CBR Data Warehouse Dissertation

Abstract

Case Based Reasoning (CBR) is one of the most widely used dependent variables in storing knowledge. The global way of thinking about this dissertation topic is the following: Companies that do not use CBR are making a mistake.

Case Based Reasoning has turned into an essential tool for many businesses. It is used everyday in the fields of medicine, education and law.

CBR has historically evolved since it was originally created in the US. This system was brought to Europe thanks to Michael Richter, Dr of Kaiserslauten University, who is mentioned in this dissertation and who will help identify several objectives for researching the benefits or advantages of CBR while comparing the CBR system with other methods. Michael Richter also helped in learning about the IICAI (Indian International Conference on Artificial Intelligence).

Artificial Intelligence is an important advance when talking about Information Systems. Firstly, the purpose of this dissertation was to write about this topic: Is it likely that companies can store all knowledge that their employees have accumulated in a lifetime? However, I quickly realized that this was such an extensive area and would not be able to analyze this topic. For this reason, this dissertation is an investigation about CBR and the proposal of this dissertation is to understand what CBR is and why a company uses it.

Markets are changing and nobody wants to take risks. For this reason companies want to store all knowledge that their employees have accumulated. Today, competition among companies is very intense. The differences between vanguard, advanced and normal companies will be analyzed.

There are many advances in the CBR field and it is growing everyday. This dissertation wants to help explain popular programs like CHEF, which is available on the Internet. Therefore, CBR can be used by anyone who does not want to make mistakes in the future.

It was found that CBR has become an important part of organisations, and can increase the finances of a company. So Companies like Siemens, in countries such as Germany, are using this type of technology.

Emerging countries, such as India, are using CBR. India is developing this system because they want to solve new problems based on the solutions of similar problems experienced in the past.

It is also important to compare different CBR processes in order to understand the CBR cycle.

It is obvious that CBR is an important aspect of companies, countries and different areas, including medicine, where it is used everyday.

1. Introduction.

Modern times are characterized by an increase in complexity: Information Technologies (IT) as well as telecommunications accelerate change.

Markets are changing, nobody wants to take a risk and competence is growing. Products and services are becoming obsolete and companies are characterized by the creation of knowledge, which is the main asset to companies. This is why companies want to incorporate new products or services. For a lot of organizations, strategic information is a very important component in order to have advantages over other companies.

This is the new economy, in fact, the traditional economic factors like work and capital are not as important as knowledge. Companies are discovering that their main asset is the knowledge of their employees. Information sciences include data, knowledge, wisdom and finally information.

Companies are beginning to realise the importance of the knowledge of their employees and are trying to manage it. Knowledge is recognized as the best and most important asset for them. So companies are trying to administrate the knowledge that the employees have stored.

It is necessary to emphasise knowledge, learning and knowledge management. These could be seen as inseparable concepts. Knowledge is defined as the organized set of data and information destined to solve problems, but learning is the process of acquiring knowledge, skill, attitudes or values through study, experience or education. On the other hand knowledge management “refers to the process of creating value from an organization´s intangible asset” (Liebowitz,2001,p.2).

According to Nonaka (1995) there are two types of knowledge, explicit and tacit. He defined explicit knowledge as something that is easy to communicate, store and distribute. On the other hand, Nonaka defined tacit knowledge as something that is difficult to explain, communicate and distribute as well.

So explicit knowledge is easy to store, in fact it has a structure and is socially constructed and tacit knowledge deals with individual references and impossible to store. It is difficult to store tacit knowledge because this concerns things which an employee has learned in his/her lifetime although some things are important and others are not. It is difficult to store in a server. To store knowledge is difficult but not impossible.

According to Serradell y Pérez (2003) in 1997, the company Arthur Andersen identified the quality of knowledge stored in a server. He projects that the amount of knowledge stored will increase every five years and by 2020 it will increase every 78 days.

But, how could knowledge be stored? There are several types of methods to store knowledge but this dissertation seeks to analyse one of the most important methods called CBR (Case Based Reasoning).

CBR is based on two tenets about the nature of the world: if the world is regular, similar problems have similar solutions; while on the other hand, the types of problems tend to repeat themselves.

So CBR, is the process of solving new problems based on the solutions of the similar past problems. That is why this process is very important. Furthermore, the impact that the CBR system will have within organisations and the necessity of Knowledge Management for understanding Information Technology will have a relevant importance.

The global way of thinking about this dissertation topic, or the research question is the following: Are companies that do not use CBR making a mistake? It is important to analyse why companies do not use CBR. One of the most important considerations when selecting the CBR system is that this choice will reduce the cost.

Furthermore it is important to know the aims, why companies use CBR and is CBR better than other methods?

To identify the five different objectives of this dissertation is really important. They are:

1. To research CBR benefits or advantages.

2. To research the amount of computer infrastructure that companies need in order to use CBR.

3. To compare the CBR system with other methods.

4. To analyse future trends in the use of CBR, and why the Internet is important for this, so as to know whether emerging economies will use CBR to store knowledge.

5. To determine if a business organization should change to this solution for storing knowledge; this is called CBR.

Michael Ricther and Jose Lusi Campuzano will participate in this dissertation and will help analyse and determine the trends in the use of CBR as well as the future trends. Michael Ricther is a professor at the University of Kaiserslauten (Germany) and Jose Luis Campuzano is an ICT (Information Communication Technology) Analyst for Siemens. He currently works for the National Assembly of Wales in Cardiff.

This dissertation is structured in twelve sections including the conclusions and interviews. The first two sections provide an overview of the methodologies. The research approach is explained and defined here in order to determine if an inductive or deductive approach should be used. This section defines Qualitative Research, Quantitative Research and Mixed Research. Finally, it analysed the objectives in relation to the methodologies.

Section three describes the literature review, identifying and evaluating different CBR systems that companies or Governments are using. The Historical evolution of CBR is also explained here.

Section four describes the process that CBR uses as Case Retrieval, Case Re-use, Case Revision and finally Case Retainment. This is the most important section in order to understand this dissertation.

Section five presents and discusses this method compared to others in the areas of law, medicine and education. This chapter analyses what CBR methods are used in each sector and why we should or shouldn’t choose CBR. Examples are given for the different sectors and each method is analysed.

Section six reflects an example application called CHEF. Section seven explains the future trends in the use of CBR and the importance of the Internet and whether it has influenced CBR. This chapter also covers the current state of CBR in Europe and the USA, which is the main market, and if other emerging countries such as India, China and Brazil are going to use it and why.

Finally section eight presents some CBR tools and analyses what tools are actually working. Section nine shows the advantages of CBR.

In section ten, a study result and analysis is carried out and these objectives will then be analysed and solved

Section eleven offers a conclusion and recommendations and finally, section twelve presents the interviews with Dr Michael Richter and Jose Luis Campuzano.

2. Methodology

This section will explain what methodology is used during the dissertation and what research approach is used. It defines the research strategy, and indicates what strategy is best to do this dissertation.

The research methodology is also be defined in this section. There are three types of Methodologies to choose from; Qualitative Research, Quantitative Research and Mixed Research. Finally the five objectives identified for this dissertation are explained.

One of the most important points when talking about Methodology is identifying what research approach is needed. There are three types of research approaches: the deductive approach, the inductive approach and the mixed approach. According to Gratton and Jones (2004), deductive research is used when an idea or hypothesis is being developed, while inductive research is used to develop a theory as the result of an analysis or hypothesis. A mixed approach could be used in order to decrease the weakness and emphasise the strength of both approaches.

Another important point when talking about the Methodology is to identify the research strategy. The research strategy is going to be used to contribute to the structure of the dissertation. According to Yin (2003), the research strategy is used in many situations to contribute to our knowledge of individual, group, organizational, social, political and related phenomena. There are several types of research strategies and the most common are: experiment, case study, action research or exploration.

Finally, the last important point in analysing the research methodology is qualitative research, quantitative research and mixed research.

According to Creswell (2003), Qualitative Research is an objective methodology; it is fundamentally interpretive. This means the researcher gives an interpretation of the data. On the other hand, Quantitative Research is a deductive form of study used as a basis for advancing research questions and hypothesis. Mixed Research is a combination of both methods, and involves objective and subjective research.

The strategy that used to write the methodology is as follows: an inductive method will be used to talk about the research approach. This is because the dissertation is going to analyse several theories or different cases. On the other hand, one important point to aid analysis in the dissertation is to choose the research strategy. The method that is going to be chosen is exploration, because this dissertation analyses application examples and tools where CBR is used by companies.

Finally, this dissertation uses a mixed methodology because this is a combination of qualitative and quantitative methodologies. Mixed methodology involves the analysis of quantitative and qualitative data.

In this dissertation there are five objectives. It is very important to explain them:

1. To research the benefits or advantages of CBR.

2. To research the amount of computer infrastructure that companies need in order to use CBR.

3. To compare CBR systems with other methods.

4. To analyse future trends in the use of CBR, and why the Internet is important for his, so as to know whether emerging economies will use CBR to store knowledge.

5. To determine if a business organization should change to the CBR solution for storing knowledge.

Objective 1- To research the benefits or advantages of CBR.

To research the benefits or advantages of CBR, it is important to use two methods. First, it is basically important to analyse the data from the interview. Michael Ricther, a professor at Kaiserslauten University (Germany) is going to help research the benefits of CBR in the education sector. Ricther was the main person that developed CBR in Europe. He developed PATDEX, but an evolution of CBR will be seen in the Literature Review. On the other hand, Jose Luis Campuzano, analyst of Siemens, is going to help by sharing information about Siemens. For example why Siemens uses CBR or the advantages of CBR.

Finally, finding information in books and on the Internet in another method.

Objective 2- To research the amount of computer infrastructure that companies need in order to use CBR.

To research the amount of computer storage that companies need, Jose Luis Campuzano (Siemens Analyst) is going to provide me with the necessary information.

Objective 3-To analyse future trends in the use of CBR, and why the Internet is important for this, so as to know whether emerging economies will use CBR to store knowledge.

To analyse the future trend in the use of CBR, there is a website for the most recent conference for CBR in the world. With this, information can be found about the emerging economies and how they use CBR to store knowledge.

Objective 4- To compare the CBR system with other methods.

To compare the CBR system with other methods, the help of Michael Ricther from the Kaiserslauten University (Germany) will be needed. He states why he used CBR instead of other systems. Information as also been obtained from books and the Internet.

Objective 5- To determine if a business organization should change to the CBR solution for storing knowledge.

This is probably the most important objective of the study; all the information gathered during the study provides the researcher with enough knowledge about CBR. To achieve this objective, qualitative and quantitative data is requested. This is also explained in the conclusion.

3. Literature Review

At the beginning of this era it was very difficult to know what two words would be used when we were talking about Information Systems or Technology: Knowledge and Internet. Knowledge can be considered as the unique resource that grows when it is shared or transferred and Internet is where it is possible to share this Knowledge.

To know the period of technological developments and industrial as well as military forces, it is important to speak about the Cold War.

This was a period of conflict, tension and competition between the United States and the Soviet Union and their allies from 1940 until 1990 (Loveman, 2006). In this Cold War period the USSR launched Sputnik and the US created the ARPA (Advanced Research Projects Agency) that finally created the Internet.

The difference between the US and the USSR was very important. The USSR wanted to protect their data and the US shared their data with every community to improve it. USSR was closed and it was impossible for them to communicate with other countries and share information. As a result, the USSR became obsolete and the US the first technological power.

Today, thanks to the USSR and other organizations that did not share their knowledge, the importance of shared knowledge has become known. For this reason companies think that shared knowledge is a competitive advantage. One of the most important assets of a company is knowledge, but contrary to other assets, sharing this knowledge allows the company to grow.

In an interesting journal Susman (2002), an employee of Microsoft, wrote that if one employee of a bank leaves his work, takes all the money and is seen by the security camera, there would not be any doubt: the company would call the police immediately. Nevertheless, every day there are workers leave their companies taking knowledge with them. Many see this as something quite normal in working world.

According to Hanley(2002), manager of the Corporative Department in Plural Electronic Commerce INC, the necessity has always existed to keep our knowledge. Nevertheless, at the present time, we consider the advantages that collaboration and shared knowledge offer.

On the other hand Davenport (2002), director of the Strategic Department of Accenture, said that when clients ask for services they are not trusting in the professional knowledge of only one worker in the company, they are interested and trusting in the experience of the company as a whole.

This is the behaviour that companies try to introduce: new tools for the management of knowledge. Briefly, knowledge management allows employees to work with an intelligent form with less effort.

In a study carried out by KPMG between 423 companies, 67 percent of the workers affirmed that they managed too much information while 56 percent disagreed with having to begin from zero when they started a new project.

Companies try to better organize their internal resources, for this reason companies install tools to store knowledge.

Therefore, companies want the employees to access the information that they need rapidly. This is why companies such as Google or Microsoft created Google Search Appliance or Microsoft Business Solutions; these tools are for the user to have access to information as soon as possible.

The Intranet of the Ford Motor Company is used by 200,000 employees to look for information. The content is updated regularly and the employees can select the main page by defect or also personalize the system.

Davis(2002), said that the automobile company has a productivity benefit of around seven million dollars a year. That is why it is important for employees to have fast access to the information that they need.

It is very important to update regularly the knowledge that the employees have accumulated.

(Figure 1). Arthur Andersen´s knowledge store.

Source: Serradell and Pérez,2003.

The storing of knowledge has been studied extensively in Information Systems (Figure1). According to Serradell and Pérez (2003) in 1997 Arthur Andersen´s company identified the quality of knowledge stored in a server. They project that the amount of knowledge stored will increase every five years and by 2020 it will increase every 78 days.

Nowadays companies have to be more efficient. Their information systems need to be more flexible and eliminate both internal and external barriers. To know where this knowledge is when needed is important to achieve a competitive advantage.

Actually, thanks to the amount of information that companies store, it is almost impossible to use the stored information.

Many authors report the difficulties in measuring the benefits from IS in a consistent and replicable way (Lautanala et al.1998)

For this reason companies are using consulting speciality systems or intelligence systems.

According to Copeland (2000), Artificial Intelligence (AI) is usually defined as the science of making computers do things that require intelligence when done by humans.

The most important components of intelligence are learning, reasoning and finally perception. This concept is very complex and is needed to understand CBR (Cased Based Reasoning) later.

According to Copeland (2000), learning is distinguished in different ways. The simplest is learning by trial-and-error. Learning is defined as the “processes that include the acquisition of new declarative knowledge” (Michalski, Mitchell and Carbonell, 1985,p.3). An interesting example of learning is how the game of chess is played, where one player moves at random until checkmating the other player. The program is going to remember the successful move and next time the computer is presented with the same problem it will be able to produce the answer immediately. So programs learn what has they have to do the next time.

According to Copeland (2000), reasoning is to draw inferences appropriate to the situation in hand. Inferences are classified as either deductive or inductive.

Finally, according to Copeland (2000), in perception the environment is scanned by means of various sense organs, real or artificial, and processes internal to the perceiver analyse the scene into objects and their features and relationships. It is very difficult to analyse because the same object may present many different appearances on different occasions, depending on the angle from which it is viewed, whether or not parts of it are projecting shadows, and so forth.

An expert system is a variety of artificial intelligence. According to Hendler (1987), **- expert systems are to come into wide use in practical applications, the problem of continuing maintenance and modification of the knowledge base must be addressed in the development stage.

Most current expert systems began as research tools, often developing in universities and maintained by their originators and their students. So, an expert system is a computer program that contains some of the subject specific knowledge, and contains knowledge and analytical skills of one or more human experts.

So, an Expert System (ES) is a computer application that solves complicated problems that would otherwise require extensive human expertise.

The ES simulates the human reasoning process by applying specific knowledge and interfaces. However ES represent the expertise knowledge as data or rules within the computer.

These rules and data can be called upon when needed to solve problems. Books and manual guides have a very large amount of knowledge but a human has to read and interpret the knowledge for to use it.

According to Kazi (2005) it is a system that uses human knowledge captured in a computer to solve problems that ordinarily require human expertise.

On the other hand Durkin (1994) said that a computer program is designed to model the problem solving ability of a human expert.

For Feigenbaum (2002) in his book Digital Rights Management, an intelligent computer program that uses knowledge and inference procedures to solve problems, where acquiring significant human expertise for their solutions was difficult.

In an interesting Journal about Experts System, Moghram, Ibrahim and Faisel (2000) presented the structure of expert systems. There are several components:

-The knowledge base is the component that contains the knowledge obtained from the domain expert. Normally the way of representing knowledge is using rules.

-The inference engine is the component that manipulates the knowledge found in the knowledge base as needed to arrive at a result or solution.

-The user interface is the component that allows the user to query the system and receive the results of those queries.

One important variety of the Expert System is CBR (Case Based Reasoning). But to better understand CBR, it is important to identify the historical evolution of CBR in the world and Europe. This it will be discussed in the next section. Another important point to review will be the process of CBR compared to other methods

There are several areas where using Case Based Reasoning is possible, such as Education, Medicine or Law. This dissertation reviews the areas of Education, Medicine and Law and presents examples of how CBR is working within these important areas.

Finally the advantages of CBR are presented as well as a broad conclusion with a several ideas of why companies that do not use CBR are they making, or not making, a mistake. Other important points included in the conclusion are why companies use CBR and is CBR better than others methods.

3.1.Historical evolution of Case Based Reasoning (CBR).

Roger Schank is the most important person when it comes to talking about CBR. Schank invented CBR when he worked with his pupils. He developed a system of solving new problems based on the solutions of similar past problems.

According to Rhem (2006) it was very difficult to develop CBR at that time, for them it was very important to investigate the analogical reasoning and the theories of concept formation, problem solving and experimental learning of Wittgenstein.

To Rhem (2006) the first system developed was CYRUS, by Janet Kolocher, a professor from YALE University. CYRUS was based on an important model of solve-learning. This model was important to develop other systems such as MEDIATOR in 1985, PERSUADER in 1988, CHEF in 1989, JULIA in 1992 or CASEY in 1989.

On the other hand, Edwina Rissland worked with her group in Massachusetts University. They were interested in the role of precedence reasoning in legal judgements.

In Europe, CBR was researched later that in the US. The early results were the work of Michael Richter and Dieter Althoff. They developed PATDEX in 1991. This was a system of technical diagnosis. To Rhem (2006), Enric and Ramon developed a system based on medical diagnosis.

Actually, Germany has several studies concerning CBR. On the other hand emerging economies such as India and countries like Japan as well as other countries on the Asian continent are developing this important system.

4. Process of CBR

As explained before “Case Based reasoning solves new problems by adapting solutions that were used to solve old problems” (Schank, 1989, p.25).

According to Plaza and Aamodt (1994) the structure of this point will have two parts. The first one analyses the cycle of CBR, identifying the main subprocesses of the CBR cycle. The second identifies a task in the structure method of CBR, so the problems to solve and task will be break down.

(Figure 2). CBR cycle according to Bergmann.

Source: Bergmann,2003.

To analyse the CBR cycle according to Bergmann (Figure 2), four general processes are identified: retrieve, reuse, revise and retain. “During retrieval the most similar case of cases in the case base are determined, based on the new problem description. During reuse the information and knowledge in the retrieved case(s) is used to solve the new problem. The new problem description is combined with the information contained in the old case to form a solved case. During revision the applicability of the proposed solution is evaluated in the real-world. If necessary and possible, the proposed case must be adapted in some way. If the case solution generated during the revise phase must be kept for future problem solving, the case base is updated with a new learned case in the retain phase” (Bergmann, 2003, p.33).

An explanation of the cycle is needed in order to understand CBR (Figure 2). A new problem called new case will go through the cycle. This case is compared with a lot of cases; this process is called retrieve because it retrieves a collection of previous cases. The retrieved case is compared and combined with a new case. So in the re-use phase, the information and knowledge are reused and an initial solution is proposed. After this, using this case is possible, but the cycle continues, a professor can apply this case to a real world scenario, and see if the option is good or not. Finally the retain phase stores this case in a database so it can be used later.

