UNIT5 Expert Systems Topics 1 Expert Systems 2

UNIT-5 Expert Systems Topics: 1. Expert Systems 2. Architecture of expert system 3. Roles of expert systems 4. Knowledge Acquisition 5. Meta knowledge 6. Typical expert systems. MYCIN, DART, XOON, Expert systems shell

Artificial Intelligence n AI n The ability of computers to duplicate the functions of the human brain 2

Interesting Statistics n It has been estimated that computers that can exhibit humanlike intelligence (including musical and artistic aptitude, creativity, physical movement physically, and emotional responsiveness) require processing power of 20 million billion calculations per second (by the year 2030? ). 3

The Difference Between Natural & Artificial Intelligence Attributes Use Sensors Creativity and Imagination Learn from Experience Human High Machine Low Low Adaptability Access external information Make complex calculations Transfer information High Low Low High 4

The Major Branches of AI(application of AI) 5

Expert Systems (ES) 6

7

Capabilities of Expert System 8

Components of Expert System 9

Components of ES 10

Components of an Expert System 11

Components of an Expert System Knowledge Base Stores all relevant information, data, rules, cases, and relationships used by the expert system. Uses • Rules • If-then Statements • Fuzzy Logic 12

The Knowledge Base n Stores all relevant information, data, rules, cases, and relationships used by the expert system n Assembling human experts n Use of fuzzy logic n A special research area in computer science that allows shades of gray and does not require everything to be simple black/white, yes/no, or true/false n Use of rules n Conditional statement that links given conditions to actions or outcomes n E. g. if-then statements n Use of cases 13

14

Components of an Expert System Inference Engine Seeks information and relationships from the knowledge base and provides answers, predictions, and suggestions the way a human expert would. Uses • Backward Chaining • Forward Chaining 15

The Inference Engine n Seeks information and relationships from the knowledge base and provides answers, predictions, and suggestions the way a human expert would n Forward chaining(Goal driven Reasoning) n Starting with the facts and working forwards to the conclusions n Backward chaining(Data driven Reasoning ) n Starting with conclusions and working backward to the supporting facts 16

The Inference Engine Figure 7. 4: Rules for a Credit Application 17

To recommend a solution, the interface engine uses the following strategies − n Forward Chaining n Backward Chaining 18

19

20

Components of an Expert System Explanation Facility Allows a user to understand how the expert system arrived at certain conclusions or results. For example: it allows a doctor to find out the logic or rationale of the diagnosis made by a medical expert system 21

Components of an Expert System Knowledge acquisition facility Provide convenient and efficient means of capturing and storing all the components of the knowledge base. Acts as an interface between experts and the knowledge base. 22

Components of an Expert System User Interface Specialized user interface software employed for designing, creating, updating, and using expert systems. The main purpose of the user interface is to make the development and use of an expert system easier for users and decision makers 23

Expert system Technology 24

Expert Systems Development Figure 7. 6: Steps in the Expert System Development Process 25

Participants in Expert System Development 26

Participants in Expert System Development n Domain n The area of knowledge addressed by the expert system n Domain Expert n The individual or group who has the expertise or knowledge one is trying to capture in the expert system n Knowledge Engineer n An individual who has training or expertise in the design, development, implementation, and maintenance of an expert system n Knowledge User n The individual or group who uses and benefits from the expert system 27

Application of ES 28

Benefits of Expert System 29

Limitations of an Expert System n Not widely used or tested n Difficult to use n Limited to relatively narrow problems n Possibility of error n Cannot refine its own knowledge n Difficult to maintain 30

Expert System Shells 31

Expert System Shells n The shell is a piece of software which contains the user interface, n a format for declarative knowledge in the knowledge base, and n an inference engine. n n The knowledge engineer uses the shell to build a system for a particular problem domain. “A collection of software packages and tools used to develop expert systems” 32

33

Components of an expert system User Expert system shell User Inter face Explanation system Case specific data: Working storage Inference engine Knowledge base editor 34

Expert System Shells n In the 1980 s, expert system "shells" were introduced and supported the development of expert systems in a wide variety of application areas. n During the work , a large amount of LISP code was written for different modules: n n n . Knowledge base Inference engine Working memory Explanation facility End-user interface

36

MYCIN 37

n MYCIN was an early expert system that used artificial intelligence to identify bacteria causing severe infections. n recommend antibiotics, with the dosage adjusted for patient's body weight n The MYCIN system was also used for the diagnosis of blood clotting diseases. n MYCIN was developed over five or six years in the early 1970 s at Stanford University. n It was written in Lisp 38

n MYCIN was a standalone system that required a user to enter all relevant information about a patient by typing in responses to questions MYCIN posed. n MYCIN operated using a fairly simple inference engine, and a knowledge base of ~600 rules. n It would query the physician running the program via a long series of simple yes/no or textual questions. 39

Tasks and Domain n Disease DIAGNOSIS and Therapy SELECTION n Advice for non-expert physicians with time considerations and incomplete evidence on: Bacterial infections of the blood n Expanded to meningitis and other ailments n Meet time constraints of the medical field n 40

