MODEL TEST PAPER (X AI)

MODEL TEST PAPER – 1

Class – X                                                                              Subject – Artificial Intelligence
Max. Time: 2 Hours                                                              Max. Marks: 50

---

General Instructions:

1. This Question Paper consists of 21 questions in two sections: Section A and Section B.
2. Section A contains Objective type questions while Section B contains Subjective type questions.
3. Out of the given 21 questions, a candidate must answer 15 questions.
4. Attempt all questions of a section in the correct order.

---

---

Section A: Objective Type Questions (24 Marks)

---

Q1: Answer any 4 out of the given 6 questions on Employability Skills (1 x 4 = 4 Marks)

i. Identify the type of skill that focuses on building and maintaining relationships, effective communication, and team collaboration.
a) Technical skills
b) Self-management skills
c) Interpersonal skills
d) Analytical skills

ii. Which of these barriers may occur when people do not share a common language?
a) Physical barrier
b) Organizational barrier
c) Interpersonal barrier
d) Linguistic barrier

iii. Self-motivation can be improved by focusing on positive affirmations and achieving small goals. What skill does this best relate to?
a) Green skills
b) Self-management skills
c) ICT skills
d) Communication skills

iv. When you think of a solution that benefits both the company and the environment, which of the following skills are you most likely using?
a) Communication
b) Green skills
c) Entrepreneurial skills
d) ICT skills

v. Which skill is used when balancing work activities and time to prevent burnout?
a) Technical skills
b) Time management
c) Interpersonal skills
d) Organizational skills

vi. Which SDG would focus on providing affordable and clean energy to communities?
a) Life on land
b) Climate action
c) Affordable and clean energy
d) Zero hunger

---

Q2: Answer any 5 out of the given 6 questions on AI Concepts (1 x 5 = 5 Marks)

i. Which of the following concepts enables AI to understand and process human language?
a) Data Science
b) Computer Vision
c) Natural Language Processing
d) Robotics

ii. Assertion (A): Machines that are capable of learning and adapting based on new information are classified as AI-enabled.
Reason (R): Only machines specifically programmed for each task are considered AI.
a) Both A and R are correct and R is the correct explanation of A
b) Both A and R are correct, but R is not the correct explanation of A
c) A is correct, but R is incorrect
d) Both A and R are incorrect

iii. An AI-powered security camera system identifies faces in real-time video feeds. This is an application of:
a) Data Science
b) Computer Vision
c) Natural Language Processing
d) AI Ethics

iv. Which of these is used to train AI models on large datasets, allowing them to improve performance over time?
a) Data Analysis
b) Manual Coding
c) Neural Networks
d) Robotics

v. What does the AI Project Cycle’s Evaluation stage involve?
a) Acquiring data for training
b) Building a predictive model
c) Testing the model’s performance
d) Identifying project objectives

vi. Which of the following is an example of a training data source in the AI Project Cycle?
a) Expert opinions
b) Sample data from past cases
c) Analytical reports
d) Theoretical frameworks

---

Q3: Answer any 5 out of the given 6 questions (1 x 5 = 5 Marks)

i. Which of the following pairs are NOT examples of AI-based applications?
a) Self-driving car and face recognition system
b) Motion sensor and chatbot
c) Image classification and voice assistant
d) Sentiment analysis and predictive text

ii. In the context of AI, the term “features” refers to:
a) The underlying technical structure of an AI model
b) Characteristics in the data that influence predictions
c) Physical elements used for data collection
d) A dataset’s accuracy measurement

iii. When analyzing large volumes of textual data for topic categorization, which AI field is most applicable?
a) Robotics
b) Natural Language Processing
c) Data Science
d) Computer Vision

iv. Which part of the AI Project Cycle involves gathering information on variables that affect the project?
a) Data Acquisition
b) Modelling
c) Data Evaluation
d) Problem Scoping

v. A system capable of learning independently and adapting its functionality over time is an example of:
a) Rule-Based Approach
b) Deep Learning
c) Reinforcement Learning
d) Data Mining

vi. Which of these describes an application that categorizes images by analyzing visual patterns?
a) Sentiment Analysis
b) Object Detection
c) Text Classification
d) Speech Recognition

---

Q4: Answer any 5 out of the given 6 questions (1 x 5 = 5 Marks)

i. What is the main difference between training data and testing data?
a) Training data is used to evaluate the model’s accuracy, while testing data builds it
b) Training data builds the model, while testing data assesses its performance
c) Testing data is typically larger than training data
d) Training data includes feedback, while testing data does not

ii. Identify the type of neural network that can generate text by predicting the next likely word in a sentence.
a) Convolutional Neural Network
b) Recurrent Neural Network
c) Deep Neural Network
d) Feed-forward Network

iii. An AI model that fails to perform well on new data is likely:
a) Underfitted
b) Overfitted
c) Untrained
d) Calibrated

iv. Pixels in an image refer to:
a) The smallest components of data processing
b) The visual blocks that make up an image
c) Software for image data extraction
d) Analytical tools used in AI models

v. Assertion (A): Data privacy issues may arise when using AI applications for personal services.
Reason (R): AI models rely on continuous data input from users to optimize performance.
a) Both A and R are correct and R explains A
b) Both A and R are correct but R does not explain A
c) A is correct, but R is incorrect
d) Both A and R are incorrect

vi. What feature allows AI systems to recognize sarcasm or tone in written text?
a) Sentiment Analysis
b) Syntax Parsing
c) Entity Extraction
d) Image Classification

