Students must start practicing the questions from Class 12 AI Important Questions and CBSE Class 12 AI Question Paper 2024 are designed as per the revised syllabus.
CBSE Class 12 AI Exam Question Paper 2024 with Solutions
Question 1.
Answer any 4 out of the given 6 questions on Employability Skills. (4 × 1 = 4)
(i) ________ is a form of communication that allows students to put their feelings and ideas on paper, to organize their knowledge and beliefs into convincing arguments, and to convey meaning through well-constructed text.
(a) Active listening
(b) Writing
(c) Absolute phrase
(d) Speeches
Answer:
(b) Writing
(ii) The term OCPD stands for _____.
(a) Obsessive ‘compulsive personality disorder
(b) Operational compulsive personality disorder
(c) Obsessive compulsive personality defect
(d) Organised compulsive professional disorder
Answer:
(a) Obsessive ‘compulsive personality disorder
(iii) Identify the incorrect statement from the following:
(a) Motivation and positive thinking can help us overcome fears and take up new challenges.
(b) Motivation and positive thinking can help us in ignoring our duties.
(c) An individual’s motivation may come from within or be inspired by others or events.
(d) Directing behavior towards certain motive or goal is the essence of motivation.
Answer:
(b) Motivation and positive thinking can help us in ignoring our duties.
(iv) Which of the following statements is NOT true for spreadsheet?
(a) A workbook has one or more worksheets.
(b) Large volumes of data can be easily handled and manipulated.
(c) Data cannot be easily represented in pictorial form like graphs or charts.
(d) Built-in functions make calculations easier, faster and more accurate.
Answer:
(c) Data cannot be easily represented in pictorial form like graphs or charts.
(v) Which of the following is one of the barriers that an entrepreneur may face?
(a) Self-confidence
(b) Availability of monetary resources on time
(c) Unavailability of monetary resources on time
(d) Availability of skilled labour/staff
Answer:
(c) Unavailability of monetary resources on time
(vi) A _____ is defined as one that helps bring about and maintain transition to environmentally sustainable forms of production and consumption.
(a) Blue collar job
(b) White collar job
(c) Yellow job
(d) Green job
Answer:
(d) Green job
Question 2.
Answer any 5 out of the given 6 questions (5 × 1 = 5)
(i) During Train-Test split evaluation, we usually split the data around _______ between testing and training stages.
(a) 90% – 10%
(b) 20% – 80%
(c) 100% – 0%
(d) 0% – 100%
Answer:
(b) 20% – 80%
(ii) With reference to Data storytelling, complete the given statement: “Data can be persuasive, but _____ are much more.”
(a) Machines
(b) Projects
(c) Stories
(d) Humans
Answer:
(c) Stories
(iii) _____ provides a useful abstract model for thinking about time series generally and for better understanding problems during time series analysis and forecasting.
(a) Decomposition
(b) Modelling
(c) Stage
(d) Building
Answer:
(a) Decomposition
(iv) The first fundamental step when starting an AI initiative is- \qquad
(a) Evaluation
(b) Testing
(c) Development
(d) Scoping
Answer:
(d) Scoping
(v) Which of the following is not one of the steps of an AI project life cycle?
(a) Problem definition
(b) Understanding the problem
(c) Data delivery
(d) Data gathering
Answer:
(d) Data gathering
(vi) Which of the following does not come under open language category.
(a) Linux
(b) Python
(c) R
(d) Scala
Answer:
(a) Linux
Question 3.
Answer any 5 out of the given 6 questions. (5 × 1 = 5)
(i) _____ is the last step in the AI project life cycle.
(a) Problem definition
(b) Data gathering
(c) Deployment
(d) Evaluation
Answer:
(c) Deployment
(ii) Identify the given element that makes a compelling data story and choose its correct name from the following options:
(a) Graphs
(b) Numbers
(c) Story
(d) Data
Answer:
(d) Data
(iii) In _______ phase, it’s crucial to precisely define the strategic business objectives and desired outcomes of the project.
(a) Design
(b) Deployment
(c) Testing
(d) Requirement analysis
Answer:
(d) Requirement analysis
(iv) Assertion (A) : With reference to Data storytelling, narrative is the way we simplify and make sense of a complex world.
