Harsh RB

CAI Quesitons

5 min read

UNIT–I: Introducing Dialogue Systems

Topics: Introduction, history, modeling, design & development

Long Answer Questions (10 Marks)

  1. Define a dialogue system and explain its components in detail.

  2. Discuss the history and evolution of dialogue systems.

  3. Explain the architecture of a dialogue system with a neat diagram.

  4. Describe different types of dialogue systems with examples.

  5. Explain how conversations are modeled in dialogue systems.

  6. Discuss the challenges in designing dialogue systems.

  7. Explain the process of designing and developing dialogue systems.

  8. Compare task-oriented and open-domain dialogue systems.

  9. Explain the role of NLP in dialogue systems.

  10. Discuss the limitations of early dialogue systems.

Medium Answer Questions (5 Marks)

  1. Explain the basic working of a dialogue system.

  2. Describe the stages involved in conversation modeling.

  3. Write a note on early dialogue systems.

  4. Explain the importance of conversational interfaces.

  5. Discuss the design considerations for dialogue systems.

Short Answer Questions (2 Marks)

  1. Define dialogue system.

  2. What is a conversational agent?

  3. What is NLP?

  4. What is human-computer interaction?

  5. What is a task-oriented dialogue system?

  6. What is an open-domain system?

  7. What is dialogue modeling?

  8. What is turn-taking?

  9. Define utterance.

  10. What is intent?

UNIT–II: Rule-Based Dialogue Systems

Topics: Architecture, tools, rule-based techniques, Alexa Prize

Long Answer Questions (10 Marks)

  1. Explain the architecture of rule-based dialogue systems.

  2. Describe rule-based techniques used in dialogue systems.

  3. Explain how to design a rule-based dialogue system.

  4. Discuss various tools used for developing dialogue systems.

  5. Explain pattern matching techniques in rule-based systems.

  6. Discuss advantages and limitations of rule-based dialogue systems.

  7. Explain the working of a rule-based chatbot with an example.

  8. Write a detailed note on Alexa Prize and its significance.

  9. Compare rule-based and statistical dialogue systems.

  10. Explain dialogue flow design in rule-based systems.

Medium Answer Questions (5 Marks)

  1. Explain rule-based systems with examples.

  2. Write a note on dialogue system architecture.

  3. List tools used for chatbot development.

  4. Explain pattern matching in dialogue systems.

  5. What are templates in rule-based systems?

Short Answer Questions (2 Marks)

  1. What is a rule-based system?

  2. What is pattern matching?

  3. What is dialogue flow?

  4. What is a chatbot?

  5. What is Alexa Prize?

  6. What is intent recognition?

  7. What is slot filling?

  8. What is a response template?

  9. What is dialogue management?

  10. What is a rule engine?

UNIT–III: Statistical Data-Driven Dialogue Systems

Topics: RL, MDP, POMDP, state tracking, policy

Long Answer Questions (10 Marks)

  1. Explain statistical data-driven dialogue systems.

  2. Describe reinforcement learning in dialogue systems.

  3. Explain how dialogue is represented as a Markov Decision Process (MDP).

  4. Discuss Partially Observable Markov Decision Process (POMDP) in dialogue systems.

  5. Explain dialogue state tracking in detail.

  6. Describe dialogue policy and its importance.

  7. Explain the transition from MDP to POMDP.

  8. Discuss challenges of reinforcement learning in dialogue systems.

  9. Compare rule-based and statistical approaches.

  10. Explain components of statistical dialogue systems.

Medium Answer Questions (5 Marks)

  1. Define reinforcement learning.

  2. Explain MDP with example.

  3. What is dialogue state tracking?

  4. Explain dialogue policy.

  5. What are the issues in RL-based systems?

Short Answer Questions (2 Marks)

  1. What is statistical dialogue system?

  2. Define reinforcement learning.

  3. What is MDP?

  4. What is POMDP?

  5. What is state tracking?

  6. What is reward function?

  7. What is policy?

  8. What is exploration vs exploitation?

  9. What is belief state?

  10. What is action selection?

UNIT–IV: Evaluating Dialogue Systems

Topics: Evaluation methods, PARADISE, QoE, interaction quality

Long Answer Questions (10 Marks)

  1. Explain the process of evaluating dialogue systems.

  2. Discuss evaluation techniques for task-oriented dialogue systems.

  3. Explain evaluation of open-domain dialogue systems.

  4. Describe the PARADISE evaluation framework.

  5. Explain Quality of Experience (QoE) in dialogue systems.

  6. Discuss interaction quality metrics.

  7. Compare different evaluation methods for dialogue systems.

  8. Explain automatic vs human evaluation methods.

  9. Discuss challenges in evaluating conversational systems.

  10. What is the best way to evaluate dialogue systems? Justify.

Medium Answer Questions (5 Marks)

  1. What is dialogue evaluation?

  2. Explain QoE.

  3. Write a note on PARADISE.

  4. What are evaluation metrics?

  5. Explain user satisfaction measurement.

Short Answer Questions (2 Marks)

  1. Define evaluation in dialogue systems.

  2. What is PARADISE framework?

  3. What is QoE?

  4. What is interaction quality?

  5. What is user satisfaction?

  6. What is task success rate?

  7. What is response accuracy?

  8. What is latency?

  9. What is human evaluation?

  10. What is automatic evaluation?

UNIT–V: End-to-End Neural Dialogue Systems

Topics: Neural models, retrieval-based, datasets, challenges

Long Answer Questions (10 Marks)

  1. Explain neural network approaches to dialogue systems.

  2. Describe end-to-end neural dialogue systems.

  3. Explain retrieval-based response generation.

  4. Discuss generative vs retrieval-based models.

  5. Explain task-oriented neural dialogue systems.

  6. Describe open-domain neural dialogue systems.

  7. Discuss issues in neural dialogue systems and their solutions.

  8. Explain datasets used in conversational AI.

  9. Discuss challenges in building neural dialogue systems.

  10. Explain recent advancements in conversational AI.

Medium Answer Questions (5 Marks)

  1. What is neural dialogue system?

  2. Explain retrieval-based models.

  3. What is generative model?

  4. Write a note on datasets.

  5. Explain challenges in neural systems.

Short Answer Questions (2 Marks)

  1. What is neural network?

  2. What is deep learning?

  3. What is sequence-to-sequence model?

  4. What is retrieval-based system?

  5. What is generative model?

  6. What is training data?

  7. What is corpus?

  8. What is chatbot dataset?

  9. What is overfitting?

  10. What is evaluation metric?

Most Expected 10-Mark Questions (Important)

  1. Explain architecture of dialogue systems.

  2. Rule-based dialogue systems and their working.

  3. Reinforcement learning in dialogue systems.

  4. MDP and POMDP in conversational AI.

  5. Dialogue state tracking and policy.

  6. Evaluation methods and PARADISE framework.

  7. Neural dialogue systems and architectures.

  8. Retrieval vs generative models.

  9. Open-domain vs task-oriented systems.

  10. Challenges in conversational AI.

Most Expected 2-Mark Questions (Repeated Topics)

  1. Define dialogue system.

  2. What is intent?

  3. What is slot filling?

  4. What is MDP?

  5. What is POMDP?

  6. What is QoE?

  7. What is PARADISE?

  8. What is chatbot?

  9. What is reinforcement learning?

  10. What is neural dialogue system?

Graph View