---

Q5: Answer any 5 out of the given 6 questions (1 x 5 = 5 Marks)

i. Natural Language Processing is used in customer support chatbots to:
a) Visualize data
b) Predict consumer trends
c) Interact with users in human language
d) Automate logistic operations

ii. Object detection in Computer Vision primarily involves:
a) Separating text from images
b) Identifying objects in images or videos
c) Translating languages
d) Performing sentiment analysis

iii. The term “Bag of Words” in NLP refers to:
a) Summarizing a document by eliminating filler words
b) Counting word frequency without considering context
c) Creating word predictions based on past inputs
d) Analyzing emotions in text

iv. The output layer of a neural network is the layer that:
a) Defines the number of hidden layers in the model
b) Takes raw data input
c) Produces the final output prediction
d) Processes input data

v. Which type of machine learning requires labeled data to learn patterns?
a) Unsupervised Learning
b) Supervised Learning
c) Reinforcement Learning
d) Active Learning

vi. In the AI model evaluation process, high recall is significant when:
a) False positives must be minimized
b) Accuracy is the main focus
c) Precision is less important
d) All outcomes are equally relevant

---

Section B: Subjective Type Questions (26 Marks)

---

Q6 – Q10: Answer any 3 out of the given 5 questions on Employability Skills (2 x 3 = 6 Marks)

6. Describe how Green Skills contribute to sustainable development.
7. Explain two ways in which ICT skills can enhance productivity.
8. Define time management and discuss its importance in achieving personal goals.
9. State two benefits of setting clear goals in entrepreneurship.
10. List two practices that can help reduce stress in a workplace environment.

---

Q11 – Q16: Answer any 4 out of the given 6 questions (2 x 4 = 8 Marks)

11. Explain how bias in AI can affect decision-making processes.
12. Briefly describe two ethical concerns associated with data privacy in AI.
13. Define overfitting and provide an example of how it might occur in an AI model.
14. Explain why the AI Project Cycle’s problem-scoping phase is crucial for project success.
15. Differentiate between Data Science and Computer Vision, including one example for each.
16. How does Natural Language Processing help improve customer service?

---

Q17 – Q21: Answer any 3 out of the given 5 questions (4 x 3 = 12 Marks)

17. Define Artificial Intelligence and explain the process by which machines become artificially intelligent.
18. Describe the role and significance of the 4W Problem Canvas in identifying project objectives.
19. Compare Supervised, Unsupervised, and Reinforcement Learning, giving one example of each.
20. Discuss the concept of precision and recall and explain when each would be prioritized in AI applications.

21: An AI model made the following predictions for a new product’s customer satisfaction  feedback (satisfied or unsatisfied) based on initial surveys. Use the confusion matrix below to answer the following:

	Actual: Yes (Satisfied)	Actual: No (Unsatisfied)
Predicted: Yes (Satisfied)	70	30
Predicted: No (Unsatisfied)	15	35

(i) Identify the total number of wrong predictions made by the model.
(ii) Calculate the Precision, Recall, and F1 Score.

---

Marking Scheme

Q1: Answer any 4 out of the given 6 questions on Employability Skills.

1. Answer: c) Interpersonal skills
2. Answer: d) Linguistic barrier
3. Answer: b) Self-management skills
4. Answer: b) Green skills
5. Answer: b) Time management
6. Answer: c) Affordable and clean energy

---

Q2: Answer any 5 out of the given 6 questions on AI Concepts.

1. Answer: c) Natural Language Processing
2. Answer: c) A is correct, but R is incorrect
3. Answer: b) Computer Vision
4. Answer: c) Neural Networks
5. Answer: c) Testing the model’s performance
6. Answer: b) Sample data from past cases

---

Q3: Answer any 5 out of the given 6 questions.

1. Answer: b) Motion sensor and chatbot
2. Answer: b) Characteristics in the data that influence predictions
3. Answer: b) Natural Language Processing
4. Answer: a) Data Acquisition
5. Answer: b) Deep Learning
6. Answer: b) Object Detection

---

Q4: Answer any 5 out of the given 6 questions.

1. Answer: b) Training data builds the model, while testing data assesses its performance
2. Answer: b) Recurrent Neural Network
3. Answer: b) Overfitted
4. Answer: b) The visual blocks that make up an image
5. Answer: a) Both A and R are correct, and R explains A
6. Answer: a) Sentiment Analysis

---

Q5: Answer any 5 out of the given 6 questions.

