Expert Systems and Geographical Information Systems for Impact Assessment - Chapter 2 pps

The same type of trees and theirassociated search methods were also used to develop game-playing methodsfor machines to play two-player games like checkers or chess, where thetree of sol

2 Expert systems and decision

2.2 EXPERT SYSTEMS AND ARTIFICIAL INTELLIGENCE

Artificial intelligence (AI) has been defined in a variety of ways, primarily

by its aims, as reflected in a number of well-known AI manuals and

text-books:

• to simulate intelligent behaviour (Nilsson, 1980);

• to “study of how to make computers do things at which, at the

moment, people are better” (Rich, 1983);

• “to understand the principles that make intelligence possible” (Winston,

1984);

• to study human intelligence by trying to simulate it with computers

(Boden, 1977)

Definitions of AI such as these tend to be based on some degree of belief

in the provocative statement made by Marvin Minsky (MIT) in the 1960sthat “the brain happens to be a meat machine” (McCorduck, 1979) which,

by implication, can be simulated The main difference between these definitions

is in their varying degree of optimism about the possibility of reproducing

in Chapter 1, by looking back at their early development and some of their

human intelligence mechanically: while the first two seem to put the

emphasis on the simulation of intelligence (reproducing intelligent behaviour), the last two – more cautious – put the emphasis rather on understanding

intelligence In fact, the tension between “doing” and “knowing” has beenone of the driving forces in the subsequent development of AI, and has alsobeen one of the root causes of the birth of expert systems

Many antecedents of AI (what can be called the “prehistory” of AI) can

be found in the distant past, from the calculators of the seventeenth century

to Babbage’s Difference Engine and Analytical Engine of the nineteenthcentury, from the chess-playing machine of Torres Quevedo at the time ofthe First World War to the first programmable computer developed in Britainduring the Second World War, together with the pioneering work of AlanTuring and his code-breaking team at Bletchley Park, part of the secret wareffort only recently unveiled in its full detail and importance (Pratt, 1987) –and popularised in the recent film “Enigma” However, the consolidation

of AI as a collective field of interest (and as a label) was very much an

American affair, and AI historians identify as the turning point the ence at Dartmouth College (Hanover, New Hampshire) in the Summer of

confer-1956, funded by the Rockefeller Foundation (McCorduck, 1979; Pratt,1987) Jackson (1990) suggests that the history of AI after the war follows

three periods (the classical period, the romantic period, and the modern

period) each marked by different types of research interests, although mostlines of research have carried on right throughout to varying degrees

2.2.1 The classical period

This period extends from the war up to the late 1950s, concentrating on

developing efficient search methods: finding a solution to a problem was

seen as a question of searching among all possible states in each situationand identifying the best The combinatorial of all possible states in all

possible situations was conceptualised and represented as a tree of successive

options, and search methods were devised to navigate such trees Searchmethods would sometimes explore each branch in all its depth first beforemoving on to another branch (“depth-first” methods); some methodswould explore all branches at one level of detail before moving down toanother level (“breadth-first” methods) The same type of trees and theirassociated search methods were also used to develop game-playing methodsfor machines to play two-player games (like checkers or chess), where thetree of solutions includes alternatively the “moves” open to each player.The same type of tree representation of options was seen as universally

Efficient “tree-searching” methods can be developed independently ofany particular task – hence their enormous appeal at the time as universalproblem solvers – but they are very vulnerable to the danger of the so-calledapplicable to both types of problems (Figure 2.1)

Expert systems and decision support 29

“combinatorial explosion”, the multiplication of possible combinations ofoptions beyond what is feasible to search in a reasonable time For instance,

to solve a chess game completely (i.e to calculate all 10120 possible sequences

of moves derived from the starting position) as a blind tree search – without

any chess-specific guiding principles – would take the most advanced puter much longer than the universe has been in existence (Winston, 1984)

com-It is for reasons like this that these techniques, despite their aspiration to

universal applicability, are often referred to as weak methods (Rich, 1983).