Reason (R) : Narrative explains what is going on within the dataset.
Select the appropriate option for the statements given above:
(a) Both (A) and (R) are true and (R) is the correct explanation of (A)
(b) Both (A) and (R) are true and (R) is not the correct explanation of (A)
(c) (A) is true but (R) is false
(d) (A) is false but (R) is true
Answer:
(a) Both (A) and (R) are true and (R) is the correct explanation of (A)
(v) AI is perhaps the most transformative technology available today. At a high level, every AI project follows total–steps.
(a) Six
(b) Seven
(c) Eight
(d) Infinite
Answer:
(b) Seven
(vi) During step 3 of AI model life cycle,——-should include all relevant subsets of training data.
(a) Relevant Data
(b) Deployment
(c) Test data
(d) Scoping
Answer:
(c) Test data
Question 4.
Answer any 5 out of the given 6 questions. (5 × 1 = 5)
(i) Match the following:
| 1. Open Frameworks | A. AutoAI |
| 2. Open Language | B. Anaconda |
| 3. Development tools | C. Python |
| 4. Productivity-enhancing capabilities | D. XGBoost |
(a) 1-D, 2-A, 3-B, 4-C
(b) 1-D, 2-C, 3-B, 4-A
(c) 1-B, 2-A, 3-D, 4-C
(d) 1-C, 2-B, 3-A, 4-D
Answer:
(b)
| 1. Open Frameworks | D. XGBoost |
| 2. Open Language | C. Python |
| 3. Development tools | B. Anaconda |
| 4. Productivity-enhancing capabilities | A. AutoAI |
(ii) Stories that incorporate ______ and analytics are more convincing than those based entirely on anecdotes or personal experience.
(a) Suspense
(b) Humour
(c) Data
(d) Energy
Answer:
(c) Data
(iii) During modeling approach of Capstone project, the data scientist will use a _________ set for predictive modeling.
(a) Training
(b) Testing
(c) Valuable
(d) Known
Answer:
(a) Training
(iv) As Data Storytelling is a structured approach for communicating insights drawn from data, and invariably involves a combination of key elements.
When the _____ is accompanied with data, it helps to explain the audience what’s happening in the data and why a particular insight has been generated.
(a) Data
(b) Visuals
(c) Narrative
(d) Story
Answer:
(c) Narrative
(v) With reference to AI Model Life Cycle, which of the following is true for Building the Model?
(a) This is arguably the most important part of your AI project.
(b) Phrase that characterizes this project stage: “garbage in, garbage out”.
(c)This stage involves the planning and motivational aspects of your project.
(d) It is essentially an iterative process comprising all the steps relevant to building the AI or machine learning model.
Answer:
(d) It is essentially an iterative process comprising all the steps relevant to building the AI or machine learning model.
(vi) RMSE stands for- _____.
(a) Root Median Squared Error
(b) Radian Mean Squared Error
(c) Root Mean Search Error
(d) Root Mean Squared Error
Answer:
(d) Root Mean Squared Error
Question 5.
Answer any 5 out of the given 6 questions (5 × 1 = 5)
(i) Good stories don’t just emerge from data itself; they need to be unraveled fromrelationship.
(a) Data
(b) Numbers
(c) Charts
(d) Computer
Answer:
(a) Data
(ii) The train-test procedure is appropriate when there is a sufficiently———data sets available.
(a) Comparative
(b) Large
(c) Small
(d) Equal
Answer:
(b) Large
(iii) During the third step of AI Model Life Cycle, the volume of test data can be large, which presents_______.
(a) Complexities
(b) Accuracy
(c) Efficiency
(d) Redundancy
Answer:
(a) Complexities
(iv) In ______ we run our modeling process on different subsets of the data to get multiple measures of model quality.
(a) Train-Test Split
(b) Regression
(c) Cross-validation
(d) Machine learning
Answer:
(c) Cross-validation
(v) The machine learning life cycle is the _____ process that AI or machine learning projects follow.
