 |
Claims  |
|
|
What is claimed is:
1. A processing system for processing knowledge data of a knowledge data
base for an electric power plant included in an inference system in which
indices representing degrees of certainty of causal relations between a
given event occurring at a device of said electric power plant and plural
events relevant to said given event are stored, said processing system
comprising:
means for receiving information regarding which of said plural events has
actually occurred with respect to said given event, said plural events
being inferred by said inference system based on said knowledge data base
by using said indices;
means for updating said indices by increasing in a predetermined mapping
relation, among said causal relations, a certainty of a causal relation
corresponding to said one of said plural events having actually occurred
with respect to said given event relative to certainties of said causal
relations of said other plural events; and
means for storing back said updated indices again in said knowledge data
base,
said indices being degrees of certainty for inferences of causes/effects of
said plural events with respect to said given event.
2. The processing system of claim 1, wherein said updating means performs
updating processing each time said causes/effects are identified after
inference of said causes/effects.
3. The processing system of claim 1, wherein said updating means performs
updating processing based on statistical data of a predetermined number of
actual occurrences.
4. The processing system of claim 1, further comprising means for providing
a user with information regarding results of inferences based on said
knowledge data base and said updated indices.
5. A processing system for processing knowledge data of a knowledge data
base included in an inference system in which certainty factors
representing degrees of certainty of causal relations between a given
event and plural events relevant to the said given event are stored, said
processing system comprising:
means for calculating a degree of frequency of actual occurrences of a
causal event (or a subsequent event) inferred by said inference system
with respect to said given event based on said knowledge data base by
using said certainty factors; and
means for updating said certainty factors of plural causal relations, which
are relevant to said given event, in accordance with said degree of
frequency of actual occurrences of said causal event (or subsequent
event), so calculated, the thus updated certainty factors being stored
back to said knowledge data base.
6. A processing system for processing knowledge data of a knowledge data
base included in an inference system in which certainty factors
representing degrees of certainty of causal relations between a given
event and plural events relevant to said given event are stored, said
processing system comprising:
means for calculating a degree of frequency of actual occurrences of a
causal event (or a subsequent event) inferred by said inference system
with respect to said given event based on said knowledge data base by
using said certainty factors; and
means for updating Said certainty factors of plural causal relations, which
are relevant to said given event, in accordance with said degree of
frequency of actual occurrences of said causal event (or the subsequent
event), so calculated, the thus updated certainty factors being stored
back to said knowledge data base,
wherein said degree of frequency of actual occurrences of said causal event
is used as a significant degree when an incrementing number of inferences
of said causal event (or the subsequent event) for said given event has
reached at least a predetermined number, said predetermined number being
variably presentable, and
wherein the difference between said certainty factor of said causal event
(or the subsequent event) and said degree of frequency of actual
occurrences of said causal event (or the subsequent event) is used for
updating said certainty factor.
7. The processing system of claim 6, wherein said updating means performs
said updating such that said difference becomes closer to 0.
8. The processing system of claim 6, further comprising means for
estimating a limit value of said frequency of actual causal event
occurrence.
9. The processing system of claim 8, wherein said updating, itself, is
modified for either suppressing or promoting updating of said certainty
factor in accordance with said difference between said frequency and said
limit value of said frequency.
10. A processing system for processing knowledge data of a knowledge data
base for an electric power plant included in an inference system in which
certainty factors representing degrees of certainty of causal relations
between a given event occurring at a device of the electric power plant
and plural events occurring at other devices of the electric power plant
relevant to said given event are stored, each of said certainty factors
being a number between zero and one, said inference system inferring a
cause/effect with respect to said given event based upon said knowledge
data base by using said certainty factors, said processing system
comprising:
certainty-factor updating means for, based upon information regarding which
one of said plural events has actually occurred, increasing, in a
predetermined mapping relation, a certainty factor of one of plural causal
relations between said given event and said one of said plural events,
said one of said plural events having actually occurred, the thus updated
certainty factors being stored back to said knowledge data base; and
means for normalizing said certainty factors of said plural causal
relations between said given event and said plural events such that a sum
of said certainty factors is equal to one.