On the other hand, they do provide a framework within which criteriaspecific to a problem can be applied One such approach adds to the searchprocess some form of evaluation at every step (an “evaluation function”),

so that appropriate changes in the direction of search can shorten it andmake it progress faster towards the best solution, following a variety ofso-called “hill-climbing” methods

2.2.2 The romantic period

This period extends from the 1960s to the mid-1970s, characterised by the

interest in understanding, trying to simulate human behaviour in various

aspects:

(a) On the one hand, trying to simulate subconscious human activities,

things we do without thinking:

from shadows and colour differences, then reconstructing shapes

Figure 2.1 Options as trees

(concavity and convexity) from those edges, and finally classifying theshapes identified and determining their exact position

pre-programming the operation of machines to perform certain tasksalways in the same way; but as the unreliability of this approachbecame apparent – robots being unable to spot small differences in thesituation not anticipated when programming them – second-generation

robotics started taking advantage of feedback from sensors (maybe

cameras, benefiting from advances in vision analysis) to make smallinstantaneous corrections and achieve much more efficient perform-ances, which led to the almost full automation of certain types of manu-facturing operations (for instance, in the car industry) or of dangerouslaboratory activities

words by spectral analysis of speech sound waves, and by trying todetermine the grammatical structure (“parsing”) of such strings ofwords leading to the understanding of the meaning of particular messages

(b) On the other hand, much effort also went into reproducing conscious

thinking processes, like:

theorems (although substantial research did concentrate on this particulararea of development) but to general logical capabilities like expressing aproblem in formal logic and being able to develop a full syllogism (i.e

to derive a conclusion from a series of premises)

actions leading to the solution of a problem, like Newell and Simon’scelebrated “General Problem Solver” (Newell and Simon, 1963)

2.2.3 The modern period

In the so-called modern period, from the 1970s onwards, many of the

trad-itional strands of AI research – like robotics – carried on but, according toJackson (1990), the main thrust of this period comes from the reaction tothe problems that arose in the previous attempts to simulate brain activityand to design general problem-solving methods The stumbling blockalways seemed to be the lack of criteria specific to the particular problembeing addressed (“domain-specific”) beyond general procedures that would

apply to any situation (“domain-free”) When dealing with geometric

wooden blocks in a “blocks world”, visual analysis might have becomequite efficient but, when trying to apply that efficiency to dealing with nutsand bolts in a production chain, procedures more specific to nuts and boltsseemed to be necessary It seemed that for effective problem-solving at thelevel at which humans do it, more problem-specific knowledge was required

Expert systems and decision support 31

than had been anticipated Paradoxically, this need for a more specific approach developed in the following years in two totally differentdirections

domain-On the one hand, the idea that it might be useful to design computersystems which did not have to be pre-programmed but which could be

trained “from scratch” to perform specific operations led – after the initial

rejection by Minsky in the late 1960s – to the development in the 1980s of

neural networks, probably the most promising line of AI research to date.

They are software mini-brains that can be trained to recognise specificpatterns detected by sensors – visual, acoustic or otherwise – so that theycan then be used to identify other (new) situations Research into neuralnets became a whole new field in itself after Rumelhart and McClelland(1989) – a good and concise discussion of theoretical and practical issuescan be found in Dayhoff (1990) – and today it is one of the fastest growingareas of AI work, with ramifications into image processing, speech recognition,and practically all areas of cognitive simulation

On the other hand, and more relevant to the argument here, the emphasisturned from trying to understand how the brain performed certain opera-

tions, to trying to capture and use problem-specific knowledge as humans

do it This emphasis on knowledge, in turn, raised the interest in methods

of knowledge representation to encode the knowledge applicable in

particu-lar situations Two general types of methods for knowledge representationwere investigated:

(a) Declarative knowledge representation methods which describe a

situation in its context, identifying and describing all its elements and

their relationships Semantic networks were at the root of this

approach; they were developed initially to represent the meaning ofwords (Quillian, 1968), describing objects in terms of the class theybelong to (which itself may be a member of another class), theirelements and their characteristics, using attribute relationships like

“colour” and “shape”, and functional relationships like “is a”, “part

Of particular importance is the is a relationship which indicates class

membership, used to establish relationships between families of objects and

to derive from them rules of “inheritance” between them If an objectbelongs to a particular class, it will inherit some of its attributes, and they

do not need to be defined explicitly for that object: because a penguin is abird, we know it must have feathers, therefore we do not need to registerthat attribute explicitly for penguins (or for every particular penguin), butonly for the class “birds”

Other declarative methods like conceptual dependency were really ations of the basic ideas used in semantic networks Frames were like “mini”

vari-semantic nets applied to all the objects in the environment being described,each frame having “slots” for parts, attributes, class membership, etc evenof” and “instance of” (Figure 2.2)

for certain procedures specific to them We can trace the current emphasis

on “object-oriented” approaches to computer technology to these frames

and networks of the 1970s Also, scripts were proposed to represent

contextual knowledge of time-related processes, standard sequences ofevents that common knowledge takes for granted, like the sequence thatleads from entering a bar to ordering a drink and paying for it As with therest of these methods, the emphasis is on common-sense knowledge that wetake for granted, and which acts as backcloth to any specific problem-solvingsituation we encounter

(b) Procedural knowledge representation, on the other hand, concentrates

not so much on the description of a situation surrounding a problem, but

on the articulation of how to use the knowledge we have (or need toacquire) in order to solve it The most prominent of these approaches has

been the use of production rules to represent the logic of problem-solving,

“if-then” rules which can be used to express how we can infer the values ofcertain variables (conclusions) from our knowledge of the values of othervariables (conditions) By linking rules together graphically, we can drawchains (“trees”) of conditions and conclusions leading to the answer for the

question at the top These inference trees do not describe the problem but

simply tell us what we need to know to solve it, so that when we providethat information, the solution can be inferred automatically For example,

a rudimentary tree to work out if a project needs an impact assessment

A tree like this is just a representation of a set of “if-then” rules whichmight be worded like this:

Figure 2.2 A semantic network.

Source: Modified from Rich, 1983.

might look like Figure 2.3

Expert systems and decision support 33

Rule 1: if the project impacts are likely to be significant

or if the project type is included in the guidelines’ list

then an impact assessment is needed

Rule 2: if the project is a nuclear reactor

or if the project is an oil refinery

or if the project is

then the project type is included in the guidelines’ list

Rule 3: if the scale of the project is of more than local importance

then the project impacts are likely to be significant

Rule 4: if the extension of the project (in hectares) is greater than 20

then the scale of the project is of more than local importance

As the values of the variables at the bottom of the tree (the “leaves”) are

obtained – normally by asking screen-questions about them – the appropriateproduction rules are “fired” sending their conclusions up the tree to activateother rules, until an answer is derived for the top question

When queried about whether “an impact assessment is needed”, theinference process will first try to find if there is any rule which has this as its

conclusion (Rule 1 in our example), and it will try to answer it by finding if

the conditions in that rule are true In this case, there are two conditions

(that the impacts are likely to be significant, or that the project is of a certaintype) and the fact that they are linked by an “or” means that either of themwill suffice Therefore, the inference will try to evaluate each condition inturn, and stop as soon as there is enough information to determine if therule is true

Figure 2.3 Inference tree

Repeating the same logic, in order to evaluate the first condition about

“the impacts being significant”, the process will look for a rule that has this

as its conclusion (Rule 2 in our example) and try to see if its condition(s)are true – in this case, the condition that “the scale is of more than localimportance” Then, in order to conclude this, it will need to find anotherrule that has this as its conclusion (Rule 3 in our example) and try to evaluateits conditions, and so on

