Real CT-AI Exam Questions are the Best Preparation Material [Q54-Q74]

Real CT-AI Exam Questions are the Best Preparation Material

Practice on 2026 LATEST CT-AI Exam Updated 162 Questions

NEW QUESTION # 54
Which ONE of the below statements BEST describes why test environments for autonomous systems might need to be different to other test environments?

• A. Tools may be required to simulate extreme scenarios
• B. AI-specific hardware may be required
• C. Non determinism may need to be introduced into the environment
• D. Tools may be required to provide explanations of system behaviour

Answer: A

Explanation:
The correct answer is A . For autonomous systems, the distinguishing test-environment issue is that the system must respond to changes in its environment without human intervention and must also recognize situations where autonomy should be ceded back to human operators. The CT-AI syllabus states that, for some autonomous systems, identifying and mimicking the circumstances for ceding autonomy may require test environments to push the systems to extremes . It also notes that some autonomous systems operate in hazardous environments, making representative real-world testing difficult or unsafe.
Option B relates more specifically to multi-agent AI systems, where the environment may need non- determinism to mimic other interacting AI-based systems. Option C is a general AI-environment consideration where AI-specific processors may need inclusion. Option D is associated with explainability, where tools may be needed to understand decisions. These are valid AI test-environment factors, but they are not the best autonomy-specific reason. Virtual environments are particularly valuable because dangerous, unusual, and extreme scenarios can be tested safely and repeatedly.
References/topics: CT-AI Syllabus Chapter 10, Sections 10.1 and 10.2 "Test Environments for AI-Based Systems."
=========

NEW QUESTION # 55
A ML engineer is trying to determine the correctness of the new open-source implementation *X", of a supervised regression algorithm implementation. R-Square is one of the functional performance metrics used to determine the quality of the model.
Which ONE of the following would be an APPROPRIATE strategy to achieve this goal?

• A. Compare the R-Square score of the model obtained using two different implementations that utilize two different programming languages while using the same algorithm and the same training and testing data.
• B. Drop 10% of the rows randomly and create another model and compare the R-Square scores of both the models.
• C. Train various models by changing the order of input features and verify that the R-Square score of these models vary significantly.
• D. Add 10% of the rows randomly and create another model and compare the R-Square scores of both the model.

Answer: A

Explanation:
This approach directly compares the performance of two implementations of the same algorithm.
If both implementations produce similar R-Square scores on the same training and testing data, it suggests that the new implementation "X" is correct.

NEW QUESTION # 56
A beer company is trying to understand how much recognition its logo has in the market. It plans to do that by monitoring images on various social media platforms using a pre-trained neural network for logo detection.
This particular model has been trained by looking for words, as well as matching colors on social media images. The company logo has a big word across the middle with a bold blue and magenta border.
Which associated risk is most likely to occur when using this pre-trained model?

• A. Improper data preparation
• B. Insufficient function; the model was not trained to check for colors or words
• C. Inherited bias: the model could have inherited unknown defects
• D. There is no risk, as the model has already been trained

Answer: C

Explanation:
A major risk when using apre-trained neural networkfor logo detection is that it mayinherit biases and defectsfrom the original dataset and training process. This means that the model could misidentify or fail to recognize certain logos due to:
* Differences in data preparation:The original training data may have used a different preprocessing method than the new dataset, leading to inconsistencies.
* Limited transparency:The exact details of the dataset and biases within it may not be known, which can cause unexpected behavior.
* Bias in logo detection:If the model was trained on a dataset with certain color or text preferences, it may disproportionately misidentify logos with similar characteristics.
This inherited bias can result in:
* False Positives:Recognizing other brand logos as the beer company's logo.
* False Negatives:Failing to detect the actual logo when variations occur (e.g., different lighting or partial visibility).
* Algorithmic Bias:The model may favor certain shapes or color contrasts due to biased training data.
Thus,the most appropriate risk associated with using this pre-trained model is inherited bias.
* Section 1.8.3 - Risks of Using Pre-Trained Models and Transfer Learningexplains how pre-trained models may inheritbiases and undocumented defectsthat affect performance in a new environment.
Reference from ISTQB Certified Tester AI Testing Study Guide:

NEW QUESTION # 57
Which ONE of the following options does NOT describe a challenge for acquiring test data in ML systems?
SELECT ONE OPTION

• A. Test data being sourced from public sources.
• B. Compliance needs require proper care to be taken of input personal data.
• C. Nature of data constantly changes with lime.
• D. Data for the use case is being generated at a fast pace.

Answer: D

Explanation:
* Challenges for Acquiring Test Data in ML Systems: Compliance needs, the changing nature of data over time, and sourcing data from public sources are significant challenges. Data being generated quickly is generally not a challenge; it can actually be beneficial as it provides more data for training and testing.
* Reference: ISTQB_CT-AI_Syllabus_v1.0, Sections on Data Preparation and Data Quality Issues.

NEW QUESTION # 58
Data used for an object detection ML system was found to have been labelled incorrectly in many cases.
Which ONE of the following options is most likely the reason for this problem?
SELECT ONE OPTION

• A. Security issues
• B. Bias issues
• C. Accuracy issues
• D. Privacy issues

Answer: C

Explanation:
The question refers to a problem where data used for an object detection ML system was labelled incorrectly. This issue is most closely related to "accuracy issues." Here's a detailed explanation:
Accuracy Issues: The primary goal of labeling data in machine learning is to ensure that the model can accurately learn and make predictions based on the given labels. Incorrectly labeled data directly impacts the model's accuracy, leading to poor performance because the model learns incorrect patterns.
Why Not Other Options:
Security Issues: This pertains to data breaches or unauthorized access, which is not relevant to the problem of incorrect data labeling.
Privacy Issues: This concerns the protection of personal data and is not related to the accuracy of data labeling.
Bias Issues: While bias in data can affect model performance, it specifically refers to systematic errors or prejudices in the data rather than outright incorrect labeling.

NEW QUESTION # 59
A transportation company operates three types of delivery vehicles in its fleet. The vehicles operate at different speeds (slow, medium, and fast). The transportation company is attempting to optimize scheduling and has created an AI-based program to plan routes for its vehicles using records from the medium-speed vehicle traveling to selected destinations. The test team uses this data in metamorphic testing to test the accuracy of the estimated travel times created by the AI route planner with the actual routes and times.
Which of the following describes the next phase of metamorphic testing?

• A. The team uses the same AI route planner to create routes that are longer and shorter but follow the same track. Finally, by driving the fast vehicles on the long routes and slow vehicles on the short routes and vice versa, the AI system will have enough information to infer travel times for all vehicles on all routes.
• B. The team decomposes each route into the relevant components that affect the travel time such as traffic density and vehicle power. The team then uses statistical analysis to characterize the influence of each component to calculate the fast and slow vehicle route times.
• C. The team uses an AI system to select the most dissimilar routes. With this information, any of the AI routes can be metaphorically transformed into a fast or slow route.
• D. The team tests the time required for the fast and slow vehicles to travel the same route as the medium vehicle. Then, by calculating the speed difference, they then predict how much faster or slower the vehicles will travel. That information is then used to verify that the arrival time of the vehicles meets the expected result.

Answer: D

Explanation:
Metamorphic Testing (MT) is a testing technique that verifies AI-based systems by generating follow-up test cases based on existing test cases. These follow-up test cases adhere to a Metamorphic Relation (MR), ensuring that if the system is functioning correctly, changes in input should result in predictable changes in output.
Metamorphic testing works by transforming source test cases into follow-up test cases Here, the source test case involves testing the medium-speed vehicle's travel time. The follow-up test cases are derived by extrapolating travel times for fast and slow vehicles using predictable relationships based on speed differences.
MR states that modifying input should result in a predictable change in output Since the speed of the vehicle is a known factor, it is possible to predict the new arrival times and verify whether they follow expected trends.
This is a direct application of metamorphic testing principles In route optimization systems, metamorphic testing often applies transformations to speed, distance, or conditions to verify expected outcomes.