(a) Irreversible
(b) Cyclic
(d) One-time
(d) Static
Answer:
(b) Cyclic
(vi) Data Modeling focuses on developing models that are either descriptive or-
(a) Inclusive
(b) Predictive
(c) Selective
(d) Reactive
Answer:
(b) Predictive
Section-B
Subjective Type Questions
Answer any 3 out of the given 5 questions on Employability Skills. Answer each question in 2030 words: (3 × 2 = 6)
Question 6.
Differentiate between ‘sentence’ and ‘phrase’ with the help of suitable example.
Answer:
In linguistics, a sentence is a grammatically complete unit of language that typically expresses a complete thought or idea. It consists of a subject and a predicate, and it can stand alone as a complete statement.
Example of a sentence:
“The cat chased the mouse.”
A phrase, on the other hand, is a group of related words within a sentence that function together as a single part of speech but does not contain both a subject and a predicate. It can’t stand alone as a complete sentence.
Example of a phrase:
“The cat”
In this example, “The cat” is a noun phrase because it consists of a noun (“cat”) modified by a determiner (“the”). It doesn’t express a complete thought on its own but acts as a part of a larger sentence.
Question 7.
Briefly explain the following terms:
(a) Personality
(b) Personality disorders
(a) Personality refers to the unique set of psychological traits, patterns of behavior, thoughts, and emotions that characterize an individual’s consistent and enduring way of interacting with the world. It encompasses various aspects of an individual’s identity, including their attitudes, values, beliefs, motivations, and interpersonal relationships. Personality traits are relatively stable over time and across different situations, influencing how individuals perceive and respond to their environment. Psychologists often study personality through various theoretical frameworks and assessment methods to understand individual differences and predict behavior.
(b) Personality disorders-Personality disorders are a group of mental health conditions characterized by pervasive and inflexible patterns of behavior, thoughts, and emotions that deviate significantly from societal norms and cause impairment in functioning and interpersonal relationships. These patterns are typically longstanding and deeply ingrained, causing distress and problems in various areas of life, such as work, social interactions, and personal development.
There are several types of personality disorders, each with its own unique set of symptoms and characteristics. Some common examples include borderline personality disorder, antisocial personality disorder, narcissistic personality disorder, and obsessive-compulsive personality disorder.
Individuals with personality disorders often struggle with understanding and regulating their emotions, maintaining stable relationships, and adapting to changing circumstances. Treatment for personality disorders typically involves psychotherapy, such as cognitive-behavioral therapy or dialectical behavior therapy, aimed at helping individuals develop insight into their patterns of behavior and learn healthier ways of coping and relating to others. In some cases, medication may also be prescribed to manage symptoms such as depression or anxiety that co-occur with the personality disorder.
Question 8.
Mr. Chowdhary wants to explain the working of a product to his clients. To make an impact on their audience, either he can use homemade charts or make a digital presentation using a computer and presentation software. Which out of the two options is more advantageous and why? Give any three points to support your answer.
Answer:
Using a digital presentation with computer software is more advantageous for Mr. Chowdhary to explain the working of a product to his clients. Here are three points to support this answer:
Visual Appeal: Digital presentations allow for the integration of multimedia elements such as images, videos, and animations, which can significantly enhance the visual appeal and engagement of the presentation. Homemade charts may be limited in their ability to convey complex information effectively, whereas digital presentations can offer dynamic visuals that better illustrate the product’s features and functionalities.
Interactivity: Presentation software often includes features that enable interactivity, such as clickable links, navigation buttons, and interactive elements. This interactivity can facilitate a more engaging and immersive experience for the audience, allowing them to explore the product’s details at their own pace and delve deeper into specific areas of interest. Homemade charts, on the other hand, may lack this level of interactivity and may not be as engaging or interactive for the audience.
Professionalism and Versatility: Digital presentations are perceived as more professional and polished compared to homemade charts. Presentation software offers a wide range of design templates, layouts, and customization options that allow Mr. Chowdhary to create visually appealing and professional-looking presentations tailored to his audience and branding requirements. Additionally, digital presentations can be easily updated, edited, and shared electronically, making them more versatile and convenient for Mr. Chowdhary to use in various settings and contexts.
Question 9.
What do you mean by interpersonal skills? Why is it importan’t for an entrepreneur to possess it? Briefly discuss.
