HSC Enterprise Computing Intelligent Systems: exam question types and certainty factors

What this guide covers

Intelligent Systems asks you to explain how expert systems, decision support systems and smart devices reason and act, to calculate with certainty factors, and to judge their effects on work and society. It mixes precise calculation with extended judgement, so it rewards a clear method. Do the practice quiz first, then use this guide to target what you missed.

Question types you should expect

1. Trace the reasoning of an expert system

You may be given facts and IF-THEN rules and asked which conclusion is reached and how. Identify the reasoning direction first: from facts to conclusion (forward chaining) or from a goal back to facts (backward chaining). Then fire the rules one at a time and show each step. Know the three components: the knowledge base (facts and rules), the inference engine (applies the rules) and the user interface.

Revise expert systems and inference engines and evolution of expert systems and intelligent agents.

2. Calculate a certainty factor

Use the method on the dot point page and show every line.

Revise building an expert system with rules and certainty factors.

3. Classify a decision and the support a DSS gives

Structured decisions follow fixed rules and can be automated (reorder stock at a set level). Semi-structured decisions follow some procedure but need judgement at points (setting a loan limit within policy). Unstructured decisions have no set procedure (entering a new market). Say what the DSS contributes (data, models, what-if analysis) and what the human still decides.

Revise decision support systems and automated smart systems and assessing DSS output.

4. Identify components of a smart or IoT system

Map the system as input, processing, output and communication: sensors and biometric inputs collect data, a microcontroller or processor decides, actuators, motors and displays act, and gateways and links move data to storage and the cloud. Edge processing keeps fast decisions local. Computational thinking terms (decomposition, pattern recognition, abstraction, algorithms) are often tested as "which skill is shown".

Revise intelligent system hardware and computational thinking and IoT networks and AI efficiency.

5. Evaluate an effect or an ethical issue

Extended short answers ask you to weigh benefits against harms, for example AI in hiring or surveillance through loyalty schemes. Name the principle at stake (fairness, transparency, accountability, privacy, safety, human oversight), connect it to the scenario, and finish with a judgement and a safeguard.

Revise disruptive effects and ethics of intelligent systems and simulation, modelling, automation and surveillance.

Worked example

Question (exam style, 3 marks). Rule 1: IF fever AND rash THEN measles (CF 0.7). A patient has fever (CF 0.9) and rash (CF 0.8). Rule 2 independently supports measles with CF 0.4. Calculate the combined certainty that the patient has measles.

Answer. Rule 1 uses the lower evidence value: $0.7 \times 0.8 = 0.56$. Combine with Rule 2: $0.56 + 0.4(1 - 0.56) = 0.56 + 0.176 = 0.736$, so the certainty is about 0.74.

Keep going

Continue with the Data Science practice and the Data Visualisation and enterprise project practice.