NEW QUESTION # 60
Which statement regarding testing transparency, explainability, or interpretability is MOST correct?

• A. LIME can precisely state the decisive reason for a change in the output
• B. Dynamic testing is one way to quantify explainability; however, each method is specific to a particular model type
• C. Since different users have different backgrounds, interpretability testing depends on the comprehensibility of the ML algorithm
• D. Tests for explainability and transparency are comparable to exploratory testing and can be performed with little information about development

Answer: C

Explanation:
The ISTQB CT-AI syllabus states in Section2.10 - Explainability, Transparency, and Interpretability that interpretability isuser-dependent, meaning different users understand explanations differently. This is because interpretability depends not only on the ML algorithm but also on the user's domain knowledge, experience, and expectations. Option B directly reflects this syllabus principle: interpretability testing must consider user background, and explanations must be comprehensible to the intended user group.

NEW QUESTION # 61
Which of the following problems would best be solved using the supervised learning category of regression?

• A. Recognizing a knife in carry on luggage at a security checkpoint in an airport scanner.
• B. Predicting shopper purchasing behavior based on the category of shopper and the positioning of promotional displays within a store.
• C. Determining if an animal is a pig or a cow based on image recognition.
• D. Determining the optimal age for a chicken's egg laying production using input data of the chicken's age and average daily egg production for one million chickens.

Answer: D

Explanation:
Understanding Supervised Learning - RegressionSupervised learning is a category of machine learning where the model is trained on labeled data. Within this category,regressionis used when the goal is to predict a continuous numeric value.
* Regressiondeals with problems where the output variable is continuous in nature, meaning it can take any numerical value within a range.
* Common examples include predicting prices, estimating demand, and analyzing production trends.
* (A) Determining the optimal age for a chicken's egg-laying production using input data of the chicken's age and average daily egg production for one million chickens.#(Correct)
* This is a classicregression problembecause it involves predicting a continuous variable:daily egg productionbased on the input variablechicken's age.
* The goal is to find a numerical relationship between age and egg production, which makesregression the appropriate supervised learning method.
* (B) Recognizing a knife in carry-on luggage at a security checkpoint in an airport scanner.#(Incorrect)
* This is animage recognition task, which falls underclassification, not regression.
* Classification problems involve assigning inputs to discrete categories (e.g., "knife detected" or
"no knife detected").
* (C) Determining if an animal is a pig or a cow based on image recognition.#(Incorrect)
* This is anotherclassification problemwhere the goal is to categorize an image into one of two labels (pig or cow).
* (D) Predicting shopper purchasing behavior based on the category of shopper and the positioning of promotional displays within a store.#(Incorrect)
* This problem could involve a mix ofclassificationandassociation rule learning, but it does not explicitly predict a continuous variable in the way regression does.
* Regression is used when predicting a numeric output."Predicting the age of a person based on input data about their habits or predicting the future prices of stocks are examples of problems that use regression."
* Supervised learning problems are divided into classification and regression."If the output is numeric and continuous in nature, it may be regression."
* Regression is commonly used for predicting numerical trends over time."Regression models result in a numerical or continuous output value for a given input." Analysis of Answer ChoicesReferences from ISTQB Certified Tester AI Testing Study GuideThus,option A is the correct answer, as it aligns with the principles of regression-based supervised learning.

NEW QUESTION # 62
Which of the following statements about ML functional performance metrics is correct?

