What is an expert system? Ane: i : ee system is a computer program that uses arlfcial intelligence (Al) technologies to simulate the judgment and behavior of a human or an organization that has expertise and experience in a particular field Expert systems are usually intended to complement, net replace, human experts. ‘The concept of expert systems was developed in the 1970s by computer scientist Edward Feigenbaum, a computer science professor at Stanford University and founder of Stanford's Knowledge Systems Laboratory. The ‘world was moving from data processing to "knowledge processing, Feigenbaum said in a 1988 manuscript. That meant computers had the potential to do more than basic calculations and were capable of solving complex problems thanks to new processor technology and computer architectures, he explained How does an expert system work? Modem ex, st knowledge systems use machine learning and arti intelligence fv simulate the behavior or judgment of domain experts. ThesePd awit an improve tholr performance Just ashumansdo, 6 they gain more spot ne and for applying cu imulate experience and facte in 3 facts ina da gine ~ a set of rules export systom: integrate them with an inference or rul the knowledge uations pro 190 base to situations provided to the program The inference engine uses or jon from erence engine ni 1@ of two methods for acquiring information fr the knowledge base: 1. For ini : ward chaining reads and processes a set of facts to make & \g/cal prediction about what will happen next. An example of forward chaining would be making predictions about the movement of the stock market. wads and processes a set of facts to reach a omething happened. An example 2. Backward chaining re: examining a set of symptoms tO logical conclusion about why s of backward chaining would be reach a medical diagnosis. con having a good knowledge base. Experts add ise the system to solve An expert system relies information to the knowledge base, and nonexperts u uire @ human expert. complex problems that would usually req yystem is called knowledae of building and maintaining an expert s} rt systems have all the Knowledge engineers ensure that exper problem. They use various knowledge +h as symbolic patterns, to do this. The g the knowledge base or The process engineering. necessary information to solve a representation methodologies, suc! system's capabilities can be enhanced by expandin: creating new sets of rules. What are the components of an expert system? n expert system: There are five main components of al/ tions and use case: yc S of expert systems per" tems can be effe - oF predictions, nese systems have °F Played a large role in many industries, including the following: + Financial servi Financia services, ver ey mate declon aout ase fransgemen ct srab-adior and mokeprditons about the various markets and other financial indicators. eo . lechanical engineering, where they troubleshoot complex electromechanical machinery. Telecommunications, were they are used to make decisions about network technologies used and maintenance of existing networks. Healthcare, where they assist with medical diagnoses. — ‘Agriculture, where they forecast crop damage, And other predictions Customer service, where they help schedule orders, route customer requests and solve problems. where they contribute in a range of areas, including control, highway design, bus and Transportation, .d aviation flight patterns and air pavement conditions, traffic light train scheduling and maintenance, an traffic control. ed to deliver legal «Law, where automation is starting to be us services, and to make civil case evaluations and assess product liability. What are some examples of expert systems?diseases. + MYCIN ide nti nuh [es bacteria such as bacteremia and meningitis, and 'ds antibiotics and dosages. “se + PXDES determi er it : ines the type and severity of lung cancer a person R1Uxt ® ‘CON is an early manufacturing expert system that automatically ‘elects and orders computer components based on customer ‘specifications. What are the advantages of expert systems? Expert systems have several benefits over the use of human experts’ + Accuracy. Expert systems are not prone to human error or emotional influence. They make decisions based on defined rules and facts. Permanence. Human experts eventually leave their role, and a lot of specific knowledge may go with them. Knowledge-based systems provide a permanent repository for knowledge and information. Logical deduction. Expert systems draw conclusions from existing facts using various types of rules, such as if-then tules. « Cost control. Expert systems are relatively inexpensive compared to the cost of employing human experts. They can help reach decisions more efficiently, which saves time and cuts costs.Pert er s. NOWled, 7 Multiple exports contribute to an oxport ayator's . ee. This Provides moro knowledge to draw from and ©xpert from skowing the decision-making. wnat are the cha AMONG expert, lenges of expert systems? systems’ shortcomings are the following: Linear thinking. Expert systems lack true proble One of the advantages of human intelligence is that it can reason in Nonlinear ways and use ancillary information to draw conclusions. Lack of