Understanding RAKT Chatbot Terms

1. **Backpropagation through time (BPTT)** is a variant of which algorithm used for training Recurrent Neural Networks (RNNs)? a) Forward propagation b) **Backpropagation** c) Tokenization d) Vectorization

• Backpropagation

• Gradient Descent

• Reinforcement Learning

• Stochastic Gradient Descent

2. Which of the following **best describes the Bag-of-words model** in natural language processing? a) A model that emphasizes grammar and word order. b) **A model representing text as an unordered collection of words, disregarding grammar but tracking word frequency.** c) An algorithm for correcting biases in datasets. d) A technique for fine-tuning language models.

• A framework for analyzing sentence structure and syntax.

• b) A model representing text as an unordered collection of words, disregarding grammar but tracking word frequency.

• A method for generating word embeddings from text.

• A model that focuses solely on semantic meaning of words.

3. What are **biases** in the context of AI and machine learning? a) Techniques to improve the accuracy of models. b) **Systematic errors in data that can lead to unfair outcomes.** c) Methods for increasing data size. d) Algorithms for optimizing performance.

• Strategies for enhancing model interpretability.

• Systematic errors in data that can lead to unfair outcomes.

• Random errors in data that improve outcomes.

• Techniques for reducing model complexity.

4. In natural language processing, what does **tokenization** refer to? a) The process of generating random text. b) **The process of breaking text into smaller units, such as words or phrases.** c) A method for analyzing sentence structure. d) A technique for summarizing text.

• The technique of identifying the main topic of a text.

• A method for generating keywords from a document.

• The process of breaking text into smaller units, such as words or phrases.

• The process of translating text into another language.

5. What is a **neural network**? a) A type of database. b) **A computational model inspired by the human brain, used for pattern recognition.** c) A programming language. d) A hardware component for data storage.

• A method for data encryption.

• A software tool for web development.

• A type of machine learning algorithm.

• A computational model inspired by the human brain, used for pattern recognition.

6. Which term describes the **process of adjusting the weights of a neural network** during training? a) Initialization b) **Learning** c) Evaluation d) Prediction

• Activation

• Regularization

• Normalization

• Learning

7. What does **overfitting** mean in machine learning? a) The model performs well on training data but poorly on new data. b) **The model learns noise in the training data instead of the actual pattern.** c) The model is too simple to capture the underlying trend. d) The model is trained on too little data.

• The model is overly complex and captures too many patterns.

• b) The model learns noise in the training data instead of the actual pattern.

• The model performs well on both training and new data.

• The model is trained on a diverse dataset.

8. In the context of chatbots, what is **natural language understanding (NLU)**? a) The ability to generate text. b) **The ability to comprehend and interpret human language.** c) A method for translating languages. d) A technique for summarizing conversations.

• A technique for analyzing user emotions.

• A method for creating chatbots.

• The ability to recognize speech patterns.

• The ability to comprehend and interpret human language.

9. What is the purpose of **training data** in machine learning? a) To test the model's performance. b) **To teach the model how to make predictions or decisions.** c) To store the final model. d) To visualize data.

• To collect user feedback on predictions.

• To improve the model's accuracy after deployment.

• To generate random data for testing purposes.

• To teach the model how to make predictions or decisions.

10. What does **reinforcement learning** involve? a) Learning from labeled data. b) **Learning through trial and error, receiving rewards or penalties.** c) Learning from static datasets. d) Learning without any feedback.

• Learning by observing others' actions.

• Learning from historical data without interaction.

• Learning through trial and error, receiving rewards or penalties.

• Learning through structured programming tasks.