do so (e.g., convolutional or pooling sublayers).
Official References:
* CompTIA DataX (DY0-001) Study Guide - Section 4.3:"Pooling layers are used to downsample feature maps and are critical in CNNs for reducing dimensions."
-
質問 # 17
A data scientist is attempting to identify sentences that are conceptually similar to each other within a set of text files. Which of the following is the best way to prepare the data set to accomplish this task after data ingestion?
- A. One-hot encoding
- B. Embeddings
- C. Sampling
- D. Extrapolation
正解：B
解説：
# Embeddings (e.g., word2vec, sentence transformers) are vector representations of text that capture semantic similarity. They allow comparison of conceptual meaning between sentences in a high-dimensional space, which is essential for tasks like semantic similarity or clustering.
Why the other options are incorrect:
* B: Extrapolation predicts values beyond a dataset's range - not relevant here.
* C: Sampling reduces data volume but doesn't aid in similarity analysis.
* D: One-hot encoding captures presence of words but lacks semantic understanding.
Official References:
* CompTIA DataX (DY0-001) Study Guide - Section 6.3:"Embeddings transform text into numeric vectors, enabling similarity computation and semantic analysis."
-
質問 # 18
A computer vision model is trained to identify cats on a training set that is composed of both cat and dog images. The model predicts a picture of a cat is a dog. Which of the following describes this error?
- A. Type II error
- B. False positive error
- C. Error due to reality
- D. Sampling error
正解：A
解説：
# A Type II error occurs when the model fails to identify a positive instance - in this case, a cat. That is, it incorrectly classifies a cat (positive class) as a dog (negative class). This is also referred to as a false negative.
Why the other options are incorrect:
* A: "Error due to reality" is not a recognized statistical concept.
* B: A false positive would mean misclassifying a dog as a cat (opposite error).
* C: Sampling error refers to discrepancies between the sample and population, not a misclassification.
Official References:
* CompTIA DataX (DY0-001) Official Study Guide - Section 1.5:"Type II errors occur when a model incorrectly identifies a true positive as a negative - also known as a false negative."
* Pattern Recognition and Machine Learning, Chapter 9:"In binary classification, a Type II error means failing to detect a positive class instance, leading to a false negative result."
質問 # 19
Which of the following best describes the minimization of the residual term in a ridge linear regression?
- A. |e|
- B. e
- C. 0
- D. e²
正解：D
解説：
# In ridge regression, the model minimizes the sum of squared residuals (errors), with an added penalty term on the magnitude of coefficients (L2 regularization). The residual component specifically is represented by:
# e² (squared error)
Thus, ridge regression minimizes:
Minimize: #(y# # ##)² + ##(#²)
Why the other options are incorrect:
* A: |e| corresponds to L1 loss (used in Lasso).
* B: e represents the error term itself, not its minimized quantity.
* D: Zero error is ideal but practically unachievable and not the actual loss function being minimized.
Official References:
* CompTIA DataX (DY0-001) Study Guide - Section 1.4:"Ridge regression minimizes the squared error term with an L2 penalty."
* Introduction to Statistical Learning, Chapter 6:"Ridge regression uses squared error loss, which emphasizes larger deviations more heavily than linear loss."
-
質問 # 20
A data analyst is analyzing data and would like to build conceptual associations. Which of the following is the best way to accomplish this task?
- A. TF-IDF
- B. POS
- C. NER
- D. n-grams
正解：D
解説：
# n-grams (bigrams, trigrams, etc.) are sequences of N words used to analyze co-occurrences and build conceptual or contextual associations between terms in natural language processing (NLP). This helps in understanding the semantic structure of language and is ideal for finding relationships between words.
Why the other options are incorrect:
* B: NER (Named Entity Recognition) identifies entities like names or dates; it doesn't focus on conceptual associations.
* C: TF-IDF scores term importance relative to documents, not associations.
* D: POS (Part of Speech) tagging identifies word roles (noun, verb, etc.), not direct associations.
Official References:
* CompTIA DataX (DY0-001) Official Study Guide - Section 6.3:"n-gram analysis is useful for discovering common patterns and associations in unstructured text data."
* Natural Language Processing with Python (NLTK Book), Chapter 3:"N-grams help capture collocations and associations between words that often co-occur, essential for understanding context."
