Web-Based Expert System - Assignment Example

Web-based systems Name: Number: Course: Lecturer: Date: Web-based Expert Systems Introduction to chapter two This paper talks of the web-based applications which is just an application that can be accessed from a given point or work station through an Internet or intranet. Another way of looking at it is that web-based application is just computer software that is basically hosted in a browser controlled environment, or even a web-based application can be coded in a language that supports browsers, the common ones are JavaScript combined with HTML. Web-based application are mostly preferred due to their ubiquity nature i.e. they have an easy interaction meaning ,i.e. their ability to update and maintaining the applications without the necessity of needing to update the other clients. The maintaining and configuring can all be done on one station (Coakes 1996). Web-based applications are mostly preferred because web browser do not occupy a large disk space unlike the other types of softwares.We also see that the web-based applications do not require upgrade because all the upgrade is done at the server-side and the thin clients are fed from the server. In most cases the applications can operate or function on various platforms of operating system. Our main point of discussion shall be web-based expert system which are a primarily advancement of the web-based systems. Web-based expert systems Expert systems (ES) came out of artificial intelligence (AI), that is, from the effort that was input by researchers in the AI field in their quest to develop computer programs that could reason like human beings. Many organizations have taken advantage of this technology to leverage their business operations with the aim of making profits. ES is considered to be one of the most successful branches of AI. This cannot rule out the instances of the failurs that have been experienced in this area. Even though this is the case, many companies have been known to develop proponents of the technology and have aggressively developed successful applications from this technology (Grove 2000). We can see that expert systems have advanced in the sense that, earlier there were just standalones based on mainframes computers which were actually far much slower. After the standalone systems, later came the LAN distributed applications (O'keefe & Rebne 1999). The Traditional Expert systems were indeed successful, even though they majorly had numerous limitations. It was reasonably hard for many experts to express explicitly their reasoning process, and it was also very hard to get information from many locations. Also another limitation is that the expert system is limited according to its coding, that is, it is limited in functionality meaning it cannot do more than it is designed to do or achieve. We see that most expert systems are very complicated; hence, upgrading may be difficult for the user because it may require fresh installations. Due to a lack of a common protocol this can discourage designs involving cooperation or dynamic information sharing. Due to over-reliance on the standalone systems, having the right knowledge to know how to use or operate the expert system can draw a huge problem. ICT technologies are changing Information system applications. Power that rapidly advances from Internet technologies have created new opportunities for the purpose of enhancing traditional Expert systems; we see that with the current Internet technology we can change the way ES is developed and modified due to its functionalities. For instance, in the current state any information or knowledge can be passed onto a group of users by the use of the web-based expert system, because its main function is to distribute knowledge to those with little expertise, that is, termed as non-experts. It is also noted that only a few of these Expert Systems use this Internet technology (Li & Fu 2002). We have commercial tools such as EXSYS CORVID and Xpert Rule Knowledge builder which usually are used in the web-based development. It is claimed that there is a large number of expert system available in the Internet. The Internet is often thought to be an ideal place for knowledge-based delivery than stand-alone platform (Metaxiotis & Psarras 2003). This is because the Internet is easily available; we see this in the sense that the Internet is now a global village, meaning one can access anything anywhere and the Internet coverage is world-wide. We also see another reason for it being an ideal place for knowledge-based delivery is the fact that all web browser have a common multimedia interface meaning that, no matter what browser one is using the output will merely be the same, and in case a change in work station occurs and one shifts to a different web browser there will be no confusion (Kunnathur, Ahmed & Charles 1996). We also see that the several knowledge based tools are compatible to the Internet and there are a few which tend not to function. Another major advantage is that Internet-based applications can work on various platforms and also we can say that they are relatively portable, that is, meaning that one can work from one workstation and if need arises one shifts to a different workstation without any inconveniences being caused. There are also protocols emerging that support the knowledge based support in the Internet (Huntington 2000). There are also a number of problematic issues involving the development of the knowledge based system, one of them being maintaining and being at par with the rapid changing technology, especially the changing in servers, interface components, inference engines and the major protocol, we see this being problematic in the sense that technology is changing everyday and one needs to be aware of the changes to ensure everything is functioning well and is up-to date. It also serves in reducing the overall delivery bottleneck which is as a result of communication load and a limited infrastructure (Grove 2000). So basically we can say that there are no many expert systems in the Internet, because when the Internet was being engineered the expert systems were not the main agenda of implementation. This can be attributed to the fact that web browser made it difficult for the ES to perform the function to which they are intended to achieve (Grove 2000). 