• A. The R-squared metric indicates how well the model distinguishes between different classes based on the ROC curve.
• B. The receiver operating characteristic curve shows, depending on parameters, how well the model distinguishes between different clusters.
• C. The silhouette coefficient describes how well the regression model fits the dependent variables.
• D. Metrics used to measure clustering include intra-cluster metrics that measure the proximity of a cluster's data points.

Answer: D

Explanation:
The ISTQB CT-AI syllabus explains ML performance metrics in Section3.2 - Evaluating ML Models. Forclustering, which is an unsupervised learning method, the syllabus lists metrics such asintra- cluster distance,inter-cluster distance, and coherence measures. Intra-cluster metrics evaluate how close data points are within a cluster, which directly corresponds to Option A.

NEW QUESTION # 63
Upon testing a model used to detect rotten tomatoes, the following data was observed by the test engineer, based on certain number of tomato images.

For this confusion matrix which combinations of values of accuracy, recall, and specificity respectively is CORRECT?

• A. 1,0.87,0.84
• B. 1,0.9, 0.8
• C. 0.84.1,0.9
• D. 0.87.0.9. 0.84

Answer: D

Explanation:
To calculate the accuracy, recall, and specificity from the confusion matrix provided, we use the following formulas:
Confusion Matrix:
Actually Rotten: 45 (True Positive), 8 (False Positive)
Actually Fresh: 5 (False Negative), 42 (True Negative)
Accuracy:
Accuracy is the proportion of true results (both true positives and true negatives) in the total population.
Formula: Accuracy=TP+TNTP+TN+FP+FN\text{Accuracy} = \frac{TP + TN}{TP + TN + FP + FN}Accuracy=TP+TN+FP+FNTP+TN Calculation: Accuracy=45+4245+42+8+5=87100=0.87\text{Accuracy} = \frac{45 + 42}{45 + 42 + 8
+ 5} = \frac{87}{100} = 0.87Accuracy=45+42+8+545+42=10087=0.87
Recall (Sensitivity):
Recall is the proportion of true positive results in the total actual positives.
Formula: Recall=TPTP+FN\text{Recall} = \frac{TP}{TP + FN}Recall=TP+FNTP Calculation: Recall=4545+5=4550=0.9\text{Recall} = \frac{45}{45 + 5} = \frac{45}{50} =
0.9Recall=45+545=5045=0.9
Specificity:
Specificity is the proportion of true negative results in the total actual negatives.
Formula: Specificity=TNTN+FP\text{Specificity} = \frac{TN}{TN + FP}Specificity=TN+FPTN Calculation: Specificity=4242+8=4250=0.84\text{Specificity} = \frac{42}{42 + 8} = \frac{42}{50} =
0.84Specificity=42+842=5042=0.84
Therefore, the correct combinations of accuracy, recall, and specificity are 0.87, 0.9, and 0.84 respectively.

NEW QUESTION # 64
Which ONE of the below is MOST likely to indicate a problem with underfitting in an ML model?

• A. The model is inaccurate on data similar to the training data
• B. The model is vulnerable to adversarial attacks
• C. The model fails to generalize on new data
• D. The model uses a large amount of resources to make a prediction

Answer: A

Explanation:
The correct answer is D . Underfitting occurs when a model is not sophisticated enough to capture the patterns in the training data. The CT-AI syllabus states that underfitting models tend to be too simplistic and may struggle to provide accurate predictions both for new data and for data very similar to the training data.
Therefore, poor performance on data similar to the training data is the strongest indicator of underfitting.
Option B is more strongly associated with overfitting when considered alone. Overfitting occurs when the model fits the training data too closely and fails to generalize properly to new data. A model that performs well on training-like data but poorly on independent new data is therefore more likely overfitted than underfitted. Option A concerns adversarial robustness, not model fit. Option C concerns performance efficiency or resource consumption, not whether the model has learned the underlying data patterns. For exam purposes, the decisive diagnostic is whether the model is also inaccurate on training-like data; if so, underfitting is the best explanation.
References/topics: CT-AI Syllabus Chapter 3, Sections 3.5.1 "Overfitting" and 3.5.2 "Underfitting."
=========

NEW QUESTION # 65
A team of software testers is attempting to create an AI algorithm to assist in software testing. This particular team has gone through over 40 iterations of testing and cannot afford to spend as much time as it takes to run the full regression test suite. They are hoping to have the algorithm reduce the amount of testing required, thus reducing the time needed for each testing cycle.
How can an AI-based tool be expected to assist in this reduction?