intuition. Human intuition enables people to use common, ‘sense and gut feelings to solve problems, Machines don't have intuition. And emulating gut-feeling decision-making using mechanical logic could take much longer than an expert using intrinsic heuristic knowledge to come to a quick conclusion. Lack of emotion. In some cases ~ medical diagnoses, for example - ~ human emotion is useful and necessary. For example, the disclosure of sensitive medical information to a patient requires emotional intelligence that an expert system may not have. Points of failure. Expert systems are only as good as the quality of their knowledge base. If they are supplied with inaccurate information, it can compromise their decisions Conventional System vs. Expert Systeme crucial difference is the method between expert systems and conventional problem-solving softwar sations, data structures and programs are both used to encode problem expertise. All problem- notin programs, in the expert related expertise is represented in data structures alone, used to encode the problem-related expertise. In traditional applic method.‘Plt 8 specie soliton t "109 nobler. Thexe ystems sim by symbolic computations using the symbors ‘oF networks, frames, serps, ete. Nevertcess, witonal systems, They o I systems. They oversimplify the isues and cannot Human Expert vs. Expert System ‘The significant distin ignificant distinction between expert systems in artificial intelligence and human experts is that expert systems process knowledge represented inthe form of rules and use representational reasoning. limited area, whereas human experts use knowledge inthe form of heuristics of rules of thumb to solve problems in a limited domain. Expert systems are always accessible, unlike human experts, who ate onl , unlike human experts, who are only available daring certain hours ofthe dy. Exper systems handle any problem ina conse amount of time, but human Characteristics of Expert system: D symbolic logic > An explicit knowledge base understandable to professionals of the field P Ability to explain its conclusions that are meaningful to its user Expert systems are used in two different ways: > Decision support: Reminding information or options to an experienced decision maker. Commonly used in medicine. P Decision making: Allowing an unqualified person to make a decision beyond his or herlovol or tr in industria it! Of exportiso, Commonly used in toms, Purpose for 80 fOr the Ci Sys {ho Construction of Export Syst port Systoms: > Recording and di ‘and distributing scarce expe tet distributing scarce export Applying the expert knovwled locations etoremee Ensuring the quality of problem solving D Training experts out of ordinary people Ability of Expert Systems: P Recognizing probloms D> Recognizing solutions > Explaining the choice of solution > Selecting applicable solutions D> Dealing with incomplete information > Restructuring problems D> Reducing the need for research Common Nature of Expert System's Tasks: > When experts out perform non experts D> When the task requires reasoning and knowledge instead of intuition or reflex > When the task could be done either in minutes or hours > When the task could be encoded D When the task is commonly instructed to beginners of the field Expert System Reasoning: ining: Data oriented > Forward Ch Existing facts matched to rule antecedents Matching rules result in consequent: conctsions tinted > Backward Chaining: Goal oriente tt goal or conclusion; match to rule consequentsPP Checking for match between rule- antecedents and facts Repeat until conclusion matches fact > Conflict Resolution: Select rule to applicate ‘Specificity, simplicity, random Development of Expert Systems Involves: > Knowledge Acquisition > Knowledge Representation > Knowledge Encoding > Expert System Testing > Expert System ImplementationeOCTURE si! RE OF AN EXPERT SYSTEM. The structure of the methods of nee S7SteMS reflect the knowledge engineers understanding of Making teske wa psenting knowledge and of how to perform intligent Sesion involve {SKS with the support of a computer based System. Commier denen compi inticate combination of factual and heuristic tnowedge Te dee he he Komputer to be able to reteve and effectively use houses, Kacetedge, we nowledge must be organized in an easily accesible format that disireeishes among data, knowledge and contol structures. For this reason expert eters ae organized in three distinct level ike (a) Knowledge Base: It consists of problem solving rules, procedures and. intrinsic data relevant to the problem domain. The knowledge base constitutes the problem solving rules, facts or intuition that a human expert might use in solving problems in a given problem domain. The knowledge base is usually stored in terms of if-then rules. The working memory represents relevant data for the current problem being solved, b) Working Memory: It refers to task specific data for the problem under consideration. This is the dynamic module of the system. It consists of an essential component called database. In general, the workspace contains a set called rule base, i.e. it contains a set of rules that to be used by a system at a given moment. ©) Inference Engine: This is a generic control