2.0 CASE STUDY OF WEB-BASED EXPERT SYSTEM We will use knowledge engineering as a frame work of our case studies. Knowledge engineering is an engineering discipline that involves the integrations of knowledge in order to solve computer system problems which are complex normally requiring high computer expertise. The knowledge engineering is mainly used in expert system, decision support system and the geographical information system (Gill 1995). Knowledge engineering mainly starts with Assessing of the problem presented to be solved, after this its followed by the development of a knowledge-based system shell/structure and then the acquisition and structuring of the related information, knowledge and specific preferences, Implementation of the structured knowledge into knowledge bases, Testing and validation of the inserted knowledge this is done in the essence so as to clarify that the inserted information has been done correctly and the Integration and maintenance of the system is where the system is integrated to its supposed functionalities so as to ensure effective functioning. Revision and evaluation of the system is done to ensure that all the things are intact and are working or functioning as planned (Durkin 1996). 2.1 KNOWLEDGE-BASED EXPERT SYSTEMS It is known that human expert knowledge is mixture of a theoretical understanding in a given sample and a collection of heuristic problem-solving rules that experience has shown to be effective. Also known as knowledge-based systems can be designed by use of the knowledge from a human expert and changing it into a form that a computer may find useful in order to solve similar problems. The ‘expert’ programme chooses among alternatives, not through brute-force calculation, but by using some of the same rules-of-thumb that human experts use; so we can say that the program does not know what it knows through the raw volume of facts in the computer's memory (Duan, Fu & Li 1996). There are several examples of expert systems one being the rule-based systems that use a set of rules to analyze information about a specific class of problems and recommend one or more possible solutions .And the case-based reasoning systems adapt solutions that were used to solve previous problems and use them to solve new problems, finally the neural networks implement software simulations of massively parallel processes involving the processing of elements that are interconnected in a network architecture (Duan, Fu & Li 1996). 2.2 DOMAIN BACKGROUND AND SYSTEM DESCRIPTION There was a research funded by the European Commission’s Leonardo Davinci which entailed web-based training and support system was developed for the purpose of training and to support small and medium-sized business enterprises with the use of ICT (Duan, Burrell & Gost 2001). The major thing that triggered the project to be developed was the statements quoted “the lack of adequate skill is major barrier that SME exhibits in successfully adopting in running e-commerce and e-business” (Darlington 2000). Due to this fact, there was an urgent need for a good education and a decision support for SME managers, so we see the e-business being mostly regarded as a business issue rather than a technical issue. Because e-business affects the whole business, managers mainly need to be able to conduct analysis and appropriate business plan (Coakes 1996). According to Duan and Mullins (2001), WITS system design consists of two components: the WITS-training and WITS-advisor. WITS-training function is to provide training Material to SME managers who need to understand the technology and its application. The WITS-training has three modules namely the Internet and WWW, electronic data interchange (EDI), and e-commerce (Duan & Mullins 2001). WITS-advisor can be regarded as a web-based ES it is the more important component. It has three subsystems that are designed to facilitate SME managers’ decision making in considering the adoption of e-commerce and e-business. WITS-advisor assists SME managers assess their company’s business activities and develop an e-commerce plan, with operational support for running the applications successfully. The components of the WITS function are Business assessment which entails helping managers valuate their organization’s business performance and make recommendations on whether it needs an e-commerce solution. Figure 1: Extended knowledge engineering process Benefits and challenges The benefits and challenges that are experienced in expert systems can be analyzed from various perspectives, that is the methodological, technological and applications. As Basden (2000)argues, one of the benefits of ES is knowledge acquisition. The role that the Internet has played in knowledge acquisition is profound. One way that this is so is that the Internet provides another source of knowledge. Then the Internet also makes knowledge elicitation from the domain expert to be something that can be achieved. The third role is that users can now be engaged in knowledge generation