Then, the second one identifies a task in the structure method of CBR, so the problems to solve and tasks are broken down.

Figure 3 shows a task method that breaks down CBR. Here it is possible to see the break down of the four developed general processes, retrieve, reuse, revise and retain.

(Figure 3). A task method of decomposition of CBR.

Source: Plaza and Aamodt(1994)

In Figure 2 a general process of the CBR cycle is identified. Now Figure 3 identifies how CBR solves a problem and learns from experience. In the drawing (Figure 3) it is possible to see the four different general processes, and analysing each one is important. Plaza and Aamodt (1994) explained each phase of this drawing (Figure 3).

Firstly, retrieve is going to retrieve a similar case. It is possible to see four different subtasks. When a new case is analysed in the retrieve phase, it has to identify features or characteristics, collect descriptors, interpret the problem or infer descriptions. So this phase interprets the new case. Another subtask in relation with the retrieve phase is the search. This is important because the search has a similar structure to the retrieve phase and has general knowledge from the database in relation to the new case. Another important subtask is initially the match, this subtask is going to calculate the similar tasks and explain them. Finally, another subtask is the select phase this subtask is going to elaborate explanations in relation to the new case.

Secondly, reuse is going to reuse the information and knowledge to solve the problem. It is possible to see two different subtasks, the first one is a copy, which is going to copy the solution and solution method, and the second one is going to modify this previous solution by analysing the new case.

Thirdly, revise is going to revise the proposed solution. There are two subtasks, evaluate solution and repair fault. This point is very important. The first case (evaluate solution) is going to be evaluated by a teacher. The teacher will be sure to evaluate the process correctly and later he will carry out a demonstration in a real scenario. Finally, he will evaluate the model. The second case will evaluate the repair fault.

Finally, retain is going to store all tasks and subtasks of this case for a future problem.. In this task three subtasks are identified: integrate, index and extract. Integrate is going to update the information in a database for use in a future problem, index is going to generalize the subtask and determine an index. Finally extract is going to analyse and extract the solution, justifications, methods and descriptions of this new case.

A practical case is analysed in the next section in order to better understand CBR.

5. Comparison to other methods

This chapter shows different areas that use CBR. There are a lot of systems CBR apply to the education fields. This section analyses case based reasoning in the law world. Secondly, it analyses CBR in the educational and medical sectors.

LAW WORLD

According to Tamayo (2007), investigation in the law world, consists in representing problems, generating and acquiring decisions. Many people think that it is very complicated to store all the law information that the law world has accumulated. But the problem is something else. How it is possible to apply CBR to law world? The first models were based on rules for representing legal reasoning. The objectives were achievements in departments such as those of public administration.

Today, this kind is done with one application; it is called SoftLaw. This program is based different tools: a legal rule to support the resolution of the cases and knowledge sharing by the system. ‘Legal rule-based systems require and use one or more computer models of the relevant legislation. Specialized for government, this platform can integrate large bodies of complex legislation, regulations, policy and administrative procedures. Its end-user interface is a simple web browser. In the United Kingdom, the Assert rule-based system has been developed using SoftLaw´s technology.’ (Wimmer, 2005, p.192).

As previously explained the CBR system refreshes the rules while the law is changing and CBR transmits the logic representation.

But what is the main asset of CBR in the law field? The CBR System, in a legal field, means that the legal decisions (excellent decisions) are stored in a database. The objective is to answer and try to solve new cases that previously have been stored.

CBR has several problems in representation of cases and structure because the systems do not understand a structure. To solve this situation, the first CBR model CBR in this field was called CATO; this intelligent system ‘involves the transition from precedents to preferences, building arguments as to which intermediate factors (called “abstract factors” in CATO) apply.’(Moens, Spyns, 2005. p.9).

On the other hand, another important system in the law world is HYPO, which is very similar to CATO because it covers a similar area. So it is very important to know that CBR is essential in the law world.

Another important point in this section is the analysis of CBR in different areas such as the military, educational, medical and technology sectors.

EDUCATION

In an interesting journal, Wai, Tahman, Zaiyadi and Aziz (2005) analyse the Expert System in Real World Applications. They consider expert systems in different fields as agriculture, education, environmental management and medicine. This dissertation discusses the educational, medical and technology sectors.

According to (Way et al, 2005), a new CBR tool called ITS (Intelligent Tutoring System) helps the student use adapting techniques for personalising the environment. With this, the student could demonstrate learning skills in the educational sector. It is important to know that there are applications for teachers, but the ITS is very interesting for students.

When a discussion uses technology, the System Expert in the educational sector is important to differentiate between agent-based and microcomputer and web based technologies. It is important to use these two tools at the same time. ‘Agent system is a platform on which agents are deployed. Software agent system… need not be large systems, requiring enterprise-class machines to execute’ (Hinchey, 2004, p58) so agent systems are going to help the user find resources via the web.

For Andersen (2005),it is possible to connect via the Internet and use the acquire skills via the Internet with web-based technologies.

(Figure 4). Structure of ITS for teaching engineering students.

Adapted from Zorica Nedic, Vladimir Nedic and Jan Machotka(2002).

Source: Wai et al(2005).

Figure 4 illustrates the architecture of ITS that explains an operation representation. This CBR process is different from the one previously explained. In the educational field, a first-year engineer student is going to do a test which is going to help him/her t understand what he is going to study during the next year or in every engineer field. In Figure 4, a student is going to do a test and consult the database. The database in turn consults the ITS architecture system, then gives an output or response about him or her.

MEDICINE

It is important to say that CBR can be used in all types of fields such as education, law or medicine. This section analyses the medical field. Medicine is an interesting field and how CBR can help medical science is reviewed. As explained before, CBR is the process of solving new problems based on the solutions of similar past problems. CBR will help medicine doctors give better diagnoses. Thus, this type of CBR System is called the Artificial Intelligence System (AIM).

‘Artificial Intelligence is the study of ideas which enable computers to do the things that make people seem intelligent… The central goals of Artificial Intelligence are to make computers more useful while making the principles which make intelligence possible understandable’ (Szolovits, 1982).

The medical field is very broad, which is why Wai, Tahman, Zaiyadi and Aziz (2005) have stated in a medical journal that many tasks are carried out within the world of medicine, such as generating alerts and reminders, diagnostic assistance, therapy critiquing and planning, agents for information retrieval, image recognition and interpretation.

-Generating alerts and reminders: AIM is going to warn doctors about the change of a patient’s condition. AIM can send emails to let doctors know that a patient has been tested before. Doctors can finally compare the results.

-Diagnostic assistance: AIM is going to help doctors know the problems of a patient.

-Therapy critiquing and planning: AIM is going to find errors in the database.

-Agents for information retrieval: AIM can to look for information on the Internet if that is what the doctor wants.

-Image recognition and interpretation: AIM tries to analyse similar X-rays to improve the doctors’ recognition and interpretation of the results.

