How do data warehouses, data mining, OLAP and expert systems extend a decision support system?
Describe data warehousing, data mining, online analytical processing, expert systems and group decision support systems, and how they analyse stored data to support decisions
A focused answer to the HSC Information Processes and Technology option dot point on data warehousing, data mining, OLAP, expert systems and group DSS. How each analyses data to support decisions, with the traps markers look for.
Reviewed by: AI editorial process; not yet individually human-reviewed
Have a quick question? Jump to the Q&A page
What this dot point is asking
NESA wants you to describe the technologies that extend a basic decision support system (DSS): the data warehouse that feeds it, the data mining and online analytical processing that find patterns in that data, the expert system that captures human reasoning, and the group DSS that supports collective decisions. This builds on the DSS and models dot point by going into the analysing tools.
The answer
Data warehousing
A data warehouse is a large, integrated store of historical data collected from an organisation's transaction systems and external sources, organised for analysis. Unlike an operational database tuned for fast daily transactions, the warehouse is structured to support querying across long time spans and many dimensions. It is the data foundation a serious DSS analyses, holding cleaned, consistent data that decisions can rely on.
Online analytical processing
Online analytical processing (OLAP) lets a user explore warehouse data interactively across multiple dimensions. Data is viewed as a cube whose dimensions might be time, product, region and customer. The user can drill down from year to quarter to month, roll up from store to region to country, and slice the cube to isolate one product line. OLAP answers planned business questions quickly, such as how sales of a product varied by region over three years.
Data mining
Data mining searches large data stores automatically for patterns, trends and relationships that no one specifically asked for. Where OLAP answers questions the user poses, data mining discovers things the user did not know to ask, such as which products are frequently bought together, which customers are likely to leave, or which transactions look fraudulent. It uses statistical and machine learning techniques to surface associations, clusters and predictions from the data.
Expert systems
An expert system is a DSS that captures the knowledge of human experts in a narrow field and applies it to advise or decide. It has a knowledge base of facts and IF-THEN rules elicited from experts, and an inference engine that chains the rules against the facts of a case to reach a conclusion, often with an explanation of its reasoning. Examples include systems that help diagnose faults or assess loan applications. Unlike a general DSS that supports a person's judgement, an expert system attempts to reproduce the expert's reasoning itself.
Group decision support systems
A group decision support system (GDSS) helps several people make a decision together, especially when they are not in the same place. It provides shared access to models and data, and tools for brainstorming, ranking options and anonymous voting, so a group can pool information and converge on a choice. It addresses the coordination and communication problems that arise when many stakeholders must decide jointly.
How they fit the DSS picture
These technologies are layers of analysis on stored data. The warehouse stores it, OLAP and data mining analyse it for patterns, expert systems reason over it with captured human knowledge, and a GDSS lets a team apply all of this collectively. Each strengthens the analysing information process that is at the heart of decision support.
Exam-style practice questions
Practice questions written in the style of NESA exam questions on this dot point, with worked answer explainers. The year tag is the paper they imitate, not the source.
2019 HSC6 marksA bank is launching new financial products and wants to analyse customers' banking history to determine which products to offer them. Explain how the use of a data warehouse and data mining could assist with decision making in this situation.Show worked answer →
For 6 marks define both and explain how they work together for the bank.
Data warehouse: a large central store that integrates the bank's historical customer and transaction data from across its systems, organised for analysis rather than daily operations. It gives the bank a single, consistent, long-term view of each customer's banking history.
Data mining: software that searches the warehoused data for patterns, trends and relationships that are not obvious. Techniques such as clustering and association find groups of similar customers and links between behaviours and product take-up.
How they assist the decision:
The warehouse supplies clean, integrated historical data to mine.
Mining identifies which customer profiles (income, spending, savings behaviour) are most likely to want each product - for example flagging customers likely to want a credit card or loan.
The bank can then target the right product to the right customer, improving marketing success and reducing wasted offers.
Markers reward correct definitions of both and a clear chain from stored history, to discovered patterns, to a targeted offering decision.
2020 HSC4 marksThe diagram shows the main components of an expert system. Using the information provided on the diagram, describe how an expert system works.Show worked answer →
For 4 marks describe the main components and how they interact to produce advice. The standard components are:
Knowledge base. Stores the domain expertise as facts and "IF...THEN" rules gathered from human experts.
Inference engine. The reasoning component that applies the rules in the knowledge base to the facts entered, to draw conclusions.
User interface. Lets the user enter information about their problem and presents the system's questions, conclusions and explanations.
Explanation mechanism. Justifies how the system reached its conclusion, so the user can trust the advice.
How it works: the user enters details through the interface; the inference engine matches these against the rules in the knowledge base; it chains the matching rules to reach a conclusion and outputs the recommendation, with the explanation facility showing the reasoning. Markers reward naming the components and showing how they interact to give advice.
2022 HSC3 marksDescribe the role of a database of facts in an expert system.Show worked answer →
For 3 marks describe what the database of facts (part of the knowledge base) holds and what it does.
It stores the factual knowledge of the domain - the established information and case data the expert system reasons about (for example symptoms, values, known conditions).
Together with the rule base, it forms the knowledge base. The inference engine compares the facts entered by the user against these stored facts and rules to draw conclusions.
It can be updated as new knowledge becomes available, so the expert system's advice stays current and accurate.
In short, the database of facts supplies the raw domain knowledge the inference engine applies its rules to, making expert reasoning possible. Markers reward identifying it as the stored domain facts used with the rules by the inference engine.