mechanism that applies the axiomatic knowledge in the knowledge base to the task specific data to arrive at some solution or conclusion. Inference in production systems is accomplished by « process of chaining through the rules recursively, either in a forward or in a backward direction until a conclusion is reached. These three pieces may very well come from different sources. The inference engit such as VP-Expert, may come from a commercial vendor. The knowledge baser cific diag ic Ba ignostic knowl wledge base compiled by a consulting firm, and the fre icin data m: popert yates, Oe repelled by the end user. A knowledge base is the nucleus of jcture. A knowledge base is created by knowledge engineers, es. These & fp translate the fe the knowledge of real human experts into rules and strategi Wailing problem scenario, THE the capability t recommend further investigation into afeas apparent to the user. ThE riles and strategies ca {IGS fa Stateses can change depending on the Bre knowledge base provides the expert system with direetions for use inquiry. The system also istgses may be important to a certain line of reasoning but no general structure of an expert system is given in figure - Knowledge Engineers Knowledge Base Data Databases Hardware Figure Expert Systems Organization ‘and Operating EnvironmentFD mmm Macnee learning Tieden strategie? Repceantation of Essciesae modularity of an expert system is an important dstineyisiing characteristics vrompared toa conventional computer program. Modularity is affected in an expert Syotem by the use of three distinct components as sow 1 fig 6.2. A good expert system is expected to grow as it fears from user feedback. Feedback is incorporated syeethe knowledge base as appropriate to make the expert sysc™ smarter. The dynamism of the application environment for expe Ser is based on the aareiiual dynamism ofthe components. This ean be classified into three categories as follows. fan expert system is always the 1g memory, sometimes called the sequently, itis the of course that it is ‘The most dynamic part o The content of the workin th each problem situation. Con ‘an expert system assuming, a) Most dynamic: working memory. data structure, changes wit ‘most dynamie component of kept current.a Moderately dynami The lie: This part in the expert system Ua ints change hs sion ke selina ledge base should be carefully evalusted before being in effect, chan, plemented. In sullation experience, es should not be based on just one cons Least dynamic: The least dynamic partis the inference engine. As the control and coding structure of an inference engine is very strict, so changes are made only if absolutely necessary to correct a bug or enhance the inferential process. Commercial inference engines, in particular, change only at the discretion of the developer. Since frequent updates can be disruptive and costly to clients, most commercial software developers try to minimize the frequency of updates. oserition otnowesse Nicos FIGURE 1-1 Major parts of an expert system. Arrows indicate information flow.A style of organization for data base:"Token Stream”. Here the data base is a linear stream of tokens, accessible only in sequence. Each production in turn is matched against the beginning of the stream. For example, if the first character of a production and the first character of the stream differ, the whole match fails. If the rule is invoked, it may add, subtract, or modify characters in the matched segment. Producten eet ec XY AGF. W 0A xz Aco 4 WY eaGURE?I Production ne and delon tre epee toes of a simple system chat replaces sequences of thice sym bots tn the data bave wlth sequences of wo ater» The above diagram shows that production rule and decision treesentation, rere data basa 2 Simple irre Fata base with omen that laces ser NO others, Wwonee OF three symbols Expert s) ystems should be Expert systems are distinct in representation, as formation systems process information, while ex ‘systems attempt to process knowledge. 7 Knowledge in an expert system may originate from many sources, such as textbooks, reports, datat bases, case studies, empirical data, and personal ‘experience. The dominant source of knowledge in today's expert systems is the dominant expert. A knowledge engineer usually obtains knowledge through direct interaction with the expert, as shown in the figure Pelov. ‘cr oe elutes it nt NGURE 1) Typical Trowkedge acquisition process (oF building an expe! system. —— —a inlended to complement, not replace, human Potential to do more than basic calculating ‘and were capable of solving complex problems thanks to new processor technology and computer architectures, he explained How does an expert system work? Modern expert knowledge systems use machine learnir intelligence to simulate the behavior or judgment of dor Systems can improve their performance over time as they gain more main experts. These experience, just as humans do. Expert systems accumulate experience and facts in a knowledge base and integrate them with an inference or rules engine --a set of rules for applying the knowledge base to situations provided to the program. The inference engine uses one of two methods for acquiring information from the knowledge base: