E. INTELLIGENT SYSTEMS AND TECHNIQUES
Artificial intelligence (AI) technology consists of computer-based systems (hardware and software) that attempt to emulate intelligent human behaviour. Such systems are able to learn languages, accomplish physical tasks, and emulate human expertise and decision-making.
While AI systems are limited to very narrow domains they play an important role in modern-day knowledge management.
Artificial intelligence and database technology provide a number of intelligent systems and techniques that organisations can use to capture individual and collective knowledge and to extend their knowledge base. Expert systems, case based reasoning, and fuzzy logic are used for capturing tacit knowledge. Neural networks and data mining are used for knowledge discovery. These can discover underlying patterns, categories, and behaviours in large quantities of data that could not be discovered by managers alone or simply through experience. Genetic algorithms have the ability to search for solutions to problems that are too large and complex for human beings to analyse on their own. Intelligent agents can automate routine tasks to help firms search and filter information for use in electronic commerce, supply chain management and other activities.
Data mining, which is discussed in Chapter 4, helps organisations capture undiscovered knowledge hidden in large databases, providing managers with new insights into problems for improving business performance. Data mining is also an important tool for management decision-making.
The following intelligent systems and techniques are discussed here:
By non-experts to improve their problem-solving capabilities
To propagate scarce knowledge resources for improved consistent results
Where the area of expertise is limited to a narrow area (referred to as the domain)
As a tool to improve productivity and quality
To support strategic decisions
An expert system is a computer program that simulates the judgement and behaviour of a human or an organisation that has expert knowledge and experience in a particular field.
Typically, such a system consists of a knowledge base containing the accumulated experience and a set of rules for applying the knowledge base to each particular situation that is
described to the program. Sophisticated expert systems can be enhanced with additions to the knowledge base or to the set of rules.
While Expert systems (Figure 10.3) lack the breath of knowledge and the understanding of a human expert, they can provide benefits, helping organisations make high quality decisions with fewer people. Today expert systems are used in business in distinct highly structured decision-making situations.
Figure 10.3: A simplified model of an Expert System The Components of an Expert System
An expert system will generally contain the following components:
Knowledge base: The knowledge base contains the knowledge of humans experts based on their experiences and knowledge built up over many years. The knowledge base also requires a set of rules that direct the use of the knowledge to solve specific problems in a particular domain.
Inference engine: The inference engine is a computer that draws inferences from the results of applying the user supplied facts to the rules in the knowledge base. It then proceeds to the next fact-rule combination. The inference engine is considered the ―brain‖ of the system.
User interface: The user interface allows the user to communicate with the system. The system communicates with the user using a question and answer format. This communication drives the inference engine to match the symptoms of the problem with the knowledge in the base so that a conclusion is drawn and a recommendation is made to solve the problem.
Explanation facility: This feature of the expert system gives it the ability to explain its recommendation.
Current Data storage: This is a storage area set aside for input data related to the current problem.
Knowledge engineer: The person who pulls the data from the human expert and fits it into the expert system is called the knowledge engineer.
Benefits of an Expert System
The benefits of an expert system include the following:
Reduced cost and reduced training time User
Improved decision making
Improved quality and services
Improved user and customer satisfaction
Organisational Intelligence: Case-Based Reasoning
Expert systems primarily capture the tacit knowledge of individual experts, but organisations also have collective knowledge and expertise that they have built up over the years. This organisational knowledge can be captured and stored using case-based reasoning systems. In case-based reasoning (CBR), descriptions of past experiences represented as cases, are stored in a database for later retrieval when the user encounters a new case with similar characteristics. The system searches for stored cases with problem characteristics similar to the new one. It finds the closest fit, and applies the solutions of the old case to the new case.
Successful solutions are tagged to the new case and both are stored together with the other cases in the knowledge base. Unsuccessful solutions are also added to the case database along with explanations as to why the solutions did not work (See Figure 10.4).
Expert systems work by applying a set of IF-THEN-ELSE rules against a knowledge base, both of which are extracted from human experts. Case-based reasoning, in contrast, represents knowledge as a series of cases, and this knowledge base is continuously updated by users of the system.
Figure 10.4: A simplified model of a Case-based reasoning system
User inputs problem description
Case Database System searches
database for similar case
System asks user for more information
System finds case that is best match &
Does solution solve problem
System stores problem and solution
Page 143 Fuzzy Logic
Fuzzy logic is a rule-based artificial intelligence technology that handles uncertainty, by mimicking the process of human reasoning and allows computers to handle incomplete or ambiguous data. Fuzzy logic represents more closely the way people actually think than traditional IF-THEN rules. Decision making often involve situations that are neither black nor white. They are grey at best with the term fuzzy often being suitable. Fuzzy logic systems are only starting to be applied to business situations.
Neural networks are systems of programs and data structures that attempt to model the capabilities of the human brain. Neural Networks are an array of interconnected processors operating in parallel in which knowledge is represented by the pattern of interconnections among them and by adjustable weights of these connections. They have good pattern recognition techniques and can identify hidden patterns in data and can also deal with incomplete input. They also have an ability to learn new information and behaviour.
A neural network uses rules it ―learns‖ from patterns in data to construct a hidden layer of logic. The hidden layer then processes more inputs and categorises them based on the experience of the model.
Difference between Neural Networks and Expert Systems
Table 10.5 provides a summary of the differences between neural networks and expert systems
Table 10.5: Summary of the differences between neural networks and expert systems
Expert Systems Neural Networks
Expert systems emulate human decision-making.
Neural networks learn human thought processes and reasoning patterns. selection and genetics. Genetic algorithms are search procedures that can be used to find the optimal solution to a specific problem by searching through a very large number of possible solutions to that problem. Genetic algorithms involve adaptive computation where possible solutions can evolve and can even be combined to form a new population of solutions. As solutions alter and combine, the worst ones are discarded and the better ones survive to go on and produce even better solutions.
Genetic algorithms are particularly suited to the areas of optimisation and search. They are used to solve problems that are complex, changing and usually involve large numbers of variables.
F. MANAGEMENT CHALLENGES AND SOLUTIONS
The difficulties of implementing knowledge management systems include:
Insufficient resources available to structure and update the stored content
Poor quality and high variability of content because of insufficient validation
Document and content stores lack context, making documents difficult to understand
Individual employees are not rewarded for contributing knowledge, and many are resistant to sharing knowledge with others
Search engines return too much information, reflecting lack of knowledge structure or mechanism for tagging documents
Laudon and Laudon, (2010) suggest that for businesses to obtain value for knowledge management systems they should use the following steps:
Develop in stages
Choose a high-value business process
Choose the right audience
Measure return on investment during initial implementation
Use the result of the measurements to establish the organisational wide values.