Answer:
Interpersonal skills, also known as people skills or social skills, refer to the ability to communicate, interact, and build positive relationships with others effectively. These skills encompass a range of behaviors and qualities, including verbal and nonverbal communication, active listening, empathy, conflict resolution, teamwork, and emotional intelligence.
For an entrepreneur, possessing strong interpersonal skills is essential for several reasons:
1. Building Relationships: Entrepreneurs need to interact with various stakeholders, including customers, employees, investors, suppliers, and business partners. Effective interpersonal skills enable entrepreneurs to establish and maintain positive relationships with these individuals, fostering trust, loyalty, and collaboration, which are crucial for the success of their ventures.
Communication: Clear and effective communication is vital for conveying ideas, sharing information, articulating vision and goals, negotiating deals, and resolving conflicts. Entrepreneurs with strong interpersonal skills can communicate persuasively, listen actively, and adapt their communication style to different audiences, ensuring mutual understanding and alignment of objectives.
Leadership and Teamwork: Entrepreneurs often lead teams of employees or collaborators to achieve common objectives. Strong interpersonal skills enable entrepreneurs to inspire, motivate, and empower their team members, fostering a supportive and collaborative work environment conducive to creativity, innovation, and productivity.
Customer Relations: Understanding customers’ needs, preferences, and feedback is critical for developing products or services that meet market demands and deliver value. Entrepreneurs with strong interpersonal skills can engage with customers effectively, gather valuable insights, build rapport, and address concerns, leading to improved customer satisfaction and loyalty.
Networking and Opportunities: Networking plays a crucial role in entrepreneurship, providing access to resources, expertise, partnerships, and opportunities for growth and expansion. Entrepreneurs with strong interpersonal skills can network effectively, establish valuable connections, and leverage these relationships to open doors, attract investment, secure partnerships, and access new markets.
Question 10.
Mention any four advantages of Green jobs.
Answer:
Environmental Benefits: Green jobs contribute to sustainable development and environmental conservation by promoting practices that minimize resource consumption, reduce pollution, and mitigate climate change. These jobs focus on renewable energy, energy efficiency, waste management, conservation, and other environmentally friendly practices, helping to protect ecosystems, preserve natural resources, and combat environmental degradation.
Economic Growth: Green jobs stimulate economic growth and create employment opportunities in emerging sectors of the green economy. Investments in renewable energy, clean technologies, green infrastructure, and sustainable practices drive innovation, entrepreneurship, and productivity, leading to job creation, business development, and economic diversification. Green industries also offer long-term stability and resilience, reducing dependency on fossil fuels and volatile markets.
Health and Well-being: Green jobs contribute to improved public health and well-being by reducing exposure to hazardous substances, pollutants, and environmental risks. Green initiatives such as energy-efficient buildings, clean transportation, and sustainable agriculture promote healthier living environments, enhance air and water quality, and reduce the prevalence of respiratory diseases, allergies, and other health issues associated with pollution and environmental degradation.
Social Equity: Green jobs offer opportunities for inclusive and equitable economic development, benefiting marginalized communities, disadvantaged populations, and vulnerable groups disproportionately affected by environmental injustices and economic disparities. By promoting inclusive hiring practices, workforce development, and community engagement, green initiatives can help address social inequalities, promote social cohesion, and empower underrepresented individuals and communities to participate in the green economy and share in its benefits.
Answer any 4 out of the given 6 questions in 20-30 words each. (4 × 2 = 8)
Question 11.
What is training set?
Answer:
In the context of machine learning and data science, a training set refers to a subset of data that is used to train a machine learning model. It consists of a collection of input data points along with their corresponding output labels or target values. The training set serves as the basis for the model to learn the patterns, relationships, and underlying structure present in the data, enabling it to make predictions or classifications on new, unseen data.
The training process involves feeding the input data from the training set into the machine learning model and adjusting the model’s parameters or weights iteratively based on the errors or discrepancies between the predicted outputs and the actual labels. Through this iterative process of optimization, the model gradually learns to generalize from the training data and make accurate predictions on unseen data.
It’s important for the training set to be representative of the overall data distribution and to cover a diverse range of examples to ensure that the model learns robust and generalizable patterns. Additionally, the training set should be sufficiently large to capture the complexity of the underlying data and prevent overfitting, where the model memorizes the training data rather than learning meaningful patterns.