MYCIN Architecture 41

Consultation System n Performs Diagnosis and Therapy Selection n Control Structure reads Static DB (rules) and read/writes to Dynamic DB (patient, context) n Linked to Explanations n Terminal interface to Physician 42

Consultation “Control Structure” n n Goal-directed Backward-chaining Depth-first Tree Search High-level Algorithm: 1. 2. 3. 4. Determine if Patient has significant infection Determine likely identity of significant organisms Decide which drugs are potentially useful Select best drug or coverage of drugs 43

Static Database n Rules n Meta-Rules n Templates n Rule Properties n Context Properties n Fed from Knowledge Acquisition System 44

Dynamic Database n Patient Data n Laboratory Data n Context Tree n Built by Consultation System n Used by Explanation System 45

Explanation System n Provides reasoning why a conclusion has been made, or why a question is being asked n Q-A Module n Reasoning Status Checker 46

DART n DART is a joint project of the Heuristic Programming Project and IBM that explores the application of artificial intelligence techniques to the diagnosis of computer faults. n The primary goal of the DART Project is to develop programs that capture the special design knowledge and diagnostic abilities of these experts and to make them available to field engineers. n The practical goal is the construction of an automated diagnostician capable of pinpointing the functional units responsible for observed malfunctions in arbitrary system configurations. 47

n Dynamic Analysis and Replanning Tool n DART uses intelligent agents to aid decision support system n Give planners the ability to rapidly evaluate plans for logistical feasibility. n DART decreases the cost and time required to implement decisions. n The field engineer is familiar with the diagnostic equipment and software testing. n Access to information about the specific system hardware and software configuration of the installation. 48

Xcon n The R 1 (internally called XCON, for e. Xpert CONfigurer) program was a production rule based system written in OPS 5 by John P. Mc. Dermott of CMU in 1978. n configuration of DEC VAX computer systems n ordering of DEC's VAX computer systems by automatically selecting the computer system components based on the customer's requirements. n XCON first went into use in 1980 in DEC's plant in Salem, New Hampshire. It eventually had about 2500 rules. n By 1986, it had processed 80, 000 orders, and achieved 9598% accuracy. n It was estimated to be saving DEC $25 M a year by reducing the need to give customers free components when technicians made errors, by speeding the assembly process, and by increasing customer satisfaction. 49

n XCON interacted with the sales person, asking critical questions before printing out a coherent and workable system specification/order slip. n XCON's success led DEC to rewrite XCON as XSELa version of XCON intended for use by DEC's salesforce to aid a customer in properly configuring their VAX. 50

XCON: Expert Configurer Stages of Expert System building n Identification: Problems, data, goals, company, people… n Conceptualization: Characterize different kinds of concepts and relations n Formalization: Express character of search n Implementation: Build the system in executable form n Testing and Evaluation: Does it do what we wanted? n Maintenance Adapt to changing environment or requirements Expert Systems 14 51

Phase 1: Identification n DEC, Digital Equipment Corporation n n n n Large computer manufacturer, started 1957 Catalogue has 40. 000 different parts Buyer (with Sales Rep) sends order, typically 100 parts Delivery and assembly by DEC personnel Too often, part collection does not allow installation Too often, installed computer does not meet requirements Remedy: Completely assemble and test system in factory Automate configuration problem; attempts with procedural languages were unsuccessful XS approach started around 1980 Expert Systems 14 52

Phase 2: Conceptualization Con. . what? Expert Systems 14 53

Phase 3: Formalization n Configuration engineers could talk well to Knowledge Engineers of the CSDG n Could explain in what stage which component should be configured how n This was expressed in production rules IF c 1, c 2 c 3 THEN a 1, a 2, a 3 n Configuration stage was explicitly represented as data: current goal or context n Changing contexts moved configuration process through all stages Expert Systems 14 54

Phase 4: Implementation into system R 1 n Language: OPS 5 (similar to CLIPS) n Conflict Resolution: MEA (extends Lex / Specificity) n Means-Ends Analysis: order by recency of first condition IF c 1, c 2 THEN. . is now different from IF c 2, c 1 THEN n Contexts are treated as special by putting them first n End-task is unspecific, thus executed last n Use MEA + Spec to concentrate on subtasks: n IF g 1, x, y THEN assert barify // Signal necessity of subtask n IF barify, a THEN p, q // Two rules perform the task n IF barify, b THEN r, s // of barification per se n IF barify THEN retract barify // Termination when ready Expert Systems 14 55

56

57

58

Important questions PART-B 1. Expert system (ES)? architecture of expert system? (components of Expert system)**** 2. Expert system shell? *** 3. MYCIN? ** 4. DART? 5. XCON? 6. Knowledge acquisition? 7. Inference Engine? Methods? (forward chaining, back ward chaining) 59

PART-A (2 marks) 1. Expert system(ES)? 2. Application of ES? 3. List advantage & disadvantage of ES? 4. List out the Components of ES? 5. Define inference engine? 6. What is knowledge base(KB)? 7. What is the role of expert engineer? 8. What is meant by knowledge acquisition? 9. Expert system shell? 10. MYCIN? 11. DART? 12. XCON? 60