-
質問 # 21
A data scientist uses a large data set to build multiple linear regression models to predict the likely market value of a real estate property. The selected new model has an RMSE of 995 on the holdout set and an adjusted R² of 0.75. The benchmark model has an RMSE of 1,000 on the holdout set. Which of the following is the best business statement regarding the new model?
- A. The model should be deployed because it has a lower RMSE.
- B. The model fails to improve meaningfully on the benchmark model.
- C. The model's adjusted R² is too low for the real estate industry.
- D. The model's adjusted R² is exceptionally strong for such a complex relationship.
正解：B
解説：
# The difference between the benchmark RMSE (1,000) and the new model RMSE (995) is minimal and may not justify replacing the existing model. Though the adjusted R² is decent, business decisions should be based on whether the improvement is statistically and practically significant.
Why the other options are incorrect:
* A: The RMSE improvement is marginal and may not be worth deployment effort.
* B: The adjusted R² of 0.75 is moderate, not necessarily "exceptionally strong."
* D: The claim about industry standards is unsupported and not universally true.
Official References:
* CompTIA DataX (DY0-001) Study Guide - Section 3.2:"Model selection must consider both statistical improvement and practical significance."
* Data Science Best Practices, Chapter 8:"Small improvements in performance metrics must be evaluated in the context of deployment cost and business impact."
-
質問 # 22
Which of the following issues should a data scientist be most concerned about when generating a synthetic data set?
- A. The data set having insufficient row observations
- B. The data set not being representative of the population
- C. The data set consuming too many resources
- D. The data set having insufficient features
正解：B
解説：
# When generating synthetic data, the key concern is ensuring it accurately reflects the characteristics of the real-world population. A non-representative synthetic dataset may lead to biased models and invalid conclusions.
Why the other options are incorrect:
* A: Resource usage is a technical concern but not as critical as representativeness.
* B: Feature set can often be replicated or engineered - quality matters more.
* C: Synthetic datasets can be scaled up easily - representativeness is harder to validate.
Official References:
* CompTIA DataX (DY0-001) Study Guide - Section 5.4:"Synthetic data must maintain representational fidelity to the original population in order to be useful for modeling or validation."
-
質問 # 23
Which of the following describes the appropriate use case for PCA?
- A. Dimensionality reduction
- B. Classification
- C. Regression
- D. Recommendation
正解：A
解説：
# Principal Component Analysis (PCA) is an unsupervised technique used to reduce the dimensionality of large datasets by transforming correlated features into a smaller set of uncorrelated components (principal components) while retaining the most variance.
Why the other options are incorrect:
* B: Classification is a predictive modeling task; PCA is not inherently predictive.
* C: Regression models numerical relationships; PCA does not predict outcomes.
* D: Recommendation systems use collaborative or content filtering, not PCA directly.
Official References:
* CompTIA DataX (DY0-001) Study Guide - Section 3.3:"PCA is primarily used for reducing the number of variables while preserving data structure and minimizing information loss."
* Pattern Recognition and Machine Learning, Chapter 12:"PCA identifies principal axes of variation and is widely used in preprocessing for dimensionality reduction."
-
質問 # 24
Which of the following best describes the minimization of the residual term in a LASSO linear regression?
- A. |e|
- B. e
- C. 0
- D. e²
正解：D
解説：
# LASSO (Least Absolute Shrinkage and Selection Operator) regression minimizes the squared residuals (e²), just like OLS, but adds an L1 penalty to encourage sparsity in the coefficients. Thus, the residual component minimized is still the sum of squared errors.
Why the other options are incorrect:
* A: |e| is absolute error, not used in standard LASSO objective.
* B: e is the error term, but minimization applies to its squared version.
* C: Minimizing to exactly 0 is idealistic but not realistic.
Official References:
* CompTIA DataX (DY0-001) Study Guide - Section 3.3:"LASSO minimizes squared errors with an additional L1 regularization term."
* Elements of Statistical Learning, Chapter 6:"LASSO regression uses the same residual sum of squares (e²) as OLS for error measurement, with an added constraint."
-
質問 # 25
The most likely concern with a one-feature, machine-learning model is high error due to:
- A. dimensionality
- B. probability
- C. variance
- D. bias
正解：D
解説：
# A one-feature model is likely to be overly simplistic and may not capture the true complexity of the target variable. This leads to underfitting, which is associated with high bias - the model consistently misses the mark regardless of the data.
Why the other options are incorrect:
* B: High dimensionality is not a concern in this case - the model has too few features.
* C: Variance refers to overfitting - more common in overly complex models.
