30-minute practice mock for Chapter 9 Artificial Intelligence and Neural Networks (ACtE09): 30 MCQs (18 conceptual, 12 numerical/code-output), five from each sub-topic - introduction to AI and intelligent agents (AI perspectives and history, applications, the Turing test, agent structure, PEAS, rationality, agent types and environment properties), problem solving and searching (state-space formulation, uninformed BFS, DFS, depth-limited, iterative deepening and uniform-cost search, greedy best-first, A* and admissible heuristics, hill climbing and simulated annealing, constraint satisfaction, minimax and alpha-beta pruning), knowledge representation (representation approaches and issues, semantic nets and frames, propositional logic, resolution and inference, predicate logic, unification, forward and backward chaining, Bayes' rule and Bayesian networks), expert systems and NLP (expert system architecture and development, MYCIN-style certainty factors, NLP terminology and stages, edit distance, n-grams and TF-IDF, machine vision stages and robotics), machine learning (supervised, unsupervised and reinforcement learning, entropy, information gain and decision trees, Naive Bayes, evaluation metrics, k-NN, k-means, Q-learning, fuzzy sets, fuzzy inference and defuzzification, genetic algorithm encoding, selection, crossover, mutation and parameters) and neural networks (biological vs artificial neurons, McCulloch-Pitts neuron, activation functions and architectures, perceptron learning, learning rate, gradient descent and the delta rule, Hebbian learning and Adaline, multilayer perceptrons and backpropagation, Hopfield networks). Search, inference, probability and learning answers were computed by running the algorithms, and every code-output answer was confirmed by running the Python code.