• A. By using A/B testing to compare the last update with the newest change and compare metrics between the two
• B. By performing Bayesian analysis to estimate the types of human interactions that are expected to be seen in the system and then selecting those test cases
• C. By performing optimization of the data from past iterations to see where the most common defects occurred and select the corresponding test cases
• D. By using a clustering method to quantify the relationships between test cases and then assigning each test case to a category

Answer: C

Explanation:
The syllabus mentions that AI can help optimize regression test suites:
"An AI-based tool can perform optimization of the regression test suite by analyzing... the information from previous test results, associated defects, and the latest changes that have been made, such as features which are broken more frequently and which tests exercise code impacted by recent changes." (Reference: ISTQB CT-AI Syllabus v1.0, Section 11.4, page 79 of 99)

NEW QUESTION # 66
Which statement regarding pairwise testing in an AI-based automotive lane-keeping assist system is correct?
Choose ONE option (1 out of 4)

• A. Pairwise testing can reduce testing efforts otherwise very high due to the large number of parameters.
• B. Pairwise testing is usually insufficient because most defects arise only from interactions of many parameters.
• C. Pairwise testing reduces the test suite so much that it is typically feasible within the available time.
• D. Pairwise testing only uses parameters directly influenced by the driver, otherwise the number of test cases becomes too large.

Answer: A

Explanation:
The ISTQB CT-AI syllabus (Section4.3 - Test Design for AI-Based Systems) highlights pairwise testing as an effectivetest-case reduction techniquefor systems with many input parameters. Lane-keeping assist systems typically include environmental, sensor, and vehicle-dynamic parameters, making exhaustive testing infeasible. Pairwise testing significantly reduces the number of test cases while still capturingall 2-way interactions, which are responsible for a large proportion of software defects.
OptionBaligns with this syllabus description: pairwise testing reduces otherwise extremely large parameter combinations, making test effort manageable.
Option A overstates feasibility guarantees; the syllabus never claims pairwise testing always makes testing
"typically feasible." Option C is unsupported and incorrect because pairwise testing doesnotrestrict parameters to driver-controlled ones. Option D is incorrect because, although some defects arise from higher- order interactions, pairwise testing captures many relevant defects and is widely recognized as a pragmatic compromise.
Thus,Option Bis the correct statement.

NEW QUESTION # 67
A motorcycle engine repair shop owner wants to detect a leaking exhaust valve and fix it before it falls and causes catastrophic damage to the engine. The shop developed and trained a predictive model with historical data files from known health engines and ones which experienced a catastrophic fails due to exhaust valve failure. The shop evaluated 200 engines using this model and then disassembled the engines to assess the true state of the valves, recording the results in the confusion matrix below.
What is the precision of this predictive model

• A. 90.0%
• B. 94.2%
• C. 94.5%
• D. 98.9%

Answer: B

Explanation:
Precision is a performance metric used to evaluate the accuracy of positive predictions in a classification model. It is defined by the formula:
Precision=TPTP+FP×100%\text{Precision} = \frac{TP}{TP + FP} \times 100\%Precision=TP+FPTP×100% Where:
* TP (True Positives)= Number of correctly predicted positive cases
* FP (False Positives)= Number of incorrectly predicted positive cases
The confusion matrix provided in the question would typically list these values. Based on ISTQB's guidelines for calculating precision, selecting the correct number of true positives and false positives from the given data should yield94.2%as the precision.
* Section 5.1 - Confusion Matrix and ML Functional Performance Metricsexplains the calculation of precisionusing the confusion matrix.
Reference from ISTQB Certified Tester AI Testing Study Guide:

NEW QUESTION # 68
Which ONE of the following statements about a system MOST describes an autonomous system?