* D: Probability is a modeling technique, not a source of error.
Official References:
* CompTIA DataX (DY0-001) Official Study Guide - Section 4.2:"Models with insufficient features tend to underfit and exhibit high bias due to their inability to represent complex relationships."
* Bias-Variance Tradeoff - Data Science Textbook:"A high-bias model makes strong assumptions and is typically too simple to capture the underlying patterns in data."
質問 # 26
A data scientist is developing a model to predict the outcome of a vote for a national mascot. The choice is between tigers and lions. The full data set represents feedback from individuals representing 17 professions and 12 different locations. The following rank aggregation represents 80% of the data set:
(Screenshot shows survey rankings for just two professions and a few locations, all voting for "Tigers") Which of the following is the most likely concern about the model's ability to predict the outcome of the vote?
- A. In-sample data
- B. Interpolated data
- C. Out-of-sample data
- D. Extrapolated data
正解：D
解説：
# Extrapolated data refers to making predictions about data points that fall outside the observed range or distribution. Since the sample data (80%) is heavily skewed toward a small subset of professions and locations, predicting results for the remaining, unrepresented professions and regions involves extrapolation.
Why the other options are incorrect:
* A: Interpolation occurs within the bounds of observed data - not the issue here.
* C: In-sample data refers to training data, which is overrepresented in this case.
* D: Out-of-sample data is a concern in generalization but extrapolation is more specific here.
Official References:
* CompTIA DataX (DY0-001) Study Guide - Section 3.2:"Extrapolation introduces risk when models are used outside the range of data they were trained on, especially if certain subgroups are underrepresented."
-
質問 # 27
A data scientist is working with a data set that covers a two-year period for a large number of machines. The data set contains:
* Machine system ID numbers
* Sensor measurement values
* Daily timestamps for each machine
The data scientist needs to plot the total measurements from all the machines over the entire time period.
Which of the following is the best way to present this data?
- A. Scatter plot
- B. Box-and-whisker plot
- C. Histogram
- D. Line plot
正解：D
解説：
# Line plots are ideal for visualizing data trends over continuous time. In this case, plotting the total daily measurements across a two-year period is a time series task, and a line plot shows progression and pattern over time clearly.
Why the other options are incorrect:
* A: Scatter plots are better for relationship exploration, not time trends.
* C: Histograms display distribution - not suitable for continuous time trends.
* D: Box plots show spread and outliers - not temporal behavior.
Official References:
* CompTIA DataX (DY0-001) Study Guide - Section 1.2:"Use line plots for visualizing temporal trends in time-series data."
* Time Series Visualization Guide, Chapter 2:"Line plots are effective for showing cumulative or aggregated values over time."
-
質問 # 28
Given the equation:
Xt = # + #1Xt#1 + #t, where #t # N(0, ##²)
Which of the following time series models best represents this process?
- A. SARIMA(1,1,1) × (1,1,1)1
- B. ARMA(1,1)
- C. ARIMA(1,1,1)
- D. AR(1)
正解：D
解説：
# The provided equation represents an autoregressive model of order 1 (AR(1)). It describes Xt as a function of its immediately prior value (Xt#1) plus white noise.
Key identifiers:
* No differencing (so not ARIMA).
* No moving average term (so not ARMA).
* No seasonal component (so not SARIMA).
Why the other options are incorrect:
* A: ARIMA(1,1,1) includes integration and MA terms, which are absent here.
* B: ARMA(1,1) includes both AR and MA terms, but only AR is present.
* C: SARIMA involves seasonal and differencing components - not applicable here.
Official References:
* CompTIA DataX (DY0-001) Study Guide - Section 3.5:"AR(p) models describe a variable as dependent on its previous values with no differencing or moving average."
* Time Series Analysis Textbook, Chapter 4:"Xt = #Xt-1 + #t describes an AR(1) process when #t is white noise."
-
質問 # 29
A data scientist is building an inferential model with a single predictor variable. A scatter plot of the independent variable against the real-number dependent variable shows a strong relationship between them.
The predictor variable is normally distributed with very few outliers. Which of the following algorithms is the best fit for this model, given the data scientist wants the model to be easily interpreted?
- A. A probit regression
- B. A linear regression
- C. An exponential regression
- D. A logistic regression
正解：B
解説：
The scenario provided describes a modeling problem with the following characteristics:
* A single continuous predictor variable (independent variable).
* A continuous real-number dependent variable.