unlike the past where knowledge only went one way. There are challenges and problems that unfortunately come with these benefits. Some of these challenges include the issue of too much information available, lack of effective data mining techniques, recognizing and locating online ES providers, sifting knowledge to get the information and knowledge which is relevant to be used in the situation that is required. There is also the issue of Knowledge representation and inference were methodologies and tools which were effectively used in traditional environments may not work in ES. There will be then be the need to rebuild these techniques so that they will be used in this platform (Athappilly 2000). Knowledge validation is one addition that will be of great value in ES systems. In this case, users can directly interact with developers through the Internet by submitting their experiences and test cases. Also, the knowledge base can be uploaded and tested by the users in this setup. In the traditional setup, users had a long process to follow in order to submit their test cases and even air their feedback. The process therefore took a long time to get accomplished. One feature which distinguishes an ES is the ability to justify results. This function is made possible with the use of Internet technology. It is also possible to get an explanation from a human expert by use of the Internet. In this regard, it can be argued that future ES shells could have functions which are in-built that will facilitate online real time communication (Adams 2001). With the advent of ES, users can now access systems globally with the use of the Internet. In this case, they do not have to have an installation in place and also do not have to be where the system is located; their location is not an issue as they can access the system from any place. With this setup, users can send their feedback by use of online forms. When updates are required in the system, this can be easily carried out from a central place. One problem that hindered initial development of web-based ES systems is that the initial perception of the web where it was meant for document sharing and distribution. This setup made it hard and difficult to deploy web-based ES. Conclusion The advent of the Internet has changed the way expert systems have been created and shared. The gist of an expert system is to imitate what an expert does and make this knowledge available in the hands of a non-expertise. This process can be enhanced a great deal with the use of the Internet. One of the questions that the paper has tried to answer is: are there any ES present in the Internet or are they there but they are not being reported? There is some problem in the methodology of developing web-based ES. We strongly believe that web-based ES will come out to be complex and sophisticated and fulfill their great promise. For this to be true there will need to be steps to solve the challenges that have been mentioned in the paper. It is the hope of many that web-based ES will bring about a new life to the field of ES and come up with a new era of applications. References Adams, J 2001, 'The feasibility of distributed web based expert systems', 2001 IEEE International Conference on Systems, Man, and Cybernetics, Tucson, AZ. Athappilly, K 2000, 'A dynamic web-based knoweldge system for prototype development for extended enterprise', 3rd International Conference on the Practical Applications of Knowledge Management, PAKEM, Manchester. Basden, A 2000, 'Some technical and non-technical issues in implementing a knowledge server, software', Journal of Information Technology, vol 30, no. 10, pp. 1127-1164. Coakes, E 1996, 'Expert systems: A survey of their use in UK business', Information and Management, vol 30, no. 5, pp. 223-230. Darlington, K 2000, The essence of expert systems, Prentice-Hall, Essex, England. Duan, Y, Burrell, P & Gost, JM 2001, 'Enhanced ewb-based interview system with expert systems and multimedia', IASTED International Conference on Applied Informatics, Innsbruck, Austria. Duan, Y, Fu, Z & Li, D 1996, 'Towards developing and using web-based tele-diagnosis in aquaculture expert systems with applications', An International Journal, vol 25, no. 2, pp. 56-63. Duan, Y & Mullins, R 2001, Project report on WITS system evaluation , University of Luton, Luton. Durkin, JD 1996, 'Expert system: A view of the field', IEEE Expert, vol 11, no. 2, pp. 56-63. Gill, TG 1995, 'Early expert systems: Where are they now?', MIS Quarterly, vol 19, no. 1, pp. 51-81. Grove, RF 2000, 'Design and development of knowledge-based systems on the web', Ninth International Conference on Intelligence Systems: Artificial Intelligence Applications for the New Millenium, Louisville, KY, USA. Grove, RF 2000, 'Internet-based expert systems', Expert Systems, vol 17, no. 3, pp. 129-136. Huntington, D 2000, 'Web-based expert systems are on the way: Java-based web delivery', PC AI Intelligent Solutions for Desktop Computers, vol 14, no. 6, pp. 34-36. Kunnathur, AS, Ahmed, M & Charles, RJ 1996, 'Experts system adoption: An analytical study of managerial issues and concerns', Information Management, vol 30, no. 1, pp. 15-25. Li, D & Fu, ZY 2002, 'Duan Fish-Expert: A web-based expert system for fish disease diagnosis', Expert Systems With Applications, vol 23, no. 3, pp. 311-320. Metaxiotis, K & Psarras, J 2003, 'Expert systems in business: Applications and future directions for the operations researcher', Industrial Management and Data Systems, vol 103, no. 5, pp. 358-361. O'keefe, RM & Rebne, D 1999, 'Understanding the applicability of expert systems', International Journal of Applied Expert Systems, vol 1, no. 1, pp. 3-24. Read More