• A. A loan approval system that can continue to process loan applications indefinitely, within a defined amount of aggregate credit, until an operator reduces the amount of aggregate credit available.
• B. A fraud detection system that alerts operators when it sees a specific fraud risk associated with a transaction, in order to obtain human input to make a final decision.
• C. A chatbot that learns the most effective responses to humans (in order to ensure the humans continue to converse) based on prior experiences.
• D. A self-driving car that automatically stops the car if the driver is not responding to voice prompts, in order to ensure they are awake.

Answer: D

Explanation:
An autonomous system is capable of performing tasks independently, without requiring human intervention. The self-driving car in option B is an example of an autonomous system because it automatically takes action (stopping the car) based on its internal decision-making, in response to the driver's behavior.

NEW QUESTION # 69
Which of the following options is an example of the concept of overfitting?
Choose ONE option (1 out of 4)

• A. A model for the recognition of dogs was trained predominantly with pictures of dogs in parks. On pictures with other animals in parks, dogs are also falsely recognized.
• B. A model for predicting IT system failures delivers too many false-negative predictions because the failures cannot be adequately explained via the log files used for training.
• C. A model for predicting academic performance was trained with data from students at one university.
  The model shows low predictive accuracy when applied to other universities.
• D. A previously trained model for recognizing cars is adapted and extended so that it can also identify the make of the car beyond its original function.

Answer: C

Explanation:
The ISTQB CT-AI syllabus definesoverfittingin Section3.2 - ML Model Evaluationas a condition where an ML model learns the training data too precisely-including noise and irrelevant detail-resulting in poor performance on unseen data. Overfitting is characterized byhigh accuracy on training data but low accuracy on validation or real-world data. OptionAperfectly matches this definition: a model trained only on one university's student data generalizes poorly to students from other universities. This is a textbook example of overfitting because the model has essentially memorized patterns unique to a narrow dataset, instead of learning generalizable relationships applicable across environments .
Option B instead describessample biasor inadequate training diversity, not overfitting. Option C involves transfer learningor model extension, unrelated to overfitting. Option D indicatesinsufficient training data qualityor lack of meaningful features, but not overfitting. Only Option A reflects the syllabus definition directly: overly specialized training leading to reduced predictive performance on new data.
Thus,Ais the correct and syllabus-aligned example of overfitting.

NEW QUESTION # 70
In which ONE of the following situations would an ML model be MOST effective at determining the criticality of new defects?

• A. An old application where defect records are linked to failed tests and production incidents
• B. An old application with few critical defect records and many non-critical defect records
• C. An old application with lots of defect records but a brand new development and test team
• D. A new application which is in the early stages of the first test cycle

Answer: A

Explanation:
An old application where defect records are linked to failed tests and production incidents would provide the most valuable data for an ML model to determine the criticality of new defects. By using historical data of defects that are linked to actual issues in production or testing failures, the model can learn patterns and correlations between defects and their criticality, making it highly effective in predicting the criticality of new defects. This type of historical data provides the necessary context for accurate predictions.

NEW QUESTION # 71
"BioSearch" is creating an Al model used for predicting cancer occurrence via examining X-Ray images. The accuracy of the model in isolation has been found to be good. However, the users of the model started complaining of the poor quality of results, especially inability to detect real cancer cases, when put to practice in the diagnosis lab, leading to stopping of the usage of the model.
A testing expert was called in to find the deficiencies in the test planning which led to the above scenario.
Which ONE of the following options would you expect to MOST likely be the reason to be discovered by the test expert?
SELECT ONE OPTION

• A. A lack of focus on non-functional requirements testing.
• B. The input data has not been tested for quality prior to use for testing.
• C. A lack of similarity between the training and testing data.
• D. A lack of focus on choosing the right functional-performance metrics.