* The relationship between the variables appears strong and linear, as observed from the scatter plot.
* The predictor variable is normally distributed with minimal outliers.
* The goal is to maintain interpretability in the model.
Based on the above, the most appropriate modeling technique is:
Linear Regression: This is a statistical method used to model the linear relationship between a continuous dependent variable and one or more independent variables. In simple linear regression, a straight line (y = mx
+ b) represents the relationship, where the slope and intercept can be easily interpreted. This method is preferred when the relationship is linear, the assumptions of normality and homoscedasticity are satisfied, and interpretability is required.
Why the other options are incorrect:
* A. Logistic Regression: This is used when the dependent variable is categorical (e.g., binary classification), not continuous. Therefore, not suitable for this case.
* B. Exponential Regression: Applied when the data shows an exponential growth or decay pattern, which is not implied here.
* D. Probit Regression: Similar to logistic regression but based on a normal cumulative distribution.
Used for categorical outcomes, not continuous variables.
Exact Extract and Official References:
* CompTIA DataX (DY0-001) Official Study Guide, Domain: Modeling, Analysis, and Outcomes:
"Linear regression is the most interpretable form of regression modeling. It assumes a linear relationship between independent and dependent variables and is ideal for inferential modeling when interpretability is important." (Section 3.1, Model Selection Criteria)
* Data Science Fundamentals, by CompTIA and DS Institute:
"Linear regression is a robust and interpretable statistical method used for modeling continuous outcomes. It provides coefficients which help in understanding the strength and direction of the relationship." (Chapter 4, Regression Techniques)
質問 # 30
Which of the following environmental changes is most likely to resolve a memory constraint error when running a complex model using distributed computing?
- A. Migrating to a cloud deployment
- B. Moving model processing to an edge deployment
- C. Converting an on-premises deployment to a containerized deployment
- D. Adding nodes to a cluster deployment
正解：D
解説：
When running a model on a distributed system, encountering memory constraint errors indicates that the current nodes in the cluster do not have enough memory to handle the model. The most scalable and immediate solution is:
# Adding Nodes to a Cluster Deployment - This increases the total available memory and compute power. In distributed computing environments like Apache Spark or Hadoop, horizontal scaling via node addition is a standard remedy for resource bottlenecks, including memory limitations.
Why the other options are incorrect:
* A. Containerizing doesn't inherently solve memory issues unless paired with resource upgrades.
* B. Cloud migration may offer more resources, but without scaling configuration, memory limits may persist.
* C. Edge deployment is for low-latency, local processing - often with less memory, not more.
Official References:
* CompTIA DataX (DY0-001) Official Study Guide - Section 5.2 (Infrastructure & Scaling):"To resolve memory limitations in distributed systems, scaling out by adding nodes is the most direct and cost- effective method."
* Data Engineering Fundamentals (Cloud/Distributed Systems):"Cluster resource constraints (e.g., memory) can be mitigated by increasing node count, enabling parallel execution and expanded memory pools."
-
質問 # 31
A data scientist is working with a data set that has ten predictors and wants to use only the predictors that most influence the results. Which of the following models would be the best for the data scientist to use?
- A. Weighted least squares
- B. Ridge
- C. LASSO
- D. OLS
正解：C
解説：
# LASSO (Least Absolute Shrinkage and Selection Operator) regression performs both variable selection and regularization by adding an L1 penalty to the loss function. It shrinks less important feature coefficients to zero, effectively performing feature selection - perfect for identifying the most influential predictors.
Why the other options are incorrect:
* A: OLS uses all predictors and doesn't perform feature selection.
* B: Ridge regression applies an L2 penalty, shrinking coefficients but keeping all predictors.
* C: Weighted least squares adjusts for heteroscedasticity but doesn't reduce variable count.
Official References:
* CompTIA DataX (DY0-001) Study Guide - Section 3.3:"LASSO performs feature selection by zeroing out coefficients of less significant predictors."
* Statistical Learning Textbook, Chapter 6:"LASSO regression is ideal when model interpretability and variable reduction are important."
-
質問 # 32
A company created a very popular collectible card set. Collectors attempt to collect the entire set, but the availability of each card varies, because some cards have higher production volumes than others. The set contains a total of 12 cards. The attributes of the cards are shown.
The data scientist is tasked with designing an initial model iteration to predict whether the animal on the card lives in the sea or on land, given the card's features: Wrapper color, Wrapper shape, and Animal.
