Exp.docx

EXPERT SYSTEM AND KNOWLEDGE PROCESSING: SURVEY AND ITS RESEARCH AREA Riddhi Sanjay Singh, B.Sc. Computer Science Student Department of Computer Science VES College of Arts, Science and Commerce, Mumbai India

Abstract: There is a huge development in the field of Artificial Intelligence. Expert system is an application of Artificial Intelligence which is having a huge impact on various fields of life. Expert system uses human understanding to solve the complex problems in various areas which includes science, engineering, commercial enterprise, medicine, weather forecasting and the companies employing the technology of expert system has seen an increase in the quality and performance. In artificial intelligence, an expert system is a computer system that emulates the decision-making ability of a human expert. There are different types of expert systems such as rule-based expert system, frame-based expert system, fuzzy logic, neural network, genetic algorithm, etc. Expert systems were among the first truly successful forms of artificial intelligence (AI) software. This paper gives a top-level view of this technology and will discuss a survey on diverse works achieved in distinct regions as agriculture, cars, biomedical, education, remedy, metallic making enterprise the usage of expert system.

I.

INTRODUCTION

The discovery and development of expert systems are recorded since inside the early 1970s after which proliferated within the 1980s till these days. Expert system is an artificial intelligence program that has expert level knowledge about a particular domain and knows how to use its knowledge to respond properly. Domain refers to the area within which the task is being performed. Ideally the expert systems should substitute a human expert. The expert system represents knowledge acquired from human expert as input (data or rules) within the computer. These rules and data can be called upon to solve the complex problems. Expert systems are designed to solve complex problems by reasoning through bodies of knowledge, represented mainly as if-then-rules rather than through conventional procedural code. Expert System has various advantages over the human expert as expert system is affordable, permanent, regular, rapid processing and brief replication whereas human expert is perishable, unpredictable, and highly-priced and has slow processing and reproduction.

Characteristics of expert systems: A. High performance: Performance of an expert system should be at the level of a human expert. B. Adequate response time: Expert system should have the ability to respond in a reasonable amount of time. Time is crucial especially for real time systems. C. Reliability: They must be consistently good in quality/performance and should not crash. D. Understandable: It should be able to justify its conclusions in the same way a human expert explains why he arrived at conclusion. Comparison between an Expert System and that of a Human Expert: Factor Time (can be obtained) Geography Safety Damages Speed and Efficiency Cost

Human Expert Working days only Local Cannot be replaced Yes Changes

Expert System

High

Intermediate

Anytime Anywhere Can be replaced No Consistent

II. LITERATURE REVIEW [1] John Durkin “Application of Expert System in Science” says that an expert system is a computer program or a software that captures the knowledge of a human expert on a given problem and uses this knowledge to solve problems like the expert. So, the expert system can assist the expert during problemsolving. [2] Yasser Abdel amid “A Proposed Methodology for Expert System Engineering” Explained that the development methodology of an expert system has two aspects: Knowledge engineering and Software engineering. In the software engineering there are five activities for expert system development: requirements, specification, design, implementation, and testing. [3] Mario A Garcia, “An Expert System in Diabetes” describes the ESDIABETES. In U.S.A. the cost of diabetes is

estimated to represent 5.8% of total personal healthcare expenditures. Also, Saudi Arabia is the third country around the world in which is where diabetes is most prevalent among the adults. ESDIABETES was developed by computer science graduate students at Texas to help people control the blood glucose level. Edward Feigenbaum of Stanford University has defined expert system as “an intelligent computer program that uses kno wledge and inference procedures to solve problems that are difficult enough to require significant human expertise for their solutions.” III.

CONFIGURATION OF EXPERT SYSTEM The procedure of constructing expert systems is regularly called knowledge engineering. The internal structure of an expert system consists of mainly three parts: knowledge base, database and inference engine. Expert system = Knowledge base + Inference engine A. Knowledge base: Knowledge base includes the domain knowledge of how that is used by the inference engine to draw conclusions. B. Inference engine: The inference engine is the generic control mechanism that applies the axiomatic understanding to the projectspecific data to arrive at some conclusion. When a user supplies information or applicable facts of query to the expert system he receives recommendation or knowledge in reaction. That is even after giving the facts it uses the inference engine which in turn uses the knowledge base to produce the solution. C. User interface: User interacts with the expert system through the user interface. In other words where questions are asked, and advice is produced. In addition to the advice, that is output, the user interface can output the justification capabilities of an expert system.

