Archive for the ‘Information System’ Category
Artificial Neural Network System (ANNS)
An information system developed with a processing and learning features similar to human beings to generate information is called an Artificial Neural Network System (ANNS) (L. J. Landau and J. G. Taylor 1998). ANNS learns by providing examples and by doing mistakes similar to human beings. It means ANNS is an attempt to develop an information system (IS) that resembles the human behavior.
Main Categories to Search Knowledge
These are the main categories to reason knowledge to have an inference:
- Optimizing Search.
- Blind Search.
- Heuristic Search.
State Space Search and Main Approaches to Search Knowledge
State Space Search
Several alternative solutions are considered to determine the best one when you face a problem in every day life. Search is a method to reach a conclusion or goal state after passing or shifting through several alternative ways or paths (Efraim Turban 1999). Any point on the path is called a state and we have several alternatives to move to next state, such as information system has several options for a user once he/she logs on. Search method is also called state space search.
Main Methods to Reason Knowledge (Part Two)
Case or Analogical Reasoning Method
Analogical reasoning method used likeliness or similarities between or among situations to make an inference. Analogical reasoning is used mostly widely for legal, disease and bank cases. Inference is made on the base of assumption that if two situations are similar in some aspects, these are similar in all aspects. The assumption used for the analogical reasoning is logically wrong. There are few advantages to develop Expert Systems based on the analogical reasoning.
Main Methods to Reason Knowledge (Part One)
These are the main methods to reason knowledge to have an inference (Efraim Turban 1999):
- Deductive Reasoning Method.
- Inductive Reasoning Method.
- Case or Analogical Reasoning Method.
- Procedural or Model Based Reasoning Method.
- Predicate Logic Calculus Reasoning Method.
Main Methods of Knowledge Representation (Part Two)
Knowledge Representation Using Frame
Frame is a data structure holding detailed knowledge of a particular object or subject, such as student or vehicle registration record. It is an implementation of an object oriented programming to develop expert systems.
Main Methods of Knowledge Representation (Part One)
The main methods of knowledge representation (Efraim Turban 1999) are as follows:
- Logic.
- Semantic Network.
- Frames.
- Production Rules.
- Data Structure.
- Algorithm.
- Data Flow Diagram (DFD).
- Unified Modeling Language (UML).
Main Areas of ES-Advantages and Disadvantages of ES-Comparison of Conventional vs. Expert System
Main Areas of Expert Systems
Expert System is always developed for a particular domain or area. Main areas of Expert System are as follows (Haag, Cummings, Dawkins 1998):
Methodologies to Develop Expert Systems (Part Three)
Rapid Prototyping
Rapid Prototyping is divided into these main sub phases:
- Develop a Prototype.
- Test the Prototype.
- Gather more Requirements.
Methodologies to Develop Expert Systems (Part Two)
System Analysis and Design
Main phases of System Analysis and Design Phase are as follows:
- Detailed Design.
- Development Strategy.
- Sources of Knowledge.
- Selection of Computer Resources.
- Feasibility Study.
- Cost-Benefit Analysis Sheet.