1. Answer: c) Interact with users in human language
2. Answer: b) Identifying objects in images or videos
3. Answer: b) Counting word frequency without considering context
4. Answer: c) Produces the final output prediction
5. Answer: b) Supervised Learning
6. Answer: a) False positives must be minimized

---

Section B: Subjective Type Questions

---

Q6 – Q10: Answer any 3 out of the given 5 questions on Employability Skills.

6. Answer: Green Skills help individuals and organizations develop environmentally responsible behaviors and make sustainable decisions, such as conserving energy, reducing waste, and recycling. These skills contribute to sustainable development by ensuring resources are used responsibly, thereby reducing environmental impact and supporting the conservation of natural resources.
7. Answer: ICT skills can enhance productivity by improving communication and collaboration among team members, especially in remote settings, through tools like video conferencing and project management software. Additionally, ICT allows for efficient data management and automation of repetitive tasks, which saves time and reduces manual errors.
8. Answer: Time management involves organizing and planning how to allocate time effectively to various tasks. It is important for achieving personal goals as it helps prioritize tasks, reduce stress, and increase productivity, ensuring that important tasks are completed within set deadlines and helping maintain a balanced workload.
9. Answer: Setting clear goals in entrepreneurship provides direction and motivation, enabling entrepreneurs to stay focused on specific objectives and measure their progress. Clear goals also facilitate strategic planning and decision-making, which are essential for the growth and sustainability of the business.
10. Answer: Two practices that can help reduce stress in the workplace include setting clear priorities to manage workloads efficiently and taking short breaks to refresh and avoid burnout. Regular team-building activities can also improve morale and provide a supportive work environment.

---

Q11 – Q16: Answer any 4 out of the given 6 questions.

11. Answer: Bias in AI can affect decision-making by leading to unfair or inaccurate outcomes if the model favors certain data points over others. For instance, if an AI system trained on historical hiring data shows a preference for a specific demographic group, it may continue this bias in future hiring recommendations, potentially leading to discriminatory practices.
12. Answer: Two ethical concerns associated with data privacy in AI include unauthorized data collection, where personal data is gathered without user consent, and data misuse, where data is shared or used for purposes outside the original intent, risking user privacy. These issues highlight the need for transparency, consent, and strict data handling practices in AI systems.
13. Answer: Overfitting occurs when an AI model performs well on the training data but poorly on new or unseen data. This might happen if the model has memorized the training examples rather than learning general patterns. For instance, an AI model trained to recognize handwritten digits might fail on new samples if it only memorized the specific examples used in training.
14. Answer: Problem-scoping is crucial in the AI Project Cycle because it helps define the project’s objectives, target audience, and limitations, ensuring that the AI model developed addresses a relevant and specific issue. Clear problem-scoping guides data collection and model selection, increasing the likelihood of project success.
15. Answer: Data Science involves analyzing large datasets to extract useful information and trends, often for predictive modeling, such as customer behavior analysis in marketing. Computer Vision, on the other hand, focuses on enabling machines to interpret visual data, like recognizing faces in images for security systems.
16. Answer: Natural Language Processing (NLP) improves customer service by allowing AI-powered chatbots and virtual assistants to understand and respond to customer queries in real-time. NLP processes human language, making it easier for businesses to provide instant support, enhancing customer satisfaction and reducing response times.

---

Q17 – Q21: Answer any 3 out of the given 5 questions.

17. Answer: Artificial Intelligence (AI) is the simulation of human intelligence in machines that are programmed to think and act like humans. Machines become artificially intelligent by learning from vast amounts of data, identifying patterns, and improving their predictions or decisions based on feedback. For example, a chatbot can become artificially intelligent by processing user input, recognizing common phrases, and gradually improving its responses.
18. Answer: The 4W Problem Canvas is a tool used in AI problem scoping to analyze four main questions: Who is affected, What is the problem, Where does it occur, and Why does it matter? It helps project teams identify key stakeholders, understand the nature and impact of the problem, and clarify the goals. This structured approach provides a comprehensive view, ensuring that the problem is well-defined and aligned with stakeholder needs.
19. Answer: Supervised Learning involves training an AI model on labeled data where each input has an associated output, such as spam email detection where emails are labeled as “spam” or “not spam.” Unsupervised Learning does not use labeled data, so the model identifies patterns or clusters within the data, such as grouping customers with similar buying behaviors. Reinforcement Learning, however, trains a model based on rewards or penalties from actions taken in a specific environment, such as a game-playing AI that improves by maximizing its score.
20. Answer: Precision measures the accuracy of positive predictions in a model and is prioritized in situations where false positives are costly, such as spam detection. Recall, on the other hand, measures the model’s ability to identify all relevant instances and is prioritized when missing positive instances is costly, such as in medical diagnoses. Choosing between precision and recall depends on the specific needs of the application.
21. Total wrong predictions: 45

Precision: 0.7 (70%)

Recall: 0.82 (82%)

F1 Score: 0.755 (75.5%)