Fig 1: Simple diagram of an Expert System

IV.

RESEARCH AREAS OF EXPERT SYSTEM A. Expert systems in the field of medicine: Technologies in medical treatment are critical as well as at some stage in preremedy of medical consultation, diagnosis ailment until the actual remedy through an expert or physician. Many expert systems have been advanced to diagnose diabetes and coronary heart diseases, where the diagnosis is complicated and involves up-to-date parameters. In medical sciences, numerous expert systems have been developed. These are:  PUFF: Pulmonary disease diagnosis Medical system for diagnosis of respiratory conditions.  GUIDON: Microbial disease education  VM: Monitoring of patients need to intensive care  ATTENDING: Anaesthesia management education  ABEL: Diagnosis of acidic materials and electrolytes  ONCOCIN: Treatment and management of patient’s chemotherapy  AI/COAG: Blood disease diagnosis  MYCIN: Microbial disease diagnosis and treatment Medical system for diagnosing blood disorders. First used in 1979.  AI/RHEUM: Rheumatic disease diagnosis  CADUCEUS: Internal medicine disease diagnosis  BLUEBOX: Depression diagnosis and treatment  ANNA: Monitoring and treatment analysis  DENDRAL: Used to identify the structure of chemical compounds. First used in 1965.



PROSPECTOR: Used by geologists to identify sites for drilling or mining B. Expert systems in the field of automobiles  Diagnosing Heavy-Duty Diesel Engine Faults: The major desire of HeavyDuty Diesel Engine Expert System (HDDE-ES) is to provide second expert guidance in dealing by the whole of HDDE problems when time is depending on and when decisions must be firm in a position where a mechanic pick up a sub-system of HDDE is not available. HDDE Expert position is off the rack on Bayesian Belief Network (BBN) technology which is a “causal reasoning” appliance and it lay out a lag as a apply of nodes interconnected mutually arcs from a dependent structured network. The HDDE development of 'thinking' computer system has a graphical addict interface that constitutes a problem assignment interface and a problem identify interface.  Car Failure Diagnosis: Car failure detection is a detailed process and requires valuable level of expertise. The Car Failure and Malfunction Diagnosis Assistance System is having three parts first is the knowledge acquisition kind of thing it gets the development from an expert source and the retained development is laid away in the knowledge base of operation storage. It has the hereafter components: human expert, knowledge engineer, external sources of data and system users. Secondly, in the Graphical User Interface (GUI), we have generally told

the required interfaces in performing position functionality. Third kind of thing is the system modules which are responsible for finding the solutions for the problems and for reasoning. System module has the four components: Reasoning Specification module, the Inference Engine, the Knowledge Base and the User Adviser module. For car defect diagnosis we further have a proposed system that divides car failures into 3 hobby types i.e. Start-Up State, problems that manage occur when a person tackle to start-up the car; Run-Stable State, problems that may occur after starting the car; and MovementState, problems that may occur while the car is moving. The system is having 150 rules for different varieties of failures and details and can detect completely 100 failures types. Through the user interface, communication between system and user is done which is represented as a menu that displays questions to users and in response the user replies yes or no. The knowledge is brought together in the form of rules in the CLIPS (C Language Integrated Production System). C. Expert systems in the field of agriculture  Diagnosis of Diseases in Rice Plant: ESTA (Expert System for Text Animation) attempt is used for development of this expert system. The way of developing an expert system using ESTA is a multi-step process which aims at developing a domain-specific knowledge base. The steps engaged in developing a