11. A processing system for processing knowledge data of a knowledge data
base for an electric power plant included in an inference system in which
certainty factors representing degrees of certainty of causal relations
between a given event occurring at a device of the electric power plant
and plural events occurring at other devices of the electric power plant
relevant to said given event are stored, said processing system
comprising:
means for updating said certainty factors based upon information regarding
which one of said plural events has actually occurred with respect to said
given event by increasing a certainty factor of a causal relation between
said given event and said one of said plural events relative to certainty
factors of causal relations between said given event and others of said
plural events, the thus updated certainty factors being stored back to
said knowledge data base;
means for storing historical information of said updated certainty factors
from said knowledge data base in said inference system;
means for classifying events into at least three levels such that events of
higher certainty factors are classified as semidetermined events, events
of lower certainty factor are classified as rare events and events of
intermediate certainty factors are classified as unstable events based
upon said historical information of said Updated certainty factors from
said knowledge data base said inference system; and
means for providing a user with results of said classification performed by
said classifying means.
12. A processing system for processing knowledge data of a knowledge data
base for an electric power plant included in an inference system in which
certainty factors representing degrees of certainty of causal relations
between a given event occurring at a device of the electric power plant
and plural events occurring at other devices of the electric power plant
relevant to said given event are stored, said processing system
comprising:
means for updating said certainty factors based upon events having actually
occurred at a device of the electric power plant, the thus updated
certainty factors being stored back to said knowledge data base;
means for storing historical information of said updated certainty factors
from said knowledge data base in said inference system; and
means for discriminating a causal relation between said historical
information of said updated certainty factors and historical information
regarding intensities of observed physical quantities, said intensities
defining individual events having actually occurred.
13. An expert system for rendering an inference, comprising:
a knowledge data base in which an inference tree diagram connects together
groups of events in at least three layers extending from causal events to
resultant events via intermediate events,
wherein degrees of causal relations between said events in said adjacent
layers are applied., as certainty factors, to corresponding individual
inter-event routes connecting events in said adjacent layers;
means for inferring events based upon said knowledge data base; and
updating means for updating said knowledge data base so that said certainty
factors for said inter-event routes relevant to actually occurred events
are increased,
wherein said updating means includes a certainty factor distribution curve
representing distribution of certainty factors for intensities of physical
quantities, said certainty factor distribution curve being provided for
each event whose certainty factor varies depending on the intensity of an
observed physical quantity,
wherein when the certainty factor for the intensity of a given physical
quantity is updated, the certainty factor distribution curve is corrected
by said updating means by conducting interpolation between the
thus-updated certainty factor and other certainty factors,
wherein each inter-event route is defined by a combination of matrix
elements by allocating said individual event items of the inference tree
diagram as elements of a matrix consisting of N rows and M columns, N
being the largest number among the numbers of the event items in the
respective layers of said inference tree diagram and M being the number of
said layers, and
wherein a dummy row is added by said updating means to said matrix
consisting of the N rows and the M columns, a certainty factor of a given
constant value is applied beforehand to an inter-event route which ends up
with an element in said dummy row, and in inferential calculation with an
event in said intermediate layer having been determined, a dummy element
in said dummy row is used in place of the element of the event item in
said intermediate layer, said dummy element being in the same column as
the last-mentioned element.
14. The expert system of claim 13, further comprising a means for recording
both times of occurrence of plural events in a common column of the
assumed inference tree diagram and events when said plural events have
occurred successively before occurrence of events in an adjacent column.
15. The expert system of claim 13, further comprising a means for notifying
an operator of occurrence of plural events in a common column of said
inference tree diagram as an abnormal state when said plural events have
occurred successively before occurrence of events in an adjacent column. |
|
 |
|
 |
Claims |
|
|
|
Description |
|
|
BACKGROUND OF THE INVENTION
a. Field of the Invention
This invention relates to a knowledge data base processing system having
functions for updating the indices of certainty (certainty factors), which
are applied to individual event propagation routes or the like, and for
generating inferential information in an inference expert system adapted
to perform inference of causes of events or inference of secondary effects
of the events.
b. Description of the Related Art
Improvements and maintenance by a knowledge engineer are indispensable for
actual updating of a knowledge data base because the preparation of a
knowledge data base is based principally on human experiences and
inference. Several proposals have hence been made from the viewpoints of
optimization and/or automation of the updating of such a knowledge data
base.
