At present the use of mobile device technology has developed rapidly and the community. Most people use it not just for the sake of communicating, but also to obtain information quickly and efficiently with Internet-oriented applications via WAP technology. Development of Artificial Intelligence technology that occurred has enabled Expert System for applied use in mobile devices with WML and PHP. One of them in the provision of information about various health problems, especially lung health problems of children. The method used is an expert system forward and backward chaining with tree creation of supporting data. With the facilities provided for users and administrators, allowing both users and administrators to use this system according to their respective needs. Users are given the ease in knowing information on various types of lung diseases of children with clinical symptoms, lung hospital information in some areas and consultation with a doctor as a child lung through several questions to be answered user to know the diagnosis. While the administrators facilitated in management systems, both processes added, as well as the latest data updates. This final project is expected to provide information on all matters relating to child lung health problems quickly and efficiently in a lead between the user and the system but remains optimal even in small devices. In general, expert systems (expert systems) are systems that try to adopt human knowledge to computer, so that the computer can solve such problems is usually done by experts / specialists. With this expert system, even a layman can solve the problem or just looking for a real quality information that can only be obtained with the help of experts in their fields. This expert system will also be able to help the activities of the expert as an assistant who is experienced and has an assistant who is experienced and has the knowledge required. In preparation, expert system combines the rules of drawing conclusions (inference rules) with a specific knowledge base provided by one or more experts in a particular field. The combination of these two things are stored in a computer, which then is used in decision-making process for solving a particular problem. Expert system is a computer-based system that uses knowledge, facts and reasoning techniques of the human as an expert on stored in the computer, and used to solve problems that typically require specific expert (Martin and Oxman, 1998). A good expert system can solve the problem with more complete, comparable to an expert who has knowledge in specific areas. Expert systems are programs "artificial Intelligence" ("artificial intelligence") that combines the knowledge base by the inference engine. This is part of high level specialized software or high-level programming language (High-level Language), which seeks to duplicate the functions of an expert in one area of expertise. This program acts as a smart consultant or adviser in a certain skill environment, as a result of the set of knowledge that has been gathered from several experts. Expert system with the correct design and a number of components that work together to form a unity integration, expert systems can be used as a tool to support the activities named as assistant Experienced. Some definitions of expert systems:
* According to Durkin: An expert system is a computer program designed to model the problem solving skills performed by an expert.
* According Ignizio: An expert system is a model and procedures relating, in a particular domain, which is the level of expertise comparable to the expertise of an expert.
* According to Giarratano and Riley: An expert system is a computer system that could match or mimic the ability of an expert.
* An expert system, which first appeared is a General Purpose Problem Solver (GPS) developed by Newel and Simon.
Example - example of an expert system:
* MYCIN is useful for diagnosing diseases
* DENDRAL Identifying the molecular structure of an unknown mixture
* XCON & XSEL Helps large computer system configuration.
* SOPHIE electronic circuit analysis
* Prospector Used in geology to help find and discover deposits
* Folio Helps give a decision for a manager in the case of stock brokers and investment
* DELTA Maintenance of diesel electric locomotives.
Categories and Expert System Problem Area:
* Interpretation, is to make conclusions or descriptions from a collection of raw data.
* Prediction, is projecting the possible consequences of certain situations.
* Diagnosis, is to determine the cause malfunctions in the situation
based on the observed symptoms.
* Design, is to determine the configuration of system components that match the specific performance goals that meet certain constraints.
* Planning, is planning a series of measures that can achieve a number of goals with specific initial conditions.
* Debugging and Repair, is decisive and menginterpresentasikan ways to cope with malfunctions.
* Instruction, is detecting and correcting deficiencies in the understanding of the subject domain.
* Control, is to regulate the behavior of an environment which
complex.
* Selection, is to identify the best option from a set
possibilities.
* Simulation, is the modeling of interactions between system components.
* Monitoring, was to compare the observations with the expected conditions.
The purpose of the expert system is to transfer expertise owned by an expert into the computer and then can be used by other people who are not experts. The advantages of ES:
* Lay people can use it
* Preserving the expertise of a specialist
* Able to operate in hazardous environments
* The ability to access knowledge
* Can work in incomplete information
* Media complementary in research
* Saves time in making a decision
* The process automatically
* Expertise with an expert
* Productivity
Lack of expert system
* The cost is very expensive to create and maintain
* Difficult developed because of limited availability of expertise and expert
* An expert system is not 100% true
Type - The type of knowledge possessed An Expert:
* The theories of the problem
* Rules and procedures are based on problem areas
* Rules to be done in the situation
* Global strategy to solve various kinds of problems
* Meta-knowledge
* The facts
Expert System Characteristics:
* Have the ability to learn or understand the problem from experience.
* Provide prompt and satisfactory response to the new situation.
* Able to handle complex problems (semi-structured).