lifestyle base are an identification of the input problem, knowledge acquisition, and representation of knowledge into the knowledge base. Simple if-do rules are used instead of if-then rules in backward chaining. Start section is the alternate section in ESTA when the demand is fulfilled advice is given. The flow of control is determined by the parameters which can be number, Boolean or text. Diagnosis of Diseases and Pests in Pakistani Wheat: We know local name of wheat are Kanak. Wheat is a Rabi crop and is grown in winters. The most commonly found disease in wheat is rust like Flag Smut, Stem Rust, Leaf Rust, etc. Wheat production rate could not catch-up with the population growth rate because of several factors such as weather, insects, viruses, fungi, bacteria and weeds etc. The rule-based expert systems have been successfully used previously. Rules are made based on the hierarchy and these rules lead to diagnoses of desire disease. For example, in diagnosing diseases of wheat, the top level involves the following typical symptoms and recommendations. Problem in leaves → disease may be Stem rust Problem in leaves → disease may be Flag smut. The knowledge base for e2glite expert system shell consists of simple if-then rules. The internal logic of inference engine determines which rules are to be fired. In rule-based expert systems forward and backward chaining are used for reasoning. Forward chaining is

data driven whereas backward chaining is goal driven. D. Expert systems in the field of steel making industry In steel making industry expert systems are preferred over the other software systems as the maintenance of expert system is easier than that for other software systems. The following four expert systems are partly in use:  The rolling mill expert system was developed to cast schedules for the steel plant.  The steelmaking plant - a scheduling expert system was developed to support the dispatchers.  A computer aided quality control (CAQC) system produces dependent on customer orders and steel grades prescriptions to achieve the desired steel quality. These are not complete in aspects like process routings and settings of process parameters for heats in steel making plants.  An expert system for the diagnosis of the blast furnace was developed. The system works interactively and does not control the furnace directly. E. Expert systems in the field of education  Teacher’s Performance Evaluation Using Fuzzy Logic: Expert system by the agency of fuzzy logic is used to evaluate teacher’s performance by per numerically weighted linguistic terms as good, very good, medium, high, low, bad, satisfied, unsatisfied etc. The approaching ES consists of at variance aspects of teachers attribute or features like research and publication, teaching-learning process, personal skill & abilities etc. To raise the value of the performance and reliability in

decision making Fuzzy logic can be incorporated directed toward ES.  Intelligent Tutoring System (ITS): This expert system is based on fuzzy logic and was developed to guide first year engineering students. This ES helps the students to help the them to have deep understanding of the topics in engineering field.  Intelligent Pascal Tutoring System (IPTS): Since the first development of CAI (Computer Aided Instruction) program, a lot of research has been done on CAI to build up expert systems which are more efficient and faster. In 1970s, a new breed of CAI became popular: Intelligent Computer Aided Instruction (ICAI) which is widely used for educational purpose. One of the ICAI that is particularly used for teaching Pascal computer language is Intelligent Pascal Tutoring System (IPTS). This expert system is used for independent teaching of Pascal to computer scholars. IPTS also contains the knowledge base for solving queries of users. IPTS teaches the students as well as keeps the students’ database to check the performance. F. Expert system in the field of management Expert system is built for the usage of different application and different group of decision makers each level in the organization, there are managers, accountant, financial analysis, consultant, strategic planners etc. In developing expert system, management teams might have difficulties during strategic plan consist of sourcing expert advisors, budget, time constraint, other management support and contribution, skill etc. An expert system as

programmed to facilitate in decision making in management and organization toward the high performance, it programs also show that the management in its decision. G. Expert system in the field of military The technologies of artificial intelligence for the most military program and hardware are wide and varied which due to robustness, reliability and durability. It showed the improvement to the military communication security, operation, control and maintenance, training areas etc. The AI application in the military field which is available in current technology for instant the CALO (Cognitive Assistant that Learns and Organizes) and PPAML (Probabilistic Programming for Advancing Machine Learning) were financed by U.S. Advanced Research Projects Agency of the Pentagon. PPAML is AI program to make accessible and effective of machine learning capability for a wide variety of information program and military weapons V. CONCLUSION There is a huge contribution to the different fields that have been made by the expert system for the past thirty years. The expert system will repeat to play an increasingly important role in the distinctive areas. In the survey done in this paper medical, automobile and agriculture areas come under the diagnosis and the other areas such as education that uses the fuzzy logic come under the decision area of the expert system. For companies, an expert system could be the master storage the knowledge and operation expertise in keeping survival the operation of the company and eliminate a problem hiring or replacing human experts. This paper is based on the review of the expert system and knowledge processing and the important research area of expert system and one can do further individual expert system research. This system is an open source, so researchers can add to it as applications of the expert system in the science are expected to increase in the near future. VI. ACKNOWLEDGEMENT This work is completed with the help of my mentor Mr. Kamlakar Bhopatkar Sir. I thank VES