Unexamined Japanese Patent Publication (KOKAI) No. 60-24647 proposes a
method for allocating intra-system resources in a system, to be shared by
plural software units, in the form of application and evaluation of a
knowledge base and, further, for creation of codes and selection of any
recessive codes on the basis of the evaluation. In addition, according to
the learning control method disclosed in Unexamined Japanese Patent
Publication (KOKAI) No. 60-8902, a response obtained when an object to be
controlled has been controlled by control information called beforehand
from a file is evaluated and the rule is then written in the file in
accordance with an index of the evaluation. The above proposals both
involve the procedures whereby a response from a series of operations for
an applied object is evaluated and a code or rule obtained by the
evaluation is added to or deleted from a data base.
In the inference of an event, mere addition or deletion of rules based on
the evaluation of an actual experience; on the object provides the
knowledge data base with no sufficient ground or flexibility as long as
the index of certainty (certainty factor) is fixed. It is also difficult
to say that there is established a method for applying the evaluation
results of actual experience to certainty factors.
As a known example of certainty factor, Unexamined Japanese Patent
Publication (KOKAI) No. 1-265311 discloses a method for determining more
practical certainty by providing the intensity of a process quantity,
specifically the derivative with respect to time, with values of certain
factors from 0 to 1. According to the method, the functional relation
itself of the certainty factor can be modified depending on the cause. The
cause is the intensity (the rate of a change) of the above process
quantity so that automatic updating of the certainty factor is not
performed based on an event actually experienced. In other words, an
operator or knowledge engineer of a plant allocates the functional
relation itself of certainty factor to each process quantity, which is to
be controlled, manually on the basis of the past experiences. This method
is therefore different in nature from the method such that, as in the
present invention, the history of a real event is positively evaluated and
is input to automatically update the certainty factor higher.
Further, Unexamined Japanese Patent Publication (KOKAI) No. 1-22933
discloses an inference system in which a conclusion inferred by an
inference engine is judged to be correct or not by a user and the
certainty factor of a rule in a knowledge base can be corrected by
inputting information on the correctness or incorrectness of the
conclusion. The certainty factor itself can also be corrected in this
known example, but the correction is governed by the user's judgment.
As has been described above, the conventional techniques involve problems
to the extent that they are insufficient in the reflection of
characteristics of an object of application, objectivity as an index of
certainty (certainty factor) is given based on experiences, both time and
labor are required for user's judgment as cause candidate items for
inferred results are provided unlimitedly, and the maintenance of
certainty factors in a knowledge data base requires both judgment and
labor on the side of a knowledge engineer.
SUMMARY OF THE INVENTION
It is therefore an object of the present invention to provide a knowledge
data base processing system for producing a more certain inference by
feeding back an actual event to the certainty factor of the event.
Another object of the present invention is to provide a knowledge data base
processing system which shows cause candidate items in groups classified
in the order of their certainty factors.
A further object of the present invention is to provide a knowledge data
base processing system which promotes user's judgment by providing
relevant historical information together with the results of inference of
an event and also a function to assist a change of a knowledge data base
by feeding back an actual event.
To achieve the above objects, the present invention provides, in one aspect
thereof, a processing system for a knowledge data base in which indices
representing the degrees of certainty of causal relations between an event
and plural events relevant to the first-mentioned event are stored. The
processing system comprises a means for inputting information regarding
those events having actually occurred among those events inferred based on
the knowledge data base; a means for updating the indices so that, among
said causal relations, the certainty of a causal relation corresponding to
the actually experienced event inputted by the input means is made higher
relative to the certainties of the other causal relations; and a means for
storing back the thus-updated indices again in the knowledge data base.
The indices representing the degrees of certainty of the causal relations
can preferably be applied to at least one of the first- and
second-mentioned events themselves, routes connecting the first-mentioned
event with the second-mentioned events, and inference lines each of which
consists of a plurality of continuous inter-event routes. Incidentally,
the term "inter-event route" as used herein means a path connecting two
events which belong to adjacent layers, respectively in an inference tree
diagram. On the other hand, the term "inference line" means a unit of
continuous inter-event routes.
The indices may be the degrees of certainty for the inference of
causes/effects of the event.