* Solving problems by reasoning.
o Using knowledge to solve problems.
- Typical features of Expert System:
* Having reliable information.
* Easily modified.
* Heuristics in using the knowledge (which often do not
perfect) to get its completion.
* Can be used in various types of computers.
* Has the ability to adapt.
The basic concept of the expert system include some fundamental problems, among others, who called the expert, what is meant by expertise, how expertise can be transferred, and how the system works. Experts are people who have the knowledge, judgments, experience, special methods, as well as the ability to apply the talent in giving advice and solving problems. Experts usually have several common concepts. First, it must be able to solve problems and achieve the performance levels significantly better than the average person. Second, an expert is relative. Experts at one time or one region may not be an expert in time or another region. For example, medical students may be referred to experts in the disease compared to administration officials, but not an expert at a leading hospital. Usually, human experts are able to do the following: Identify and formulate problems, solving problems quickly and accurately, Explaining the solution, Learning from the experience, knowledge Rearranging, Dividing up the rules if necessary, enact relevant expertise is extensive knowledge specific to the task owned expert. Human expert-knowledge system has improved: that they can analyze his own knowledge and its uses, learn from them, and increase it for future consultation. Similarly, evaluation is necessary in learning the computer so the program can analyze the reasons for success or failure. This can lead to increased knowledge base so as to produce a more accurate and more effective consideration. These components are not available in commercial expert system at the moment, but is being developed in experimental ES at several universities and research institutions. Expert system (Expert System) is an information system that has artificial intelligence (Artificial Intelegent) that resembles human intelligence. Expert systems are similar to the DSS that is aimed at providing support for high-level problem solving for users. Differences ES ES and DSS is the ability to explain the flow of reasoning in reaching a specific solution. Very often solutions that explanation was more valuable than the solution itself. Expertise is often achieved from training, reading, and practice. Expertise includes explicit knowledge, such as the theory is learned from textbooks or the classroom, and implicit knowledge gained from experience. Development of expert system is divided into two generations. Most first-generation expert system using a rule if it is to represent and store knowledge. Second generation expert system is much more flexible in adopting many methods of knowledge representation and consideration. The transfer of expertise from experts to electronic media such as computer and then transferred again to the people who are not experts, is the main aim of the expert system. This process requires four activities: additional knowledge (from experts or other sources), knowledge representation (to computer), inference knowledge, and transfer of knowledge to the user. Knowledge that is stored on a computer called a knowledge base, namely: the facts and procedures (usually in the form of rules). One feature that should be owned by an expert system is the ability to reason. If the skills are stored as knowledge base and available programs that can access the database, then the computer must be programmed to make inferences. This inference process is packaged in a motor inference (inference engine). And each sub-system has the properties of the system to run a particular system function and influence the overall system. There are several reasons for a company to adopt expert system. First, an expert in a company / institution could retire, quit, or have died. Secondly, knowledge needs to be documented or analyzed. Third, education and training is important but it is a difficult task. Expert system knowledge is transferred more easily allows a lower cost. System explanation facility is an additional component of an expert system which functions to deliver information to the user why a question asked by an expert system, what conclusions can be obtained, why a particular solution is rejected, and what the plan is to reach a solution. Inference is a process to generate information from known facts. Inference is a logical solution or implications based on the information available. In an expert system, inference process carried out in a module called the inference engine. When representation pengetahaun in the knowledge base is complete, or at least have been on a level that is quite accurate, then the representation of knowledge is ready for use. Inference engine is a module that contains the program about how to control the reasoning process. There are two important methods of inference in an expert system, namely trace forward (forward chaining) and trace back (backward chaining). "Brain" ES is the inference engine, known also as the control structure or translator rules (in rule-based ES). This component is actually a computer program that provides a methodology to consider the information in the knowledge and the workplace, and formulate conclusions. Inference engine is stored in the skills required in a knowledge base (knowledge base), the computer is programmed so as to produce a solution. There are two ways (methods) the mechanism for inference in rule-based expert system, namely: Inferencing with Rule: Forward and Backward Chaining Inference with the rules is the implementation of component mode, which is reflected in the mechanism of search (the search). Can also check all the rules in knowledge base in a direction forward or backward. Search process continues until no rule that can be used or to a destination (goal) is reached. There are two methods of inferencing with the rules, ie forward chaining or data-driven and backward chaining or goal-driven. Trace forward (forward chaining) trace forward is the rules are tested one by one in the sequence (data driven). Forward chaining is a group of multiple inference that a search of a problem to the solution. If the premise clauses in accordance with the situation (TRUE), then the process will be to assert the solution. Forward chaining is data-driven because the inference starts with the information available and newly obtained solutions. If an application produces a wide tree and not in, then use the forward chaining. Trace backward (backward chaining) reasoning trace back is started from the conclusion and will be verified (goal driven). Using a goal-driven approach, starting from the desired expectations of what happens (the hypothesis), then check on the causes that support (or contradictory) of these expectations. If an application tree produces a narrow and deep enough, then use backward chaining. Both of the above is influenced by the kinds of searches that consists of 3 kinds / search techniques:
* Depth first search, search techniques of code into the code to move down to the level in the sequence.
