ScholarQuill logoScholarQuillUniversity Notes
  • Notes
  • Past Papers
  • Blogs
  • Todo
Login
ScholarQuill logoScholarQuillUniversity Notes
Login
NotesPast PapersBlogsTodo
More
SubjectsDiscussionCGPA CalculatorGPA CalculatorStudent PortalCourse Outline
About
About usPrivacy PolicyReportContact
Notes
Past Papers
Blogs
Todo
Analytics
    Current Subject
    🧩
    Artificial Intelligence Past Papers
    PCOMP2121
    Progress0 / 8 topics
    Topics
    1. Artificial Intelligence SEM7 20252. Artificial Intelligence 2024 repeater3. Artificial Intelligence 20224. Artificial Intelligence 20235. Artificial Intelligence SEM6 20256. Artificial Intelligence SEM3 20257. Artificial Intelligence SEM7 20248. Artificial Intelligence SEM6 2024
    PCOMP2121›Artificial Intelligence 2024 repeater
    Artificial Intelligence Past PapersTopic 2 of 8

    Artificial Intelligence 2024 repeater

    11 minread
    1,824words
    Intermediatelevel

    Section II (Short Answer)

    Q.2- i. What is Artificial Intelligence? What are the Types of Artificial Intelligence? Definition: Artificial Intelligence (AI) is the simulation of human intelligence processes by machines, especially computer systems. It involves the ability to learn from data, reason to draw conclusions, and self-correct to solve problems. Types of AI:

    • Narrow AI (Weak AI): AI trained to perform a single, specific task (e.g., Apple's Siri, facial recognition, or spam filters). This is the only type of AI that exists today.
    • General AI (Strong AI): A theoretical machine that possesses human-level cognitive abilities and can perform any intellectual task a human can.
    • Super AI: A hypothetical system where machines surpass human intelligence in every aspect, including creativity and problem-solving.

    Q.2- ii. What are some real-life applications of Artificial Intelligence? AI is deeply embedded in our daily routines and multiple industries to improve efficiency and user experience. Real-Life Examples:

    • Entertainment & Content: Recommendation engines used by Netflix and YouTube to suggest content based on your past viewing behavior.
    • Navigation: Google Maps uses AI to analyze traffic speed, historical data, and route options to find the fastest path.
    • Healthcare: AI algorithms analyze X-rays and medical records to help doctors detect diseases earlier and more accurately.

    Q.2- iii. Which algorithm does Facebook use for face verification and how does it work? Facebook (Meta) developed a deep learning-based facial recognition algorithm called DeepFace. How it Works:

    1. Alignment: It detects a face in an image and maps it to a 3D model so the face is aligned perfectly straight.
    2. Feature Extraction: A deep neural network analyzes the facial features and generates a unique, numerical 128-dimensional representation (a "facial signature").
    3. Verification: It compares this numerical signature against stored templates of the user. If the similarity score crosses a specific threshold, it verifies the face with near-human accuracy.

    Q.2- iv. How can AI be used in detecting fraud? AI is highly effective at detecting fraud because it can analyze massive volumes of transaction data in real-time to identify anomalies that a human might miss. Example: When you make a transaction, a bank's AI checks your typical spending habits, location, and device history. If your credit card is suddenly used to buy expensive electronics in another country while your phone is still connected locally, the AI instantly flags this behavior as fraudulent and blocks the transaction.

    Q.2- v. Mention Some Popular Domains of AI? AI is a broad field made up of several specialized domains:

    • Machine Learning (ML): Teaching machines to learn and improve from data without being explicitly programmed.
    • Natural Language Processing (NLP): Enabling machines to understand, interpret, and generate human text and speech (e.g., ChatGPT).
    • Computer Vision: Giving machines the ability to "see," analyze, and interpret visual data from images or videos.
    • Robotics: Combining AI with physical hardware so machines can interact intelligently with the physical world.

    Q.2- vi. What is an expert system? What are its characteristics? Definition: An Expert System is a computer application that utilizes a knowledge base and a set of rules to emulate the decision-making capabilities of a human expert in a specific field. Characteristics:

    • High Performance: It provides highly accurate and reliable answers within its specific domain.
    • Understandable: It features an "explanation module" that can clarify its reasoning and how it arrived at a specific conclusion.
    • Domain-Specific: It is highly specialized (e.g., the MYCIN system used for diagnosing blood infections) and cannot solve problems outside its programmed scope.

    Q.2- vii. What are the limitations of current AI technology? Despite rapid advancements, current AI systems have notable limitations. They completely lack common sense, true human intuition, and emotional intelligence. They are heavily dependent on vast amounts of training data; if the data is biased, the AI will confidently produce biased results. Furthermore, AI models are often "brittle," meaning they easily fail or make absurd mistakes if presented with a scenario even slightly outside their training environment.

    Q.2- viii. Explain rational agents and rationality? Rational Agent: In AI, an agent is an entity (like a self-driving car or a software bot) that perceives its environment through sensors and acts upon it through actuators. A rational agent is one that always takes the action that achieves the best possible outcome. Rationality: This refers to the capacity to make logical decisions based on the information available. For an AI, rationality means choosing the action that maximizes its expected performance measure, given its current knowledge and built-in goals.


    Section III (Essay Type)

    Q. #.3. Give a brief introduction to the Turing test in AI?

    Introduction The Turing Test is a method of inquiry in artificial intelligence for determining whether or not a computer is capable of thinking like a human being. It was introduced by British mathematician and computer scientist Alan Turing in 1950 in his paper "Computing Machinery and Intelligence."

    The Setup The test involves three participants in isolated rooms:

    1. A human judge (the interrogator).
    2. A human responder.
    3. A machine (the AI) responder.

    The human judge communicates with both the human and the machine using only text-based messages (like a chat interface).

    How it Works The judge's goal is to determine which of the two participants is the human and which is the computer by asking them a series of questions. The machine's goal is to trick the judge into believing it is the human. The human's goal is to help the judge figure out the truth.

    Significance If the human judge cannot reliably tell the machine from the human after a series of conversations, the machine is said to have "passed" the Turing Test. Passing the test indicates that the machine exhibits intelligent behavior indistinguishable from that of a human. While it does not prove the machine possesses actual consciousness, it remains a foundational benchmark in the philosophy of AI.


    Q. #.4. What are some search strategies used in AI? Explain Depth First Search in detail?

    Search Strategies in AI In AI, search strategies are algorithms used to navigate through a problem space (often represented as a tree or graph) to find a solution path from an initial state to a goal state. They are broadly divided into two categories:

    1. Uninformed Search (Blind Search): These strategies have no additional information about the problem other than the problem definition. They search exhaustively. Examples include Breadth-First Search (BFS) and Depth-First Search (DFS).
    2. Informed Search (Heuristic Search): These use domain-specific knowledge (heuristics) to guide the search more efficiently toward the goal. Examples include A* Search and Greedy Best-First Search.

    Depth-First Search (DFS) in Detail Depth-First Search is an uninformed search algorithm that explores as far as possible along each branch before backtracking.

    • Mechanism: DFS uses a Stack data structure, which follows a Last-In, First-Out (LIFO) principle.
    • How it Works:
    1. It starts at the root node and selects one of its children to explore.
    2. It continues to move deeper and deeper down that specific path until it reaches a "leaf" node (a node with no children) or a dead end.
    3. Once it hits a dead end, it "backtracks" to the most recent node that has unexplored branches and continues the search down the new path.
    • Advantages: It requires significantly less memory than Breadth-First Search because it only needs to store the current path from the root to a leaf node, rather than all nodes at a given level.
    • Disadvantages: It is not guaranteed to find the shortest path (it is not optimal). Furthermore, if the search tree is infinitely deep, DFS can get trapped in an infinite loop and never find a solution (it is not complete).

    Q. #.5. Explain Markov's chain process with an example?

    Introduction to Markov Chain A Markov Chain is a mathematical system that experiences transitions from one state to another according to certain probabilistic rules. It is heavily used in AI for reinforcement learning, natural language processing, and predictive modeling.

    The Markov Property (Memorylessness) The defining characteristic of a Markov Chain is the "Markov Property." This property states that the future state of the process depends only on the current state, and not on the sequence of events that preceded it. In simple terms, the system is memoryless—how you arrived at the current state does not matter for predicting the next step.

    Components

    • State Space: The set of all possible states the system can be in.
    • Transition Probabilities: The probability of moving from one specific state to another.

    A Real-World Example: Predicting the Weather Imagine a simple weather model with only two states: Sunny and Rainy. According to historical data, we establish the following transition probabilities:

    • If today is Sunny, there is an 80% chance tomorrow will be Sunny, and a 20% chance it will be Rainy.
    • If today is Rainy, there is a 60% chance tomorrow will be Rainy, and a 40% chance it will be Sunny.

    If today is Sunny, the AI can use this Markov Chain to calculate the probability of the weather for the next few days. Because of the Markov property, to predict tomorrow's weather, the AI only needs to know that today is Sunny; it does not care if it rained for a whole week prior to today.


    Q. #.6. If you are starting a new business, how Will you use AI to promote your business?

    Introduction Integrating Artificial Intelligence into a new business strategy is a highly effective way to automate workflows, reduce operational costs, and rapidly scale promotional efforts. If I were launching a new business, I would deploy AI across several key marketing and growth vectors.

    1. Optimizing Online Visibility and SEO Search Engine Optimization (SEO) is critical for a new business to be found online. I would use AI-driven, plain-English SEO audit tools to quickly analyze my website's architecture and content. AI can automatically identify missing keywords, track competitor rankings, and suggest content modifications to ensure the business ranks higher on search engines, driving organic traffic without needing an expensive marketing agency.

    2. Viral Content Creation & Distribution Short-form video content is incredibly powerful for building an audience. I would use AI video editing and scripting tools to create engaging short-form content for platforms like YouTube and TikTok. AI can analyze current viewing trends, auto-generate captions, and optimize posting schedules, allowing a small business to quickly generate millions of views and a massive subscriber base with minimal manual editing.

    3. Automated Customer Service (Chatbots) To keep customers engaged, I would implement an AI-powered NLP chatbot on the business website. Instead of answering basic inquiries manually, the AI can provide instant, 24/7 support for FAQs, product availability, and order tracking. This creates a highly responsive customer experience that builds trust.

    4. Data-Driven Targeted Marketing Instead of guessing who my customers are, I would use Machine Learning algorithms to analyze customer data. The AI can identify hidden patterns in buying behavior and segment the audience. This allows the business to run highly targeted social media ad campaigns, ensuring that promotional budgets are only spent on the demographics most likely to convert into paying customers.

    Previous topic 1
    Artificial Intelligence SEM7 2025
    Next topic 3
    Artificial Intelligence 2022

    Past Papers

    Open this section to load past papers

    Click on Show Past Papers to see past papers.
    On This Page
      Reading Stats
      Est. reading time11 min
      Word count1,824
      Code examples0
      DifficultyIntermediate