Once the training process is complete, the performance of the trained model is typically evaluated using a separate validation set or test set to assess its accuracy, generalization ability, and performance on unseen data.
Question 12.
Name the two categories of loss functions.
Answer:
The two categories of loss functions in machine learning are:
Regression Loss Functions: These loss functions are used when the machine learning task involves predicting continuous numerical values. Regression loss functions measure the discrepancy between the predicted values generated by the model and the actual target values in the dataset. Common examples of regression loss functions include Mean Squared Error (MSE), Mean Absolute Error (MAE), and Huber loss.
Classification Loss Functions: These loss functions are used when the machine learning task involves predicting discrete class labels or probabilities. Classification loss functions measure the difference between the predicted class probabilities or scores assigned by the model and the true class labels in the dataset. Common examples of classification loss functions include CrossEntropy Loss (also known as Log Loss), Hinge Loss (used in Support Vector Machines), and Binary Cross-Entropy Loss.
Question 13.
“Stories create engaging experiences that transport the audience to another space and time” Justify this statement.
Answer:
Emotional Connection: Stories have the ability to evoke emotions and resonate with the audience on a deep, personal level. By weaving together characters, plotlines, and settings, stories can tap into universal themes and human experiences, allowing audiences to empathize with the characters and become emotionally invested in their journey.
Imagination and Creativity: Stories stimulate the imagination and engage the audience’s senses, transporting them to different worlds, cultures, and time periods. Through vivid descriptions, rich imagery, and compelling narratives, stories spark curiosity and creativity, inviting audiences to explore new perspectives and possibilities beyond their own lived experiences.
Suspension of Disbelief: Stories suspend the audience’s disbelief and create a sense of immersion, allowing them to temporarily escape from reality and enter the fictional universe of the story. Whether it’s a fantastical adventure, a historical drama, or a science fiction epic, stories have the power to captivate the imagination and make the audience feel like they are part of the narrative unfolding before them.
Shared Experience: Stories create shared experiences that bring people together and foster a sense of community and connection. Whether it’s through books, films, theater, or oral storytelling traditions, stories have been passed down through generations, uniting people across cultures, languages, and generations through a common thread of shared narratives and shared emotions.
Overall, stories have a profound impact on the human psyche, offering a means of escape, exploration, and connection that transcends the boundaries of time and space. Through storytelling, audiences are transported to another world where they can laugh, cry, learn, and grow alongside the characters, creating memorable and meaningful experiences that linger long after the story ends.
Question 14.
What is a Capstone project? Give any two examples.
Answer:
A Capstone project is a culminating academic project that integrates and applies the knowledge and skills acquired throughout a course of study or program. It serves as a comprehensive demonstration of the student’s mastery of the subject matter and their ability to solve real-world problems, conduct independent research, and produce high-quality work. Capstone projects are often undertaken in the final year or semester of an academic program and may take various forms depending on the discipline and requirements of the institution.
Examples of Capstone projects:
Business and Management: In a business or management program, a Capstone project might involve developing a comprehensive business plan for a startup venture or an existing company. Students would conduct market research, analyze industry trends, develop financial projections, and create a strategic roadmap for launching or expanding the business. The project may culminate in a formal presentation to a panel of industry experts or potential investors.
Engineering and Technology: In an engineering or technology program, a Capstone project could involve designing and prototyping a new product or system to address a specific engineering challenge or societal need. For example, students might design a renewable energy system, develop a mobile application to improve healthcare delivery, or build a prototype for a sustainable transportation solution. The project would typically involve research, design, testing, and iteration, with the goal of producing a functional prototype or proof-of-concept demonstration.
These are just two examples of the many types of Capstone projects that students may undertake across various academic disciplines. The specific nature and scope of a Capstone project can vary widely depending on factors such as the academic program, student interests, faculty expertise, and industry partnerships.
Question 15.
Name the two techniques that can be used to validate AI model quality.