The index updating means can perform updating processing each time the
causes/effects are identified after the inference of the causes/effects.
As an alternative, the updating processing can be performed based on
statistical data of a predetermined number of actual extent occurrences.
Desirably, the processing system further comprises a means for providing a
user with information based on inference results, which in turn are based
on the knowledge data base and the updated indices.
In another aspect of the present invention, there is also provided a
processing system for a knowledge data base in which certainty factors
representing the degrees of certainty of causal relations between an event
and plural events relevant to the first-mentioned event are stored. The
processing system comprises a means for calculating the frequency of
occurrence of a causal event (or a subsequent event) inferred with respect
to a given event on the basis of the knowledge data base; and a means for
updating the certainty factors of plural causal relations, which are
relevant to the given event, in accordance with the degree of the
frequency so calculated.
The degree of frequency of the causal event (or the subsequent event) can
be used as a significant degree when the number of inference of the causal
event for the given event has reached at least a pre-determined number.
Preferably, the predetermined number is variably presettable.
Further, the difference between the certainty factor of the causal event
(or the subsequent event) and the degree of frequency of occurrence of the
causal event can be used for updating the certainty factor. For example,
the updating means can perform the updating such that the difference
becomes closer to 0.
It is possible to additionally provide a means for estimating a limit value
of the frequency. In this case, the manner of updating itself can be
modified in a direction toward :suppressing or promoting the updating of
the certainty factor in accordance with the difference between the
frequency and the limit value of the frequency.
In a further aspect of the present invention, there is also provided a
processing system for a knowledge data base in which certainty factors
representing the degrees of certainty of causal relations between an event
and plural events relevant to the first-mentioned event are stored. The
processing system comprises a certainty-factor updating means for
increasing, in a predetermined mapping relation, the certainty factor of a
causal relation corresponding to an actually experienced event among
plural causal relations relevant to the actually experienced event on the
basis of information relating to the actually experienced event; and a
means for normalizing the certainty factors among the plural causal
relations relevant to the actually experienced event.
In a still further aspect of the present invention, there is also provided
a processing system for a knowledge data base in which certainty factors
representing the degrees of certainty of causal relations between an event
and plural events relevant to the first-mentioned event are stored. The
processing system comprises a means for updating the certainty factors on
the basis of an actually experienced event; a means for storing historical
information on the certainty factors so updated; and a means for
classifying events into at least three levels--events of higher certainty
factors as semidetermined events, events of lower certainty factor as rare
events and events of intermediate certainty factors as unstable events--on
the basis of the historical information on the certainty factors.
Desirably, the processing system is additionally provided with a means for
providing a user with the results of classification by the classifying
means.
In a still further aspect of the present invention, there is also provided
a processing system for a knowledge data base in which certainty factors
representing the degrees of certainty of causal relations between an event
and plural events relevant to the first-mentioned event are stored. The
processing system comprises a means for updating the certainty factors on
the basis of actually experienced events; a means for storing historical
information on the certainty factors so updated; and a means for
discriminating the causal relation between the historical information on
the certainty factors and historical information regarding the intensities
of observed physical quantities, said intensities characterizing the
actually experienced, individual events. In a still further aspect of the
present invention, there is also provided an expert system for producing
an inference. The expert system comprises a knowledge data base in which
an inference tree diagram is assumed to connect together groups of events
in at least three layers extending from causal events to resultant events
via intermediate events. The degrees of causal relations between the
events in the adjacent layers are applied, as certainty factors, to the
corresponding individual inter-event routes connecting the events in the
adjacent layers. Further provided are a means for inferring events on the
basis of the knowledge data base and a means for updating the knowledge
data base so that the certainty factors for the inter-event routes
relevant to the actually experienced events are increased.
In the above expert system, a certainty factor distribution curve
representing the distribution of certainty factors for the intensities of
physical quantities may be provided for each event whose certainty factor
varies depending on the intensity of an observed physical quantity so
that, when the certainty factor for the intensity of a given physical
quantity is updated, the certainty factor distribution curve can be
corrected by conducting interpolation between the thus-updated certainty
factor and other certainty factors.