Currently, there are a lot of systems that doctors use in the hospital, which is why there are new intelligent systems that AIM has developed for solving problems.

Wai, Tahman, Zaiyadi and Aziz (2005) explain a new important system called CaDet. This is a computer based on a clinical decision support system for Early Cancer Detection. This System is going to store the information for its future use.

6. Example application

To better understand this dissertation, this section investigates a real CBR tool. It was very difficult to find a current tool because many tools are not up-to-date. There was a lot of information in terms of cases, but understanding a new programming language was very difficult. This is why each code line is explained here.

There are many CBR tools in use but this section analyses CASPIAN, a tool developed by Wales University Aberystwyth. It can resolve real problems of reasoning based on cases. CASPIAN is a tool that allows for using CBR and is based on the CHEF System. CHEFF was created by Hammond in 1989 (Hammond, 1999) and is based on cases for the creating recipes.

CHEF creates a recipe by using a plan that was used in a similar situation while changing it to adapt this new situation.

It is important to analyse a complex case programming in CASL with CASPIAN technology.

In Figure 5 an example in CASL language is offered.

(Figure 5) Program built in CASL programming language.

Source: University Ryerson (Canada).

In figure 5 it is possible to see:

“case instance chicken_green_beans_stir_fry is

cook_method = stir_fry;

ingredients = [chicken green_beans];

taste = sweet_n_sour;

meat = chicken;

vegetable = green_beans;”

The first line wants a recipe for chicken with green beans. Between the second and the last line, ingredients needed are listed in order to complete the cooking task. In addition, the taste the client wants can also be chosen, , in this case sweet. Finally the program offers the solution.

solution is

meat_preparation = bone;

vegetable_preparation = shell;

recipe = [[meat_preparation the meat]

[vegetable_preparation the vegetable]

[cook_method the meat 'and the' vegetable together

'using a wok']];

The solution that should be applied is seen above

Finally, there are several code lines where the rule of definition can be applied For example, a recipe for chicken and broccoli can be acquired. The system is going to change green beans for broccoli.

case instance test2 is

cook_method = stir_fry;

taste = sweet_n_sour;

ingredients = [ ];

vegetable = broccoli;

meat = chicken;

solution is

meat_preparation = bone;

vegetable_preparation = chop;

recipe = [ [ meat_preparation the meat ]

[ vegetable_preparation the vegetable ]

[ cook_method the meat 'and the' vegetable separately 'using a wok' ] ];

end;

The program changes the one step to be carried out and “cook_method the meat 'and the' vegetable separately 'using a wok'” is seen, meaning that the meat and the vegetable should be cooked separately.

Finally the CBR system is going to store two points:

-The retrieve case is going to store this information replaces the previous ingredients with the new ingredients.

-The old conditions used in this case will be replaced with the new conditions.

CHEF was created based on the CBR system, and uses the same process, in other words, the CHEF system uses retrieve, reuse, revise and retain. There are many applications on the Internet about CBR and the CHEF system is one of them.

7. Future trends in the CBR field

CBR is present in all areas. Different trends used by CBR have been analysed and the use of CBR has been explained in different areas including law, education and medicine. A practical case has been explained in a previous section, identifying and explaining the code lines that the CBR System can use. Furthermore, we learned that the CHEF system is easy to download from the Internet.

This section is going to analyse the use of CBR in Europe and the US, and why it is important that the emerging economies such as India, Brazil or China use CBR technology. Finally, different types of companies are going to be analysed, while trying to explain why some companies are interested, or not, in acquiring this new technology.

We already know that CBR is a System that allows for using Artificial Intelligence (AI). AI tries to solve and learn about determinate specific topic. AI is going to use specific knowledge and combine different reasoning methods. It is important to know where Intelligent Systems should be used.

According to Cepelia (2006), Europe and the US use CBR to earn profit. But it is interesting to know that CBR and AI are being used in risky professions such as medicine and aeronautics

On the other hand, companies are expectant to find out the benefits that can be obtained by using expert system.

(Figure 6). Benefits of knowledge-based formulation.

Source: Colbourn and Rowe,2001.

Thanks to the development of Expert Systems, improving different areas, as seen in Figure 6, is possible. The graph shows how the accuracy of decision making in a company is elevated when using Expert Systems. On the other hand, they improve the quality of work and cost effectiveness. Finally, staffing is the last position reduced. According to Colbourn and Rowe (2001) Expert Systems are going to help a company get quantitative benefits.

Harmon and King (1988), distinguish between three types of companies. The first group is called vanguard companies, those companies that want to invest in new technologies as soon as possible. Some of these companies don’t take risks to save money with the Expert Systems, while others try to save by investing in new technologies. Other types of companies are called advanced companies, and they want reliable technology. Then there are the normal companies that invest only when necessary. REVIEW

Finally, emerging technologies in countries such as India are trying to use Artificial Intelligence in companies or the government. India is very interested in AI, which is why they held the 3rd Conference in India called IICA (Indian International Conference on Artificial Intelligence).

8. Some tools of CBR

This section analyses several tools used in the CBR area. For this dissertation, several tools that work today have been analysed. It was very difficult to find different tools since many have become obsolete. Below there are several CBR tools that can be seen, such as ART Enterprise, CBR 3, ReMind and CASPIAN. CASPIAN, which was previously explained in the CHEF process.

-ART Enterprise: According to Oteo (2006), ART Enterprise is a tool that was developed in the mid 80s by Brightware. This tools is currently in use. The price of this tools is between $60,000 and $85,000 . It allows for representing production rules and logic science. ART provides a CBR kernel for the clients who want to incorporate this into their application.

-CBR 3: CBR3 was developed by Inference Corp. The CBR3 items are the most successful. According to Price (1999) the CBR3 tools are:

-CBR Express

-CasPoint

-Generator

-Tester

-Case PointWebServer.

On the other hand, CBR3 uses a simple registry structure for storing information in a database.

However, developers who use CBR3 do not need experience in programming since anyone can apply this tool.

ReMind: this tool was developed by Cognitive System, a US company. This is very similar to CBR Express and is accepted by SQL consulting.

CASPIAN: this tool was developed by Wales University Aberystwyth, and resolves real problems of Case Based reasoning. CASPIAN is a tool that allows for CBR and is based on the CHEF System. CHEF was created by Hammond in 1989 (Hammond, 1999), and is based on cases for creating recipes.

9. Advantages of CBR (Case Based Reasoning)

Price (1999, p 48) stated that ‘CBR in literature research (Hammond, 1989 ;Kolodner,1993) works in the following way:

-select a suitable case from a case-base of past solutions for the problem, by matching a description of the new problem that needs to be solved with the stored description of already solved problems.

-Adapt the solution used by the selected past case to suit the requirements of the current situation.