Answer:
Two techniques commonly used to validate the quality of an AI model are:
Cross-Validation: Cross-validation is a technique used to assess how well a predictive model generalizes to new, unseen data. It involves splitting the available dataset into multiple subsets, typically referred to as folds. The model is trained on a portion of the data (training set) and evaluated on the remaining data (validation set). This process is repeated multiple times, with each fold serving as both the training and validation set in turn. Cross-validation helps to estimate the model’s performance more accurately by reducing the variance introduced by a single train-test split.
Holdout Validation: Holdout validation is a simple technique where the available dataset is divided into two subsets: a training set and a validation set (or test set). The model is trained on the training set and then evaluated on the validation set to assess its performance. Holdout validation provides a straightforward way to estimate how well the model generalizes to new data. However, the performance estimate may be sensitive to the particular split of the data, especially with smaller datasets.
Question 16.
Name any two open frameworks and two development tools that can be used to build an AI model.
Answer:
Two open frameworks commonly used for building AI models are:
TensorFlow: TensorFlow is an open-source machine learning framework developed by Google Brain. It provides a comprehensive ecosystem of tools, libraries, and resources for building and deploying machine learning models across a range of domains, including deep learning, neural networks, and reinforcement learning. TensorFlow offers flexibility, scalability, and support for both research and production-level applications.
PyTorch: PyTorch is an open-source machine learning library developed by Facebook’s AI Research lab (FAIR). It is known for its dynamic computation graph, which allows for more flexibility and intuitive model development compared to static graph frameworks like, TensorFlow. PyTorch is widely used for research, prototyping, and production deployment of deep learning models.
Two development tools commonly used for building AI models are:
Jupyter Notebook: Jupyter Notebook is an open-source web application that allows users to create and share documents containing live code, equations, visualizations, and narrative text. It supports various programming languages, including Python, R, and Julia, making it a popular choice for interactive data analysis, prototyping, and collaborative research in AI and machine learning projects.
scikit-learn: scikit-learn is an open-source machine learning library for Python that provides simple and efficient tools for data mining and analysis. It includes a wide range of algorithms for classification, regiression, clustering, dimensionality reduction, and model evaluation, as well as utilities for data preprocessing, feature extraction, and model selection. scikit-learn is widely used for building and deploying machine learning models in both research and production environments.
Answer any 3 out of the given 5 questions in 50-80 words each. (3 × 4 = 12)
Question 17.
List any four importance of Data Storytelling.
Answer:
Data stopytelling is a powerful technique that enables organizations to effectively communicate insights and findings derived from data. Here are four key reasons why data storytelling is important:
Enhances Understanding: Data storytelling helps simplify complex data and analytical findings into compelling narratives that are easy to understand and interpret by non-technical audiences. By presenting data in the form of a story, with context, characters, and a plot, it enables stakeholders to grasp the significance of the data and its implications more intuitively, leading to better decision-making.
Drives Engagement: Data storytelling captivates the audience’s attention and fosters emotional engagement by weaving together data, visuals, and narrative elements into a cohesive and persuasive storyline. By appealing to both the rational and emotional aspects of human cognition, data storytelling makes data more relatable, memorable, and impactful, encouraging active participation and buy-in from stakeholders.
Facilitates Action: Effective data storytelling goes beyond merely presenting information-it motivates and inspires action by framing data-driven insights in the context of real-world challenges, opportunities, and goals. By highlighting the implications of the data and suggesting actionable recommendations or next steps, data storytelling empowers decision-makers to translate insights into tangible outcomes and drive positive change within their organizations.
Builds Trust: Data storytelling builds trust and credibility by providing transparency and context around the data analysis process, assumptions, and limitations. By openly acknowledging uncertainties and potential biases in the data, and presenting findings in a clear, honest, and compelling manner, data storytellers can earn the trust of their audience and foster a culture of data-driven decision-making within their organizations.
Question 18.
What is Design Thinking? List its main stages.
Answer:
Design Thinking is a human-centered approach to problem-solving and innovation that emphasizes empathy, creativity, and collaboration to develop: innovative solutions to complex challenges. It involves understanding the needs and perspectives of end-users, generating creative ideas, prototyping and testing solutions iteratively, and refining designs based on feedback. Design Thinking is widely used in various fields, including product design, service design, business strategy, and social innovation.
The main stages of Design Thinking typically include:
Empathize: In this stage, designers seek to understand the needs, motivations, and behaviors of the people they are designing for. This involves conducting research, observing users in their natural environment, and engaging in empathy-building activities to gain insights into users’ experiences and pain points.
Define: In this stage, designers synthesize the research findings from the Empathize stage to define the problem’statement or design challenge they are addressing. This involves reframing the problem from the perspective of the user and identifying specific needs, goals, and constraints that the design solution should address.
Ideate: In this stage, designers brainstorm and generate a wide range of creative ideas and potential solutions to address the problem defined in the previous stage. This involves encouraging divergent thinking, suspending judgment, and exploring unconventional ideas through techniques such as brainstorming, mind mapping, and rapid prototyping.
Prototype: In this stage, designers create tangible representations of their ideas and concepts to test and refine them. Prototypes can take various forms, from rough sketches and mockups to interactive prototypes and physical models. The goal is to quickly and cheaply explore different design possibilities and gather feedback from users to inform further iteration.
Test: In this stage, designers test the prototypes with real users to gather feedback, evaluate usability, and validate assumptions. This involves conducting user tests, usability studies, and observational research to identify strengths and weaknesses in the design and iteratively refine the prototypes based on user insights.
Iterate: Design Thinking is an iterative process, and the stages described above are often repeated multiple times as designers refine and improve their solutions based on feedback and insights gained through testing. Iteration allows designers to continuously learn and evolve their designs to better meet the needs of users and address emerging challenges and opportunities.
By following these stages of Design Thinking, designers can develop innovative solutions that are user-centered, feasible, and viable, ultimately leading to more successful outcomes and greater impact.
Question 19.
Explain the ‘Design/Building the Model” step of the AI Model lifecycle in detail.
Answer:
The “Design/Building the Model” step in the AI Model lifecycle involves the process of designing and constructing the machine learning model based on the defined problem statement or objective. This step is crucial as it lays the foundation for the entire model development process and directly influences the model’s performance, accuracy, and effectiveness in solving the problem at hand. Here is a detailed explanation of this step:
Problem Definition: The first step in designing the model is to clearly define the problem statement or objective that the model aims to address. This involves understanding the business context, identifying key stakeholders, and specifying the goals, constraints, and requirements of the project. The problem definition serves as a guiding framework for designing the model and determining the appropriate approach and techniques to be used.
Data Collection and Preparation: Once the problem is defined, the next step is to collect and preprocess the relevant data needed to train and evaluate the model. This involves gathering data from various sources, cleaning and preprocessing the data to handle missing values, outliers, and inconsistencies, and transforming the data into a suitable format for model training. Data preparation is critical as the quality and completeness of the data directly impact the performance and generalization ability of the model.
Feature Engineering: Feature engineering involves selecting, creating, and transforming the input features (variables) in the dataset to enhance the predictive power of the model. This may include selecting relevant features, encoding categorical variables, scaling numerical features, and creating new features through transformations, aggregations, or interactions. Effective feature engineering plays a crucial role in improving the model’s performance and robustness.
Model Selection and Architecture Design: In this step, the appropriate machine learning algorithm or model architecture is selected based on the nature of the problem, the characteristics of the data, and the desired outcome. This may involve choosing between different types of models such as regression, classification, clustering, or deep learning models, as well as selecting the specific parameters and hyperparameters of the chosen model. The model architecture is designed to optimize performance metrics such as accuracy, precision, recall, or AUC (Area Under the ROC Curve).
Model Training: Once the model architecture is defined, the next step is to train the model using the prepared dataset. This involves feeding the training data into the model, adjusting the model parameters iteratively through optimization algorithms such as gradient descent, and minimizing the loss function to learn the underlying patterns and relationships in the data. The model training process aims to optimize the model’s performance on the training data while avoiding overfitting, where the model memorizes the training data and fails to generalize to new, unseen data.
Model Evaluation: After the model is trained, it is evaluated using a separate validation dataset to assess its performance, generalization ability, and robustness. This involves measuring various performance metrics such as accuracy, precision, recall, F1-score, or ROC-AUC on the validation data and comparing them against predefined thresholds or benchmarks. Model evaluation helps identify any issues or deficiencies in the model’s performance and guides further iterations or improvements to the model design.
