ISTQB ISTQB-CT-AI Online Practice Questions and Exam Preparation

ISTQB ISTQB-CT-AI Online Questions & Answers

• Question 101:

  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 against the actual routes and times.

  Which of the following describes the next phase of metamorphic testing?

  A. 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 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.
  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 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.
  A. 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 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.

  ---

  The ISTQB CT-AI syllabus describes metamorphic testing as a technique where metamorphic relations (MRs) are defined and then applied to verify that predictable input transformations lead to predictable output changes, even when exact expected outputs are not available.

  In this scenario, vehicle speed is the defining metamorphic relation. By applying the same route to vehicles with different speeds and predicting the expected change in travel time based on speed differences, the testers can verify whether the AI system behaves consistently with the defined relation.

  Therefore, applying speed-based transformations to the same route and validating the resulting arrival times correctly represents the next phase of metamorphic testing, making Option A the correct answer.

• Question 102:

  A company producing consumable goods wants to identify groups of people with similar tastes for the purpose of targeting different products for each group. You have to choose and apply an appropriate ML type for this problem. Which ONE of the following options represents the BEST possible solution for this above-mentioned task?

  A. Regression
  B. Association
  C. Clustering
  D. Classification
  C. Clustering

  ---

  Regression is used to predict continuous numerical values and is not suitable for grouping people based on similar tastes.

  Association techniques identify relationships or co-occurrence patterns between variables (for example, market basket analysis), but they do not directly group individuals into segments.

  Clustering is an unsupervised learning technique that groups data points based on similarity in their features without requiring predefined labels. This makes it ideal for identifying groups of people with similar tastes for targeted marketing purposes.

  Classification is a supervised learning approach that assigns data points to predefined classes and requires labeled training data. In this scenario, the goal is to discover groups rather than assign known labels.

  Therefore, Clustering (Option C) is the most appropriate and syllabus-aligned solution.

• Question 103:

  Which of the following statements about the structure and function of neural networks is true?

  A. The bias of a neuron is determined by the activation values of the neurons in the previous layer.
  B. Training a neural network only changes the values of the weights at the connections between neurons.
  C. A single-layer perceptron is NOT a neural network.
  D. The input layer of a deep neural network must have at least as many neurons as its output layer.
  B. Training a neural network only changes the values of the weights at the connections between neurons.

  ---

  Section 1.7 - Neural Networks of the ISTQB CT-AI syllabus explains that neural networks consist of neurons connected by weighted links. During training, learning occurs by adjusting the weights on these connections. This is the essence of gradient descent and backpropagation. Option B correctly states this behavior: only the weights are modified, not the activation functions, the number of neurons, or the architectural structure.

  Option A is incorrect because a neuron's bias is not determined by the activations of neurons in the previous layer; it is an independent trainable parameter added to the weighted sum of inputs. Option C is incorrect because the syllabus states that a single-layer perceptron is a valid type of neural network, although it is limited to linearly separable problems. Option D is incorrect because there is no requirement that the number of input neurons be greater than or equal to the number of output neurons. Input

  neurons correspond to the number of features, while output neurons correspond to tasks or classes.

  Therefore, Option B precisely reflects the syllabus definition of what changes during neural network training.

• Question 104:

  Consider an AI-based system in which the complex internal structure has been generated by another software system. Why would a tester choose to perform black-box testing on this particular system?

  A. Test automation can be built quickly and easily from the test cases developed during black-box testing.
  B. The tester wishes to better understand the logic of the software used to create the internal structure.
  C. Black-box testing allows the tester to check the transparency of the algorithm used to create the internal structure.
  D. Black-box testing eliminates the need for the tester to understand the internal structure of the AI-based system.
  D. Black-box testing eliminates the need for the tester to understand the internal structure of the AI-based system.

  ---

  The ISTQB CT-AI syllabus explains that many AI-based systems have internal structures that are highly complex, automatically generated, or not interpretable by humans. In such cases, understanding the internal structure does not provide additional useful information for testing.

  The syllabus states that when the internal workings of an AI-based system are too complex or opaque, testing must be performed using a black-box approach, focusing on inputs and outputs rather than internal logic. Therefore, black-box testing is chosen specifically because the tester does not need to understand the internal structure of the AI-based system.

  Options A, B, and C are incorrect because black-box testing is not intended to simplify automation creation, reveal internal logic, or assess transparency of the algorithm itself.

  Thus, Option D correctly reflects the rationale for using black-box testing in this context.

• Question 105:

  Which ONE of the following statements is a CORRECT description of adversarial examples in the context of machine learning systems that work on image classifiers?

  A. Black-box attacks based on adversarial examples create an exact duplicate model of the original.
  B. These attack examples cause a model to predict the correct class with slightly lower accuracy, even though they look like the original image.
  C. These attacks cannot be prevented by retraining the model with these examples augmented to the training data.
  D. These examples are model-specific and are not likely to cause another model trained on the same task to fail.
  D. These examples are model-specific and are not likely to cause another model trained on the same task to fail.

  ---

  Adversarial examples are inputs that are intentionally modified in a subtle way to cause a machine learning model to make an incorrect prediction, even though the input appears unchanged to a human observer.

  Option A is incorrect because black-box attacks do not create an exact duplicate of the original model. Instead, attackers may approximate behavior by querying the model, without knowing its internal structure.

  Option B is incorrect because adversarial examples are designed to cause misclassification, not merely a slight reduction in accuracy while still predicting the correct class.

  Option C is incorrect because adversarial attacks can be mitigated through techniques such as adversarial training, where adversarial examples are included in the training data to improve model robustness.