-Produce a new solution and evaluate it.

-If it is a good solution, store the description of the new problem just solved, together with the new solution, as a case in the case-base.

CBR is applicable to many types of problem-solving, but has proven extremely popular in the area of diagnosis, especially as technical support for help desk applications.’ But now CBR is used for diagnosing in medicine..

The advantages that CBR enders a company is also analysed in this section, which is one of the main objectives of this dissertation as well as why CBR is important in our society. The many advantages of CBR include:

- CBR is similar to human reasoning; this characteristic is an advantage for understanding how the system works.

- CBR proposes many solutions; this is very important for obtaining goals.

- It can be used in many areas, such medicine, education, law…

- CBR allows for making predictions about the possible success of a given solution.

- CBR allows clients to identify many conditions that they may not remember.

- Cased-Based Reasoning allows for making predictions based on previous cases.

- CBR avoids making the same mistakes as in the past.

10. Study results and analysis

This section discusses, analyses and interprets the different objectives and aims.

This dissertation has been divided into twelve important points. The first section, the introduction, discussed the general structure, as well as the aims and objectives. In the methodology section, the type of strategy and methodology were explained, as well as the objectives needed to be carried out in this dissertation. A literature review was explained in order to better understand CBR. This section also included a historical evolution of CBR.

Section four explained the CBR process for storing knowledge and how this knowledge can be re-used. This was the most important point of the dissertation because understanding this point makes understanding other examples of CBR methods easier.

Another important point was the “comparison to other methods.” This section analysed the different areas where CBR is used, such as the medical, law and educational sectors.

In section six a real application example that CBR uses was analysed This application was the CHEF system. This is a system based on cases for creating recipes. You can download this application from the Internet.

Another point was identifying future trends and why emerging economies are going to use CBR. This section analysed this and why emerging economies want to use CBR. This dissertation also explained some CBR tools and its advantages, which is very important and an important objective. The next section gave a broad conclusion and some recommendations. Lastly, the interviews and references were displayed.

This section discusses, , analyses and interprets the different objectives and aims. Analysing the objectives is very important.

Objective 1- To research the benefits or advantages of CBR.

The first objective is to analyse the data thanks to the interview. Michael Ricther is a professor at Kaiserslauten University (Germany). Ricther helped us find information related to CBR applications or Conferences about CBR in emerging economies, such as India. On the other hand, Jose Luis Campuzano is a Siemens analyst and he helped us identify another CBR system that Siemens uses called Knoxx. To resolve this point, the interview was important. According to the interview with Jose Luis Campuzano, in relation to the advantages of CBR, it is important to say that CBR is an easier system to use and quickly search for information. On the other hand, Michael Ricther said that CBR is good for areas such as Medicine. In this dissertation, in section nine Price (1999), discusses CBR. He analyses the different advantages, and the most important are:

CBR produces new solutions and evaluates them. The solutions used are adapted by selecting past cases that suit the requirements of the current situation. CBR is applicable in many types of problem-solving cases, but has proven extremely popular in the area of diagnosis, especially as a technical support for help desk applications. Therefore, this objective has been achieved since both Price, in his book Computer-Based Diagnostic System, and Dr. Richter, in his interview, as agree that CBR is important in medicine (diagnosis). The author of this dissertation also believes that this objective has been achieved.

Objective 2- To research the amount of computer infrastructure that companies need in order to use CBR.

Michael Richter and Jose Luis Campuzano helped research the amount of computer storage needed for CBR. This dissertation analysed several areas that use CBR such as the law, educational and medical sectors. Dr Michael Richter said that it depends on where you want to use CBR. He said that it is not necessary to have a big server if a company doesn’t need it. Dr Richter said that it is obvious for medicine or law to need a larger computer infrastructure, for example the CHEF program. Meanwhile, Jose Luis Campuzano gave his opinion and said that Siemens uses many big Windows 5.0 servers with IIS (Internet Information Service); this is similar to the Apache server. The difference is that this server is for Windows. He coincided with Dr Richter, in that it depends on what area wants to use CBR. He also said that the army, for example, does not use the same technology. Obviously, this tool can be used in more areas than in others.

Information was found on the Internet in order to solve this point and we analysed the three typical companies of today’s society: vanguard, advanced and normal companies. Obviously the computer infrastructure that a normal company uses in comparison with vanguards companies is different.

Objective 3-To analyse future trends in the use of CBR, and why the Internet is important for this, so as to know whether emerging economies will use CBR to store knowledge.

To analyse this objective, it was divided into two points in this dissertation. The

first point was called ‘Same tools’, section eight, and the second point was called future trends, section seven. In section eight, the author wrote about the several tools that are being used today by companies. Tools such as ReMind, ART Enterprise and CBR3, which analyses future trends in the use of CBR. On the other hand, and thanks to help of Dr Michael Richter, this dissertation identified the importance of emerging technologies in the CBR field. Richter provided the information that the author needed to achieve this objective. That is why emerging economies such as India held the IICAI (Indian International Conference on Artificial Intelligence) about CBR. Therefore, the author knows that countries like India want to develop CBR and other AI (Artificial Systems) to develop its economy. Thanks to Michael Richter, the author researched the Internet about how this system developed in the EU (European Union) in comparison with the US. According to www.ec.europa.eu , the European Union wants to become the first and most competitive economy in the world based on knowledge by 2010. To accomplish this objective, the Internet phenomenon is basically needed.

Objective 4- To compare the CBR system with other methods.

In section six, several CBR systems were compared with different areas such as medicine, education and law. The different CBR programs were explained here, for example, when talking about the law world, it was very important to analyse the CATO System that involves the transition from precedents to preferences, building arguments and intermediate factors. Another important method in medicine was called the CaDet System, which is a computer based clinical decision support system for Early Cancer Detection. This System is going to store the information for its future use. This point also analysed the educational sector. The tool was called ITS (Intelligent Tutoring System) which helps students to using adapting techniques for personalising the environment. With this, the student could demonstrate the skills for learning in the educational sector. This is why the author believes that this point has been achieved.

Before mentioning the CBR cycle process, understanding this point is essential.

To write information about different CBR systems, the author had to carry out in-depth research using books and the Internet. This was a difficult task there is not a lot of information about this subject in literature.

Also in section six, a complex program called CHEF was explained.

Objective 5- To determine if a business organisation should change to the CBR solution for storing knowledge.

This is probably the most important objective of the study; all the information gathered during the study provided the researcher with enough knowledge about CBR. This point will also be explained in the conclusion. This objective is very complex and there are different points of view that can be adopted. The first reason to determine if a business organization should change to the CBR solution for storing knowledge is not easy to determine. There are four different CBR systems for storing knowledge and they do not necessarily need to be called CBR. The Siemens case is a good reference. Siemens uses an Artificial Intelligence program called Knoxx. Another important point of view is the kind of company we are talking about. If a ‘normal’ or ‘small’ company wants a CBR System it may not be necessary. However, if this company wants to be a ‘vanguard’ company, it will need to store knowledge that their employees have accumulated. That is why this company should use a CBR system or another system with Artificial Intelligence in order to store knowledge. CBR is a good election because the cycle used is easy to understand and programmers can develop it easily.