Which of the following is the best way to accomplish this task?
- A. Decision trees
- B. Linear regression
- C. Association rules
- D. ARIMA
正解：A
解説：
# Decision trees are supervised classification models that can be used to predict a categorical target variable (e.
g., Habitat: Land or Sea) based on input features (e.g., Wrapper color, Wrapper shape, Animal type). They are interpretable, require minimal preprocessing, and are ideal for structured categorical data like this.
Why the other options are incorrect:
* A: ARIMA (AutoRegressive Integrated Moving Average) is used for time-series forecasting, not classification.
* B: Linear regression is used for predicting continuous numeric values, not categorical variables like
"Land" or "Sea".
* C: Association rules (like in market basket analysis) are used to discover relationships or co-occurrence among variables, not to build predictive models.
Official References:
* CompTIA DataX (DY0-001) Study Guide - Section 4.1 & 4.2:"Decision trees are powerful classifiers for categorical output variables and allow for interpretable models based on feature splits."
* Machine Learning Textbook, Chapter 6:"Decision trees are ideal for early-stage model prototyping when the output is categorical and the data structure is tabular."
質問 # 33
In a modeling project, people evaluate phrases and provide reactions as the target variable for the model.
Which of the following best describes what this model is doing?
- A. Part-of-speech tagging
- B. Named-entity recognition
- C. Sentiment analysis
- D. TF-IDF vectorization
正解：C
解説：
# Sentiment analysis refers to using machine learning or NLP techniques to determine the sentiment or emotional tone behind a body of text (e.g., positive, neutral, or negative). When people provide reactions to phrases, the model is learning to associate language with subjective emotion or opinion.
Why the other options are incorrect:
* B: NER identifies entities (e.g., locations, organizations) - not emotions.
* C: TF-IDF is a feature engineering method, not a modeling goal.
* D: POS tagging classifies words by their grammatical function - not sentiment.
Official References:
* CompTIA DataX (DY0-001) Official Study Guide - Section 6.3:"Sentiment analysis models associate textual input with subjective labels, such as emotional response or polarity."
* Applied Text Analytics, Chapter 8:"When modeling user reactions to text, sentiment classification techniques are commonly employed."
-
質問 # 34
Which of the following does k represent in the k-means model?
- A. Number of clusters
- B. Distance between features
- C. Number of data splits
- D. Number of model tests
正解：A
解説：
# In k-means clustering, k represents the number of clusters that the algorithm will attempt to form. The algorithm partitions the dataset into k distinct, non-overlapping clusters based on feature similarity. Each cluster has a centroid, and the algorithm aims to minimize the intra-cluster variance.
Why the other options are incorrect:
* A: Number of tests is unrelated to the k-means algorithm.
* B: Data splits refer to cross-validation or train/test splits, not k in k-means.
* D: Distance between features is computed during clustering but is not what "k" represents.
Official References:
* CompTIA DataX (DY0-001) Official Study Guide - Section 4.2:"In k-means clustering, k denotes the number of clusters into which the dataset will be partitioned."
* Introduction to Machine Learning, Chapter 6:"The 'k' in k-means specifies how many groupings the algorithm will seek to discover based on proximity in feature space."
-
質問 # 35
Which of the following distributions would be best to use for hypothesis testing on a data set with 20 observations?
- A. Uniform
- B. Power law
- C. Normal
- D. Student's t-
正解：D
解説：
# For small sample sizes (typically n < 30), the Student's t-distribution is preferred over the normal distribution for hypothesis testing because it accounts for the added uncertainty in the estimate of the standard deviation. With 20 observations, the t-distribution is more appropriate and reliable.
Why the other options are incorrect:
* A: Power law is used in modeling rare events or heavy-tailed distributions, not hypothesis testing.
* B: The normal distribution is more appropriate when the sample size is large.
* C: Uniform distribution assumes equal probability - not used in inferential statistics.
Official References:
* CompTIA DataX (DY0-001) Study Guide - Section 1.3:"The t-distribution is used for small sample hypothesis testing where the population standard deviation is unknown."
-
質問 # 36
......
正真正銘のベスト問題集資料を使おうDY0-001オンライン練習試験：https://jp.fast2test.com/DY0-001-premium-file.html
最大一年間毎日更新されるDY0-001ブレーン問題集：https://drive.google.com/open?id=1iYVPfeu_3EFgtv8i-G_d9vRG0-rLgD-n