When, at the end of this chain of conclusions and conditions, the process

finds some conditions to be evaluated for which there are no rules, the

evalu-ation of those conditions has to be undertaken outside the rules The usual

way will be to find the information in a database or to ask the user In the latter case, the user will simply be asked to quantify the extension of the

project (in hectares) and, if the answer is greater than 20, then the chain of

inference will derive from it that the project needs an impact study, and thiswill be the conclusion

The logic followed in this example is usually referred to as chaining” inference, which derives what questions to ask (or what condi-tions to check) from the conclusions being sought in the correspondingrules Another possible approach is usually referred to as “forward-chaining”inference, by which information or answers to questions are obtained first,and from them are derived as many conclusions as possible.4 This type ofinference is also embedded in similar trees as shown above, but it can also

“backward-be useful to represent it with simpler flow diagrams showing the succession

of steps involved in the inference process The “data-first” diagram for

tree diagram, even if both represent basically the same deductive process ofderiving some conclusions from answers to certain questions, following thesame logical rules

Inference trees have the inherent appeal of having two-in-one uses: theyrepresent the logic of analysing a problem, and at the same time they showthe steps necessary to solve it But their visual effectiveness diminishes rapidly

as the complexity of the problem increases, as the number of “links”between levels increases and lines begin to cross A clear understanding ofsuch complex trees would require an impractical three-dimensional repre-sentation, therefore trees tend to be used only to describe relatively simpleprocesses – or, as here, to illustrate the principle – and flow diagrams areoften preferred in practical situations

It is not by chance that the development of these methods was concurrent

with the growing interest in expert systems in the 1970s Semantic nets and

classificatory trees were often used in the first expert systems to representrelationships between types of problems or aspects of the problem, and

4 Also, backward and forward chaining can be combined, so that, at every step of the ence, what information to get is determined by backward chaining and, once obtained, all its possible conclusions are derived from it by forward chaining

infer-such approach (Figure 2.4) would look quite different from the previous

Expert systems and decision support 35

production rules were used to derive conclusions to solve them CASNET(developed in the early 1970s at Rutgers University to diagnose and treatglaucoma) used semantic nets as a basis of a model of the disease, linkingobservations to disease categories and these to treatment plans.INTERNIST (also known as CADUCEUS, developed at the same time atCarnegie-Mellon University in Pittsburgh for general medical diagnosis)had its central knowledge represented by a disease tree linked to sets ofsymptoms, to be matched to the data about the patient PROSPECTOR

Figure 2.4 Data-first flow diagram

(developed at Stanford University in the late 1970s to help field geologistsassess geological deposits) contained a taxonomy of the geological world in

a semantic net, and a series of geological “states” connected by rules MYCIN(developed also at Stanford in the early 1970s to help doctors diagnose andtreat infectious diseases) organised its substantive knowledge about types

of patients, symptoms and diseases into classificatory trees, and applied theactual consultation using connected sets of rules Although quite a fewexpert systems caught the attention in the 1960s and early 1970s, it isprobably fair to say that PROSPECTOR and particularly MYCIN bestexemplify the potential of production rules for this new approach to problem-solving and, in so doing, also provide a paradigm for the development ofmost expert systems today

2.3 EXPERT SYSTEMS: STRUCTURE AND DESIGN

The idea that the methodology for solving a particular type of problem can

be represented by a set of connected rules (and an inference diagram),which can then be applied to a particular case, has been at the root of theappeal and of the development of expert systems from the beginning and,

to a certain extent, has given shape to what is still considered today a

“standard” structure for these systems (Figure 2.5):

• The knowledge needed to solve a problem is represented in the form of

if-then rules and kept in what is known as the knowledge base

• To “fire” the rules and apply the inference chain, an inference engine

Expert systems and decision support 37

• If some of the information needed is to come from existing data instead

of the user (or if the output from the system is to be stored), a database

appropriate to the problem must be connected to the system

MYCIN applied “backward-chaining” inference – deriving the necessaryconditions from the conclusions sought and working out that way whatinformation is needed – in what is now a well-established approach In thiscontext, the inference engine’s role is:

• to derive what conditions need to be met for an answer to the mainquestion to be found;

• to identify what rules may provide values for those conditions;

• to derive from those rules, in turn, what other conditions are needed todetermine them;

• when no rules are found to derive information needed, to either find it

in the database or ask appropriate questions of the user;

• once all the information needed has been found, to infer from it theanswer to the overall question;

• finally, to advise the user about the final conclusion

What was important and innovative at the time from the computingpoint of view, was that which part of the knowledge base would be used atany time, while running the system (the order of “control” evaluating the

rules) was not pre-determined as in conventional computer programs – by

writing the program as a particular sequence of commands – but woulddepend on how the inference was going in each case.5 As information con-cerning that specific case was provided, the successive rules applicable atevery stage of the inference would be “found” by the inference enginewhatever their location in the knowledge base, without the need for theprogrammer to pre-determine that sequence and to write the rules in anyparticular order

Although initially this type of inference logic was embedded in theMYCIN expert system linked to its rules about infectious diseases, it wassoon realised that it could be applied to other problems as long as theycould be expressed in the form of if-then rules of a similar kind This led

to the idea of separating the inference engine from a particular problem and giving it independence, so that it could be applied to any knowledge

base, as long as its knowledge was expressed in the form of if-then rules.The new system developed along these lines became known as EMYCIN(“empty” MYCIN), and this idea has since been at the root of the proliferation

5 This style of program writing was taking one step further the growing preference in the

computer-programming industry for so-called structured programming, which replaced

traditional control changes using commands like “go to” by making all the parts of a puter program become integrated into one overall structure

com-(commercially and for research) of a multitude of expert-system toolscalled “shells”, empty inference engines that can be applied to any rule-based knowledge base As these “shells” became more and more user-friendly, they contributed substantially to the diffusion of expert systems

and of the idea that anybody could build an expert system, as long as they

could express the relevant problem as a collection of linked if-then rules.When applying an expert system to the solution of a particular problem,the inference may be quite complicated “behind the scenes” (as encapsulated

in the knowledge base), but what the user sees is only a series of relatively

simple questions, mostly factual Because of this black-box approach, the

user may be unsure about what is going on or about the appropriateness ofhis answers, and it is common for expert systems to include some typicaladditional capabilities to compensate for this:

(a) Explanation, the capacity of the expert system to explain its logic to the user, usually taking two forms: (i) explaining why a particular question

is being asked, normally done by simply detailing for the user the chain ofconditions and conclusions (as in the rules) that will lead from the present

question to the final answer; (ii) explaining how the final conclusion was

reached, done in a similar way, spelling out what the deductive chain was(what rules were applied) going from the original items of information tothe final answer to the main question For instance, in the example of theset of rules shown before to determine if a project needs an impact assess-ment, when the user is asked to quantify “the extension of the project (in

hectares)” he/she could respond by asking the expert system Why? (why do

you ask this question?) and what the system would do is to show how theanswer is needed to determine a certain rule, in turn needed to evaluate

another, and so on, leading to the final answer The answer to the Why?

question could look something like:

the area of the project in hectares is necessary to evaluate the rule that says that

if the extension of the project (in hectares) is greater than 20

then the scale of the project is of more than local importance

which is necessary to evaluate the rule that says that

if the scale of the project is of more than local importance

then the project impacts are likely to be significant

which is necessary to evaluate the rule that says that

if the project impacts are likely to be significant

or if the project type is included in the guidelines’ list

then an impact assessment is needed

which is necessary to evaluate the final goal of whether an impact assessment

is needed

In a similar way, if the answer to the question was, for instance 23 (hectares), the system would conclude (and tell the user) that an impact