Model Deployment: Once the model has been designed, trained, and evaluated, it is ready for deployment in real-world applications. Model deployment involves integrating the trained model into the production environment, deploying it on the appropriate infrastructure (e.g., cloud, edge devices), and implementing mechanisms for monitoring, maintenance, and updates. Successful deployment of the model enables it to generate predictions or insights in real-time and deliver value to end-users or stakeholders.
Overall, the “Design/Building the Model” step in the AI Model lifecycle is a systematic and iterative process that involves defining the problem, collecting and preparing data, engineering features, selecting and designing the model architecture, training and evaluating the model, and deploying it for practical use. Each sub-step in this process plays a crucial role in ensuring the effectiveness, accuracy, and reliability of the AI model in solving real-world problems and delivering actionable insights.
Question 20.
Expand and explain the term MSE. Give the mathematical formula to calculate MSE. Why use MSE? Briefly discuss.
Answer:
MSE stands for Mean Squared Error, which is a commonly used metric in statistics and machine learning to evaluate the performance of regressiort models. It measures the average squared difference between the predicted values generated by a model and the actual observed values in a dataset.
The mathematical formula to calculate MSE is as follows:
Where:
- n is the total number of data points in the dataset.
- y i represents the actual (observed) value of the target variable for the i th data point.
- ^ y^ i represents the predicted value of the target variable for the i th data point.
To calculate MSE, you take the squared difference between each actual and predicted value, sum these squared differences across all data points, and then divide by the total number of data points.
MSE is commonly used for several reasons:
Sensitivity to Errors: Squaring the errors in the MSE formula amplifies larger errors, making MSE particularly sensitive to outliers or large deviations between actual and predicted values. This property makes it useful for detecting and penalizing models that perform poorly on specific data points, helping to identify areas where the model needs improvement.
Differentiability: MSE is a differentiable function, which means it can be easily optimized using gradient-based optimization algorithms like gradient descent. This makes MSE suitable for use as a loss function during the training of machine learning models, allowing the model parameters to be updated iteratively to minimize the MSE and improve model performance.
Mathematically Intuitive: MSE has a clear and intuitive interpretation, representing the average squared deviation between predicted and actual values. This makes it easy to understand and interpret, both for practitioners and stakeholders involved in model evaluation and decisionmaking.
Widely Used: MSE is a widely used and accepted metric in the fields of statistics and machine learning, making it easy to compare model performance across different studies, datasets, and applications. Its ubiquity also makes it a standard choice for model evaluation and selection in many machine learning projects.
Overall, MSE provides a simple yet effective measure of the accuracy and reliability of regression models, allowing practitioners to assess model performance, identify areas for improvement, and make informed decisions based on the quality of predictions generated by the model.
Question 21.
(a) Why Storytelling is so powerful and cross-cultural? Explain
(b) Which of the following is a better data story? Give reasons.
image 1.
image 2.
Answer:
(a) Stories are a powerful communication tool in a multicultural workplace, because they enable listeners to make connections between what is said and their own experiences, facilitating understanding of important meanings, beliefs, and behaviors from different cultures.
Cross-cultural communication means being fully aware of the differences and having respect for the values, customs, beliefs, and behaviours that shape the way people see in the world and interact with others.
The Key Elements of Effective Cross-Cultural Communication
Awareness: Understanding of how the societal norms, beliefs, rituals and etiquettes of individuals from different cultures might impact communication.
Language: Being able to communicate effectively in the same language as your audience or making use of interpreters to avoid misunderstandings.
Listening: Being present and listening to what someone is saying, while also paying attention to non-verbal forms of communication like facial expressions, hand gestures, posture, eye contact, tone of voice, etc.
Respect: Being mindful of and respecting the differences in cultural backgrounds and making efforts to avoid using offensive language or performing actions that might be deemed insensitive.
Openness: Being open and adaptable to new ways of communicating and expressing oneself, with a willingness to learn about other ways of doing things.
Answer:
(b) Because Image 2’s data story on the observations drawn from the image is well articulated, it has a superior data story than Image 1.
The post CBSE Class 12 AI Question Paper 2024 with Solutions appeared first on Learn CBSE.