  Option D is correct. Adversarial examples are often model-specific, exploiting weaknesses in a particular model's learned decision boundaries. While transferability can occur, many adversarial examples do not reliably cause failures in other models trained on the same task.

  Therefore, Option D is the correct statement.

• Question 106:

  Which ONE of the following options describes the LEAST LIKELY usage of AI for detection of GUI changes due to changes in test objects?

  A. Using a pixel comparison of the GUI before and after the change to check the differences.
  B. Using computer vision to compare the GUI before and after the test object changes.
  C. Using vision-based detection of GUI layout changes before and after test object changes.
  D. Using an ML-based classifier to flag whether GUI changes should be reviewed by humans.
  A. Using a pixel comparison of the GUI before and after the change to check the differences.

  ---

  Using a pixel-by-pixel comparison of the GUI before and after a change is a traditional image comparison technique and does not involve AI. It performs a direct comparison of pixel values without applying learning, inference, or intelligent interpretation. Therefore, it is the least likely usage of AI for detecting GUI changes.

  Using computer vision techniques to compare GUIs involves AI-based image understanding and is commonly applied to detect visual changes beyond simple pixel differences.

  Using vision-based detection of GUI layout changes leverages AI to analyze structural and spatial relationships within the interface, which is a typical AI-based approach.

  Using an ML-based classifier to determine whether detected GUI changes should be flagged for human review is also a valid AI usage, as it applies learned patterns to support decision-making.

  Therefore, Option A correctly describes the least likely usage of AI for detecting GUI changes.

• Question 107:

  Which ONE of the following describes a situation of back-to-back testing the LEAST?

  A. Comparison of the results of a current neural network ML model implemented on platform A (for example, PyTorch) with a similar neural network ML model implemented on platform B (for example, TensorFlow), using the same data.
  B. Comparison of the results of a home-grown neural network ML model with the results of a neural network ML model implemented using a standard framework (for example, PyTorch), using the same data.
  C. Comparison of the results of a neural network ML model with a decision tree ML model using the same data.
  D. Comparison of the results of the current neural network ML model using the current dataset with a slightly modified dataset.
  C. Comparison of the results of a neural network ML model with a decision tree ML model using the same data.

  ---

  Back-to-back testing is a technique in which the same test cases and input data are executed against multiple implementations that are intended to provide equivalent behavior, and their outputs are compared to identify inconsistencies.

  Option A describes comparing two implementations of the same type of model (neural networks) built on different platforms using identical data, which is a typical back-to-back testing scenario. Option B describes comparing a custom implementation with a standard reference implementation using the same data, which is also a common use of back-to-back testing to validate correctness.

  Option C describes comparing a neural network model with a decision tree model using the same data.

  Because these models are fundamentally different in structure and behavior, they are not expected to produce equivalent outputs, making this the least appropriate scenario for back-to-back testing.

  Option D compares the same model using different datasets. This represents robustness or sensitivity testing rather than back-to-back testing, but it still does not involve comparing different model types.

  Therefore, Option C is the correct answer, as it least describes a back-to-back testing situation.

• Question 108:

  Which of the following problems would best be solved using the supervised learning category of regression?

  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
  B. Recognizing a knife in carry-on luggage at a security checkpoint using airport scanner images
  C. Determining whether an animal is a pig or a cow based on image recognition
  D. Predicting shopper purchasing behavior based on the category of shopper and the positioning of promotional displays within a store
  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

  ---

  The ISTQB CT-AI syllabus defines supervised learning as being divided into classification and

  regression problems.

  Regression is used when the model must predict a continuous numeric value, rather than a discrete category.

  In this scenario:

  Option A involves predicting an optimal numeric value (age) based on historical numerical data (age and egg production), which is a classic regression problem.

  The remaining options represent classification tasks:

  Recognizing a knife in luggage

  Distinguishing between animal types

  Predicting purchasing behavior categories

  These involve assigning inputs to discrete classes rather than predicting continuous values.

  This aligns with the syllabus definition:

  "Regression is used when the problem requires the ML model to predict a numeric output, for example predicting the age of a person based on their habits."

  (ISTQB CT-AI Syllabus v1.0, Section 3.1.1, Page 26 of 99)

• Question 109:

  Which ONE of the following is the MOST appropriate metric to prioritize for a cancer detection system with very low prevalence?

  A. Accuracy
  B. Recall
  C. Specificity
  D. R-squared
  B. Recall

  ---

• Question 110:

  Which statement regarding data preparation in the ML workflow is correct?

  A. A key challenge in data transformation is the removal or correction of erroneous data.
  B. Since data preparation is time-consuming, all steps should be automated.
  C. One challenge of data gathering is obtaining high-quality data from multiple sources.
  D. Sampling is so well researched that it is no longer considered risky.
  C. One challenge of data gathering is obtaining high-quality data from multiple sources.

  ---

  The ISTQB CT-AI syllabus describes the ML data preparation workflow in Section 2.2 - Data Preparation. Data preparation consists of data gathering, cleaning, transformation, and sampling. The syllabus emphasizes that one significant challenge during data gathering is combining data from multiple heterogeneous sources, which often differ in structure, quality, and format. Ensuring that the resulting data set is accurate, complete, and representative can be complex, making this a critical challenge in the machine learning workflow. This aligns directly with Option C.

  Option A is incorrect because the removal or correction of erroneous data is part of data cleaning, not data transformation. Option B contradicts the syllabus because, although automation can support data preparation, not all steps should be fully automated due to the need for expert judgment and oversight, particularly in identifying subtle data quality issues. Option D is incorrect because sampling remains risky, especially with respect to representativeness and bias, and the syllabus explicitly highlights the need for caution when sampling data.

  Thus, Option C is the only statement that accurately reflects the syllabus.