In a still further aspect of the present invention, there is also provided
an expert system for performing inference by using a knowledge data base
in which an inference tree diagram is assumed to connect together groups
of events in at least three layers extending from causal events to
resultant events via intermediate events and the degrees of causal
relations between the events in the adjacent layers are applied, as
certainty factors, to the corresponding individual inter-event routes
connecting the events in the adjacent layers. Each inter-event route is
defined by a combination of matrix elements by allocating the individual
event items of the inference tree diagram as elements of a matrix
consisting of N rows and M columns, N being the largest number among the
numbers of the event items in the respective layers of the inference tree
diagram and M being the number of the layers.
Preferably, a dummy row is added to the matrix consisting of the N rows and
the M columns. A certainty factor of a given constant value is applied
beforehand to an inter-event route which ends up with an element in the
dummy row, and in inferential calculation with an event in the
intermediate layer having been determined, a dummy element in the dummy
row is used in place of the element of the event item in the intermediate
layer. Said dummy element is in the same column as the last-mentioned
element.
It is desirable that the expert system further comprise a means for
recording both the times of occurrence of plural events in a common column
of the assumed inference tree diagram and the events when the plural
events have occurred successively before occurrence of the events in the
adjacent column.
It is also possible to provide the expert system with a means for notifying
an operator of the occurrence of plural events in a common column of the
assumed inference tree diagram to be an abnormal state when the plural
events have occurred successively before occurrence of the events in the
adjacent column.
A description will hereinafter be made of how each of the above means works
in the corresponding knowledge ,data base processing system of the present
invention.
The means for updating an index of certainty (certainty factor) given to a
tree diagram allows the processing system to perform processing by using
an actual cause or secondary effect or statistical data thereof as a
parameter of certainty for updating mapping such that the certainty factor
for an actual cause item or an actual secondary effect item, among plural
cause candidate items and secondary effect candidate items upon
observation of the same event, becomes relatively larger in order to
reflect characteristics of an object to which an event inference expert
system is applied. This makes it possible to determine a true cause for
the event or a true secondary effect of the event on the basis of more
certain inference.
The means for classifying or ranking cause items of the event or secondary
effect items of the event inferred with the certainty factors, which have
been updated by the above certainty factor updating mapping, performs the
classification or ranking by setting a multi-stage classification standard
such that the items can be ranked into semi-determined causes whose
certainty factors gradually approach toward approximately 1 as the actual
occurrence of the event increases, rare causes whose certainty factors
gradually converge at about 0 as the actual occurrence of the event
increases, and unstable causes whose certainty factors fluctuate. Thus it
is possible to provide a user with relatively certain and important
inferred information in a form successively ranked in classes.
As information accompanying the thus-ranked individual cause effect items
of the event is compiled, it is also possible to provide, as needed, the
user with historical information on the certainty factor and/or diagnostic
efficiency as approximate indication for the certainty of the
retrieval/display and inference of similar information actually occurred
in the past.
In addition, as assist information for a knowledge engineer, it is possible
to achieve the automation of maintenance of each certainty factor and also
to provide, as needed, past comparison and/or relation information such as
information on causes/effects of past events, historical information on
certainty factors and historical information on the frequency of actual
experiences. This makes it possible to assist the knowledge engineer in
connection with the partition/unification of event items in a tree
architecture or modifications of the tree architecture based on :new
relations between independent event inference lines.
Owing to the constructions described above., the present invention can
exhibit inter alia the following advantageous effects.
i) Inference with higher certainty can be realized owing to the certainty
factor updating processing function based on actual experiences.
ii) The conventional certainty factor maintenance by a knowledge engineer
can be automated by the ,certainty factor updating processing function.
iii) Objective certainty factor maintenance is feasible by the certainty
factor updating processing function.
iv) User can take necessary action quickly based on the certainty factors
displayed in ranks.
v) The frequency of actual experiences can be used for a method for the
evaluation of inference itself (or stable updating processing of a
certainty factor can be performed by feeding back to for processing the
difference between actually-experienced frequency and the certainty
factor).
vi) New knowledge assist information can be provided based on information
on the history of the certainty factory.