After reading this dissertation, evaluating whether a business organisation should change to the CBR solution for storing knowledge is easy. The author believes that there is enough information in this dissertation in order to evaluate these objectives.

11. Conclusion and recommendations.

In our society we have been witnesses of big advances which have taken place in Information Technologies and telecommunications. The development of new communication technologies and the result of globalizing the economy have transformed the contemporary world and the social processes of communication. In the past, mankind would never have imagined this tremendous capacity and unpublished potential of communicating with each other, by overcoming limits of space and time, sending and trying information as well as developing and deepening technological capacities and knowledge as we do today.

Recognizing the value of knowledge is one of the most important things that companies and countries have achieved. Markets are changing and nobody wants to take a risk, which is why companies develop systems for storing knowledge so they can re-use it again and again.

One of the most important infrastructures developed by companies was related to the e-learning areas. Companies such as Siemens and British Airways use CBR (Case Based Reasoning) to store knowledge.

The aim of this dissertation is to opine about the functionality of CBR. The functionality of CBR is positive since I have not found any negative information against using CBR. Thanks to Dr Richter and Jose Luis Campuzano, and before reading a lot about this topic, I detected some disadvantages in using CBR such as: the possibility of using previous cases blindly, trusting previous experience without checking beforehand, or the possibility that the appropriate case might not be available.

But in my conclusion, I would like explain that CBR it is a very good system when controlled by a teacher. It is better to compare a final solution with a teacher and analyse this solution since a different solution can be used in each case.

This dissertation has analysed five objectives: researching the CBR benefits or advantages, researching the amount of computer infrastructure that companies need in order to use CBR, comparing the CBR system with other methods, analysing future trends in the use of CBR, and why the Internet is important for this, so as to know whether emerging economies will use CBR to store knowledge, and finally determining whether or not a business organization should change to the CBR solution for storing knowledge. These objectives have been resolved in the dissertation.

The aim of this dissertation is to find out why companies use CBR and, whether or not CBR is better than other methods. First, I would like analyse the former and then the latter for a more global opinion. When we discussed why companies use CBR, the answer was simple. Companies want to obtain benefits. CBR is a System that can store all knowledge from cases and use this knowledge as many times as you want.

Secondly, the answer the question whether or not CBR is better than other methods? I have to say no. CBR is not better or worse that any other method. This is an amazing topic because you can adjust an Expert System such as CBR to what you need.

If we ask what company should use CBR, the answer would be any company.

This dissertation has focused on the following research question: Are companies that do not use CBR making a mistake? There are several types of companies, in this dissertation we classified them as vanguard, advanced and finally normal. I think that a vanguard company like Siemens or Motorola are making a mistake if they do not use CBR or any Expert System to store knowledge so they can use it when they want. On the other hand, advanced companies are making a mistake if they do not use CBR because the employees will use this system to look for information rapidly and can use it again and again. Nevertheless, when we are talking about normal companies, it is very difficult to know. We would have to analyse whether these normal or small companies want to growing the future. If they do want to grow, I recommend storing the knowledge that employees accumulate so it can be used in the future. If these companies don’t aim to expand, than I believe that using CBR is not a good idea.

I would like finish this dissertation by talking about the EU and the importance of knowledge. According to the European commission www.ec.europa.eu/ knowledge is the first asset for gaining competitive advantage over the US. The European commission is driven toward four important points: eLearning , eComerce, eHealth and finally eBusiness. The European initiative strives to connect Europe to the net as soon as possible. The European Union wants to invest in the Internet and technology. Investing in these areas will help create more jobs.

This is why CBR and other Experts Systems are very important for improving our society

.

12. Interviews

Interview No.1

1) What is your name?

Jose Luis Campuzano.

2) Profession?

Analyst.

3) What is the name of your company?

Siemens.

4) What industry is it related to?

Technological sector.

5) Are you using in some manner CBR (case based reasoning) in your company or university?

Yes, we are using CBR in our company. The department that uses CBR is called e-learning area.

6) When did you start using it?

When I came to Siemens everyone was using CBR, Siemens has developed a Knoxx system which makes storing all the knowledge that we have accumulated possible.

7) How are you using it?

Really Knoxx is easy to use. You can introduce interesting information and make a search in a database. It is very simple to use and rapidly look for information. It took several minutes for the people to perform these tasks before the system was put in place.

8) Have you changed CBR to CBR Express and why not?

We have a different program called Knoxx.

9) Are you using CBR or a combination with CBR express?

No, as I explained before we used CBR in combination with Knoxx

10) What difficulties have you faced since its implementation?

The problem is that when you are working rapidly you can not introduce the information as soon as possible for this reason forget to introduce this information

11) What benefits (if they exist) has using CBR given you?

There are several benefits, the first one is the time that you save in searching for information, and I can get information about the financial department but I am sure that Siemens earns more money using this program because now we are working faster than before.

12) Have the end-users expressed any disagreement with the use of CBR?

No.

13) What kind of CBR applications are you using?

Knoxx.

14) What cost benefits (if any) has CBR given you since its implementation?

Basically we can quickly access the information. I cannot access financial department.

15) Have you made any improvements since you started using it?

I don´t know if Knoxx has improved with any version, sorry.

16) What advise would you give to someone that is just about to start a CBR project?

I am only a user.

17) What disadvantages does the CBR system have for your Company?

Only one, it is very uncomfortable to have to introduce data on Knoxx.

18) Thank you very much and I hope you enjoy your holidays.

Thank you very much.

Interview No.2

1) What is your name?

Michael Richter.

2) Profession?

Professor.

3) What is the name of your company or University?

Kaiserslauten.

4) What industry is it related to?

Education area and technology.

5) Are you using in some manner CBR (case based reasoning) in your university?

Of course, I am using CBR in my university.

6) When did you start using it?

I cannot remember.

7) How are you using it?

Basically I do tests.

8) Have you changed CBR to CBR Express or why not?

I need to use both. CBR because I do tests with it and CBR Express because it is easy to get information via Internet

9) Are you using CBR or a combination with CBR express?

Really I introduce the information via CBR Express and I can find information via CasPoint.

10) What difficulties have you faced since its implementation?

It is easy. You do not need to know about programming, you can access CBR3 via CBR Express. This tool is good for users that do not have any idea about programming.

11) What benefits (if they exist) has using CBR given you?

Yes there are a lot benefits, one of them is share information in internet via CBR Express.