* Breadth first search, search techniques on all code in one level before moving onto the level below it.
* Best first search, the combination of depth first search and breadth first search.
Acquisition of Knowledge (Knowledge Acquisition) Acquisition of knowledge is the collection of data from an expert into a system (computer program). Material knowledge can be obtained through books, scientific journals, literature, an expert, internet browsing, and other reports. Sources of knowledge from books, scientific journals, literature, an expert, browsing the Internet, the report used as documentation to be studied, processed and collected with a structured knowledge base (knowledge base). The sources of knowledge acquired in order to produce good data that it needs to be processed with good skills as well so as to produce an efficient solution. Because of the ability of the things that principal / mandatory system is needed by a developer. Gaining knowledge from experts is a complex task that often cause bottlenecks in the construction of the ES. In building a large system, one requires a knowledge engineer or expert elicitation knowledge to interact with one or more human experts in building the knowledge base. Usually the knowledge engineer to help experts formulate the problem area by interpreting and integrating human answers, arrange analogy, filed a comparative example, and explain the conceptual difficulties. Expert System Structure Expert systems are composed by two main parts, namely the development environment (development environment) and environmental consultation (consultation environment) (Turban, 1995). Expert system development environment is used to incorporate expert knowledge into an expert system environment, while environmental consultancy used by users who are not experts in order to acquire expert knowledge. The three main components that seem to virtually every expert system is a knowledge base, inference engine and user interface.
1. 1. Knowledge Base (Knowledge Base)
Knowledge base is the core of an expert system, namely the representation of expert knowledge. Knowledge base consists of facts and rules. The fact is the information about objects, events, or situations. The rule is a way to generate a new fact of facts already known. Base includes two basic elements:
* The fact, such a situation the problem and the theory of problem areas, and
* Heuristic or special rules that directs the use of knowledge to solve specific problems in a particular domain.
In addition, the inference engine can include general-purpose problem-solving and decision rule). Heuristic states peniliaian informal knowledge in the areas of application. Knowledge, not only the facts, is the primary raw material in an expert system.
1. 2. Machine Inference (Inference Engine)
Inference engine acts as the brain of an expert system. Inference engine serves to guide the reasoning process of a condition, based on the available knowledge base. Inside there is a process of inference engine to manipulate and direct rule, model, and the fact stored in the knowledge base in order to reach a solution or conclusion. In the process, the inference engine using reasoning strategies and control strategies. Strategy reasoning consists of reasoning strategy plan (Exact Reasoning) and uncertain reasoning strategy (Inexact Reasoning). Exact reasoning will be done if all the data needed to draw a conclusion is available, while inexact reasoning done on a state sebaliknya.Strategi direction control serves as a guide in conducting the process of reasoning. There are three techniques frequently used controls, ie forward chaining, backward chaining, and a combination of both these control techniques.
1. 3. Database (Data Base)
The database consists of all the necessary facts, which facts were used to fulfill the conditions of the rules in the system. The database stores all the facts, whether the fact early on when the system began operating, and the facts obtained during the process of drawing conclusions being implemented. The database used to store data from observation and other data required for processing.
1. 4. User Interface (User Interface)
User Interface, is the part that allows managers mamasukan instructions and information into and receive information from an expert system. This facility was used as a communication mechanism between the user (users) with the system. User interface (User Interface) to receive information from the user (user) and provide information to the user (users) to help direct the path search problem to find a solution. The user interface, functions to input new knowledge into the knowledge base of expert system (ES), featuring an explanation system and provide overall system usage guide step by step so that the user understands what will be done on a system. Most important in building the user interface is easy to use / run the system, interactive, communicative, while the difficulty in developing / building a program is not too be disclosed. Knowledge Representation Techniques Knowledge representation is a technique to represent the acquired knowledge base into a scheme / diagram so that it can be seen certain relationships / connectedness between the data with other data. This technique helps the knowledge engineer in understanding the structure of knowledge that will make an expert system. There are several knowledge representation techniques commonly used in the development of an expert system, namely
1. a. Rule-Based Knowledge
Knowledge is represented in a form of facts (facts) and rules (rules). The form of this representation consists of premise and conclusion.
1. b. Frame-Based Knowledge
Knowledge is represented in a form of hierarchy or network frames. c. Object-Based Knowledge Knowledge is represented as a network of objects. Object is a data element consisting of data and methods (processes).
1. d. Case-Base Reasoning
Knowledge is represented in the form of conclusions cases (cases).