College of Arts, Science and Commerce for giving me an opportunity to be part of the ACP (Additional Credit Programme). VII. REFERENCES [1] Avneet Pannu, “Survey on Expert System and its Research Areas,” Volume 4, Issue 10, April 2015 [2] Ahmad T. Al-Taani, “An Expert System for Car Failure Diagnosis,”Vol: 1, 20-12-2007. [3] Balram Kishan, Varun Chadha and Chamandeep Maini, “A Review of Development and Applications of Expert,”International Journal of Advanced Research in Computer Science and Software Engineering, Volume 2, Issue 10, October 2012. [4] Durkin, John, “Research Review: Application of Expert Systems in the Sciences,” Ohio Journal of Science: Volume 90, Issue 5 (December 1990). [5] Fahad Shahbaz Khan, Saad Razaaq, Kashif Irfan, Fahad Maqbool, Ahmad Farid, Inam Illahi and Tauqeer Ul Amin,“Dr. Wheat: A Web-based Expert System for Diagnosis of Diseases and Pests in Pakistani Wheat,” Proceedings of the World Congress on Engineering 2008 Vol.1, WCE 2008, July 2-4, 2008, London, U.K. [6] Hilda Ramirez, “ESDIABETES (AN EXPERT SYSTEM IN DIABETES),” Journal of computer Science, March 2001. [7] John ca Bullinaria, “IAI: Expert Systems,”2005. [8] Jurgen Dorn“Expert Systems in the Steel Making Industry”. [9] Mario A Garcia,”ESDIABETES (AN EXPERT SYSTEM IN DIABETES),”Journal of computer Science, March 2001. [10] Nabende Peter “An Expert System for Diagnosing Heavy Duty Diesel Engine Faults”. [11] Salama A. Mostafa, Mohd Sharifuddin Ahmad, Mazin Abed Mohammed and Omar Ibrahim Obaid,“Implementing an Expert Diagnostic Assistance System for Car Failure and Malfunction,”IJCSI International Journal of Computer Science Issues,Vol 9,Issue2,No 2,March 2012. [12] Samy S. Abu Naser, Abu Zaiter A.Ola, Al-Azhar University and Gaza, Palestine,“An Expert System For Diagnosing Eye Diseases Using CLIPS,” Journal of Theoretical and Applied Information Technology, 2005- 2008 JATIT.

[13] Satvika Khanna, Akhil Kaushik, Manoj Barnela, “EXPERT SYSTEMS ADVANCES IN EDUCATION,” NCCI 2010 -National Conference on Computational Instrumentation, 19-20 March 2010 [14] Shikhar Kr. Sarma, Kh. Robindro Singh and Abhijeet Singh,“An Expert System for Diagnosis of Diseases in Rice Plant,” International Journal of Artificial Intelligence, Volume 1, Issue1. [15] Yasser Abdelhamid, Hesham Hassan and Ahmed Rafea,“A Proposed Methodology for Expert System Engineering” Central Laboratory for Agricultural Expert Systems. [16] C. F. Tan1 , L. S. Wahidin1 , S. N. Khalil1 , N. Tamaldin1 , J. Hu2 and G.W. M. Rauterberg2, “THE APPLICATION OF EXPERT SYSTEM: A REVIEW OF RESEARCH AND APPLICATIONS”, VOL. 11, NO. 4, FEBRUARY 2016