BRIEF DESCRIPTION OF THE DRAWINGS
The above and other objects, features and advantages of the present
invention will become apparent from the following description and the
appended claims,, taken in conjunction with the accompanying drawings in
which:
FIG. 1 is a block diagram showing the positioning of a knowledge data base
processing system according to the present invention in an expert system
for the inference of causes/effects of an event;
FIG. 2 is a block diagram illustrating the overall construction of the
knowledge data base processing system of the present invention;
FIG. 3 is a schematic illustration of an expression method for a tree
diagram;
FIG. 4 is a schematic illustration of an exemplary interference among
inference trees starting with different end events;
FIG. 5 is a conceptual diagram of updating processing for a certainty
factor distribution curve;
FIG. 6 is a functional diagram of updating processing for certainty
factors;
FIG. 7 is a schematic illustration of one example of updating map for
certainty factors;
FIG. 8 is a flow diagram of updating processing of certainty factors;
FIG. 9 is a functional diagram of a certainty factor sorter;
FIG. 10 is a relation diagram between historical information on a certainty
factor and the intensities of physical quantities observed; and
FIG. 11 illustrates an exemplary application of the present invention to a
thermal electric power plant.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
A description will hereinafter be made of a knowledge data base processing
system according to one aspect of the present invention, as incorporated
in an expert system which pertains to another aspect of the present
invention and is adapted to perform inference of causes/effects of an
event so as to diagnose any abnormality.
FIG. 1 illustrates the positioning of a knowledge data base processing
system 1000 of the present invention in the overall system.
A conventional expert system 5000 for the inference of causes/effects of an
event is constructed in the following manner. Via an interface 5100, a
user 6100 gives an inference command 4010 to an inference engine 5200
through a bus 4030. The inference engine 5200 delivers a retrieval command
4050 for knowledge data, which are required for the inference, to a
knowledge data base 5300, whereby desired knowledge data 4060 are read in
the inference engine 5200. At the inference engine 5200, an inference
operation is performed, for example, based on indices for certainty
(hereinafter called "certainty factors") in a tree diagram commonly
employed for the inference of causes/effects of an event, and the results
4040 of the inference are furnished to the user 6100 through a bus 4020.
In the meantime, a knowledge engineer 6200 per se investigates whether the
results of the inference are good or not. If he comes to the judgment that
there is a room for improvements in the knowledge data base 5300, the
knowledge engineer 6200 reads out knowledge data base information 4090 via
an interface 5400 and sends a knowledge data base maintenance command
4100--which instructs, on the basis of the results of his investigation,
updating of the certainty factors or modification of the architecture of
the tree diagram for the inference of causes/effects of the event in the
knowledge data base 5300--to the knowledge data base 5300 through a bus
4080, whereby maintenance is performed.
The knowledge data base processing system 1000 of the present invention is
shown within the area enclosed by alternate long and short dash lines in
FIG. 1. The system is basically constructed of two main units, one being a
certainty factor updating processor 2000 and the other a knowledge data
base processing information generator 3000 which serves to generate
historical information on the certainty factors or the like. These two
units are collectively called the "knowledge data base processing system
1000". In FIG. 1, with a view toward avoiding confusion, information input
to the system 1000 from the user 6100 or the inference engine 5200 and
that output from the former to the latter are designated by 4200 and 4300
respectively, while information input to the system 100 from the knowledge
engineer 6200 or the knowledge data base 5300 and that output from the
former to the latter are indicated by 4400 and 4500, respectively.
The certainty factor updating processor 2000 makes certainty factors, which
are to be used for inference, reflect each event having actually occurred
on an application object of the expert system 5000, whereby certainty
factors permitting more certain inferences are redefined.
As a main function, the knowledge data base processing information
generator 3000 generates secondary information such as information on its
own history by certainty factor updating and information of their
relations, in other words, information or the like as a result of finding
common trends from plural items of historical information. Further, as
another main function, the generator 3000 combines the thus-generated
information with inference result information output from the inference
engine 5200 and data 4110 observed from the application object 6000 of the
expert system 5000. The generator 300 then arranges them in a form
convenient for use by the user 6100 and the knowledge engineer 6200, for
example, in a diagrammatic representation with plural parameters
simultaneously displayed as functions of a common axis of abscissas, and
then furnishes the thus-arranged items of information to the user 6100 and
the knowledge engineer 6200.
FIG. 2 is a functional diagram of the knowledge data base processing system
1000. This figure clarifies the details of the information items 4200,
4300,4400,4500 input or output between the interfaces 5100 and 5400 and
also shows the flow of information between the individual processing
units.
Details of information 4200 input to the knowledge data base processing
system 1000 via the interface 5100 will next be described. The input
information 4110 represents (historical) data observed on the application
object of the expert system 5000, and is sent to a certainty factor
updating processor 2100, an event correlation discriminator 3300, an
inference information arrangement processor 3110, and the like. The
certainty factor updating processor 2100 performs updating of certainty
factors in accordance with the intensities of the observed data. Input
information 4210 consists of identification numbers of event cause/effect
items inferred at the inference engine 5200 and also calculated values of
their corresponding certainty factors. Input information 4220 consists of
actual data or log relating to an event, cause/effect items chosen as an
object for inference or as an object for the discrimination of event
correlations. Input information 4230 is event determination information
from the user. Input information 4240 consists of observed data for the
calculation of certainty factors of event effects, said certainty factors
corresponding to certainty factors for the inference of causes of the
events. The input information 4240 is delivered to an event prediction
processor 2300. Input information 4225 consists of an inference result
display format menu, which is used to produce at the inference information
arrangement processor 3100 an inference result display format desired by
the user, and a guidance for the operation of inferred causes.
Output information 4300 from the processing system 1000 to the interface
5100 are inference results of causes/effects of the event, said results
having been obtained at an inference information arrangement processor
3100.
Input information 4400 to the processing system 1000 from the interface
5400 consists of an assist information display selection menu 4410 and
initialization information 4460 to be used for the initialization of
certainty factors. The menu 4410 is used to produce assist information,
which is desired by the knowledge engineer, at a knowledge engineer assist
information arrangement processor 3200.
Output information 4500 from the processing system 1000 to the interface
5400 consists of data base storage information 4420, which in turn
consists of updated certainty factors and accompanying information on
events having actually occurred, knowledge engineer assist information
4430 produced at the knowledge engineer assist information arrangement
processor 3200, and certainty factor initialization information 4480 to be
stored in the knowledge data base.
A description will next be made of the individual function units depicted
in FIG. 2.
The calculated values 4210 of the certainty factors for the inference lines
(including the certainty factors for the individual inter-event routes),
said values having been calculated by the inference engine 5200 (see FIG.
1), are input to a certainty factor sorter 2400 and a diagnostic
efficiency calculation unit 2200. The calculated values 4210 are arranged
in order and then input, as ranked certainty factor information 4215, to
the inference information arrangement processor 3100.
A memory 1100 for history of actual events stores actual event information
4220 derived from the knowledge data base, and outputs a determined
frequency 4221 of ocurrence of each event cause/effect item especially to
the diagnostic efficiency calculation unit 2200 and historical actual
event information 4222, namely, the relevant certainty factors, observed
data and logs at the times of inference in the past to both the inference
information arrangement processor 3100 and the knowledge engineer assist
information arrangement processor 3200.
The certainty factor updating processor 2100 receives certainty factor
information 4210, the experienced event information 4230 input by the
user, and the observed data 4110 from the application object 6000 of the
expert system, performs updating on the basis of a preset certainty factor
updating map, and outputs updated certainty factors 4250.
Initial values 4470 of the certainty factors and the updated certainty
factors 4250 are input to and stored in a memory 1200 for history of
certainty factors. They are output as historical certainty factor
information 4440 as needed. It is however to be noted that no problem or
inconvenience will arise even when the historical certainty factor
information 4440 itself is stored in the knowledge data base 5300. In such
case, such historical certainty factor information should be directly
input to and output from the knowledge data base 5300 (see FIG. 1) instead
of the historical certainty factor information memory 1200.
The diagnostic efficiency calculation unit 2200 receives the determined
frequency 4221 of experience of each event cause/effect item and the
certainty factor information 4210 and outputs, as calculation results, a
diagnostic efficiency 4260 of the inference at the particular time.
The diagnostic efficiency 4260 is input to a memory 1300 for history of
diagnostic efficiencies, and is output as historical diagnostic efficiency
information 4450 as needed. The storage of the historical diagnostic
efficiency information 4450 in the knowledge data base 530 | | |