12) Have the end-users expressed any disagreement with the use of Yes, it is necessary to evaluate the situation before solve a problem

13) What kind of CBR applications are you using?

CBR, CBR Express, CasPoint, CasPoint Web server.

14) What cost benefits (if any) has CBR given you since its implementation?

Time, really important to invest.

15) Have you made any improvements since you started using it?

Yes, several implementations with CBR Express and CBR.

16) What advise would you give someone that is just about to start a CBR project?

Power and motivation.

17) What disadvantages does the CBR system have for your Company?

I spend a lot of time introducing data and information.

18) Thank you very much and I hope you enjoy your holidays.

Thanks Juan. Hasta Pronto!!

References

-Andersen K.,(2005). Database and Expert Systems Applications. Springer, Copenhagen, Denmark. 2^st edition

-Bergmann R.,(2003). Developing Industrial Case Based Reasoning Applications: The Inreca Methodology. Springer, Washington DC, US 2^st edition.

-Creswll J.,(2003). Research Design: Qualitative, Quantitative, and Mixed Method Approaches. Auckand, New Zealand.

-Gratton C., Jones I.,(2004). Research Methods for Sport Studies. London, UK. Routledge.4 ^st edition.

-Harmon P., King D.,(1988). Sistemas expertos de la inteligencia artificial en la actividad empresarial. Madrid, Spain. Ediciones Díaz de Santos 14^st edition.

-Haton J., Keane M., Manago M.,(1995). Advances in Case Based Reasoning. Chantilly, France. SPRINGER. 8 ^st edition.

-Hendler A.,(1987). Expert Systems: The User Interface. Norwood,NJ, US. 2 ^st edition.

-Hinchey M.,(2005). Formal Approaches to Agent-Based-Systems. Springer, Greenbelt,US. 4 ^st edition.

-Durking J.,(1994). Expert Systems: Design and Development. Macmillan Publishing Company, New York, US. 3 ^st edition.

-Feigenbaum J.,(2003). Digital Rights Management: Acm Ccs-9 Workshop Drm 2002. Springer, Washington DC, US 2^st edition.

-Kazi A.,(2005). Knowledge Management in the Construction Industry: A Socio-technical Perspective. New York, US. IGI Global. 4^st edition.

.

-Liebowitz J.,(2001). Learning from Knowledge Engineering. Maryland,USA .CRT Press.1^st edition.

-Loveman B.,(2006). Addicted to Failure: U.S. Security Policy in Latin America and the Andrean Region. Plymouth, UK. Rowman & Littlefield Publishers. 1 ^st edition, pp 71-72.

-Michalski R., Michael T., Carbonell J.,(1985). Machine Learning. London,UK. Morgan Kaufmann. 2 ^st edition.

-Oteo A.,(2006). Gestión clínica: Desarrollo e instrumentos. Madrid, Spain. Ediciones Díaz de Santos 4^stedition.

-Price C.,(1999). Computer-Based Diagnostic Systems. London, UK. Springer 4^stedition.

-Rhem J.,(2006). UML for Developing Knowledge Management Systems. Chicago, Us. CRC Press. 2 ^st edition.

- Schank R.,(1991). Inside Case Based Reasoning. New Jersey, US. Lawrence Erlbaum Associates, Publichers,1^st edition.

-Yin K.,(2003). Case Study Research: Design and Methods. SAGE. 2 ^st edition.

-Wimmer M.,(2005). Electronic Governmen: 4^th International Conference, Egov 2005. Copenhagen, Denmark. Springer 4^st edition

Internet References

-Cepedia L.,(2006). Cited in: ‘Sistemas expertos’, online, available from(http://www.monografias.com/trabajos16/la-inteligencia-artificial/la-inteligencia-artificial.shtml), access 22/07/07.

-University Ryerson(Canada) http://www.ryerson.ca/~dgrimsha/courses/cps820/Resources/cbr/CHEF3.TXT

-Copeland J.,(2000). Cited in: ‘What is Artificial Intelligence?’, online, available from(http://www.alanturing.net/turing_archive/pages/Reference%20Articles/What%20is%20AI.html ), access 12/07/2007

- Colbourn E., Rowe R., (2001). Cited in: ‘Expert systems’, online, available from (http://www.chemsoc.org/chembytes/ezine/2001/colbourn_jul01.htm ), access 22/07/07.

-Davenport T.,(2002). Cited in: ‘Compartir conocimiento, ventaja competitiva’, online, availabefrom:(http://www.microsoft.com/spain/enterprise/perspectivas/numero_3/estrategia.mspx), access 9/07/2007.

--Davis M.,(2002). Cited in: ‘Compartir conocimiento, ventaja competitiva’, online, availabefrom:(http://www.microsoft.com/spain/enterprise/perspectivas/numero_3/estrategia.mspx), access 9/07/2007.

-Hammond K.,(1999). Cited in: ‘Case- Based Planning: Viewing Planning as a Memory Task, Chapters2, academia Press’, online, available from: (http://www.cc.gatech.edu/~jimmyd/summaries/hammond1989a.html ), access 22/07/07.

-Hanley S.,(2002). Cited in: ‘Compartir conocimiento, ventaja competitiva’, online, availabefrom:(http://www.microsoft.com/spain/enterprise/perspectivas/numero_3/estrategia.mspx), access 9/07/2007.

- Lautanala, (1998). Potential benefits of Information Technology in construction in Finland. Proceedings CIB W78 Workshop on Computer Integrated Construction, available from(http://cic.vtt.fi/vera/document/timi_eng/TIMI_CIB_W78.doc), access 14/07/07.

-Plaza E., Aamodt A.,(1994). Cited in: ‘Cased Based Reasoning: Foundational Issues, Methodological Variations and Systems Approaches’, availabe from(http://www.iiia.csic.es/People/enric/AICom.html), access 19/07/07.

-Moghram ., Ibrahim S., Faisel A.,(2000). Cited in: ‘An expert system-aided tool for teaching the designo f PID-based controllers’, online, available from: (http://findarticles.com/p/articles/mi_qa3792/is_200010/ai_n8907582/pg_1 ), access 14/07/07.

-Serradell E and Pérez A(2003). La gestión del conocimiento en la nueva economía(http://www.uoc.edu/dt/20133/index.html), access 2/07/2007.

-Susman D.,(2002). ‘Compartir conocimiento, ventaja competitiva’, online, availabe from:(http://www.microsoft.com/spain/enterprise/perspectivas/numero_3/estrategia.mspx), access 9/07/2007.

-Szolovits P.,(1982). ‘Artificial Intelligence and Medicine’, online, available from(http://groups.csail.mit.edu/medg/ftp/psz/AIM82/ch1.html ), access 21/07/07.

Tamayo M.,(2007), ‘La investigación científica’, online available from(http://www.monografias.com/trabajos15/invest-cientifica/invest-cientifica.shtml ), access 20/07/07.