Answer: C

Explanation:
The question asks which deficiency is most likely to be discovered by the test expert given the scenario of poor real-world performance despite good isolated accuracy.
* A lack of similarity between the training and testing data (A): This is a common issue in ML where the model performs well on training data but poorly on real-world data due to a lack of representativeness in the training data. This leads to poor generalization to new, unseen data.
* The input data has not been tested for quality prior to use for testing (B): While data quality is important, this option is less likely to be the primary reason for the described issue compared to the representativeness of training data.
* A lack of focus on choosing the right functional-performance metrics (C): Proper metrics are crucial, but the issue described seems more related to the data mismatch rather than metric selection.
* A lack of focus on non-functional requirements testing (D): Non-functional requirements are important, but the scenario specifically mentions issues with detecting real cancer cases, pointing more towards data issues.
References:
* ISTQB CT-AI Syllabus Section 4.2 on Training, Validation, and Test Datasets emphasizes the importance of using representative datasets to ensure the model generalizes well to real-world data.
* Sample Exam Questions document, Question #40 addresses issues related to data representativeness and model generalization.

NEW QUESTION # 72
Which statement about using AI to analyze reported defects is MOST correct?

• A. ML models trained with critical defect tickets can identify defects that cause serious consequences.
• B. ML models can support duplicate defect identification when checking defect criticality.
• C. ML models identify developers who should handle a defect based on ticket content.
• D. ML models can identify categories for a reported defect during assignment.

Answer: D

Explanation:
The ISTQB CT-AI syllabus (Section5.3 - AI Support for Defect Analysis) explains that AI can categorize defect reports using natural language processing or classification models.
Categorization helps route defects efficiently and determine which areas of the system are affected. Thus, Option C is correct: AI canidentify defect categories, supporting assignment and triage.

NEW QUESTION # 73
Which performance metric is BEST suited to assess the quality of trained models detecting fraudulent credit card transactions?
Choose ONE option (1 out of 4)

• A. F1 value
• B. Accuracy
• C. Sensitivity
• D. -

Answer: A

Explanation:
The ISTQB CT-AI syllabus explains in Section3.2 - Functional Performance Criteria of ML Modelsthat accuracy becomes unreliable whenclass imbalanceexists. In fraud detection, more than 99% of transactions are non-fraudulent, meaning the dataset is extremely imbalanced. Because accuracy counts all correct non- fraudulent classifications, it will appear artificially high, even if the fraud detection performance is poor.
Therefore, accuracy is not suitable for evaluating fraud detection systems.
The syllabus further explains thatsensitivity (recall)captures the proportion of correctly identified fraudulent cases. This metric is important, as missing fraudulent events can cause high financial loss. However, the client also stresses thatlegitimate transactions must be correctly identified, meaningfalse positives must be minimizedto maintain customer satisfaction.
TheF1 score, defined as the harmonic mean ofprecision and recall, balances both:
* Precision protects legitimate customers by minimizing false alarms.
* Recall ensures fraudulent transactions are detected.
Section 3.2 emphasizes that when both false positives and false negatives have significant consequences, and the data is highly imbalanced,F1 is the most appropriate metricbecause it reflects the combined importance of detecting fraud while avoiding unnecessary alerts. Thus,Option Cis the correct choice.

NEW QUESTION # 74
......

Authentic CT-AI Exam Dumps PDF - Sep-2026 Updated: https://www.passcollection.com/CT-AI_real-exams.html

Download Latest CT-AI Dumps with Authentic Real Exam QA's: https://drive.google.com/open?id=1aAvKqFqktZOrQzKamDCeZtcWhUTSeSYi