JPH11175344A - Knowledge base system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a knowledge base system capable of sufficiently withstanding even input/output specifications requiring the updating and utilization of a knowledge base at a high speed or in real time in the construction of the knowledge base. SOLUTION: The required data set of the knowledge base 16 present on an external storage is read by a knowledge base operation part 15 and loaded on an object knowledge operation part 14 as knowledge data described as a variable structure body. An inference part 12 performs the processing of inference or the like by utilizing the knowledge data of a variable structure placed in the object knowledge operation part 14 and outputs the result through an input/output part 11 again. Also, in the case of performing change such as the change, insertion and addition, etc., of the knowledge data by the processing of the inference, the inference part 12 performs it only to the object knowledge operation part 14. At the time of ending system utilization, the knowledge base operation part 15 preserves the set of the knowledge data placed in the object knowledge operation part 14 in the knowledge base 16.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a knowledge base system, and more particularly to a knowledge base system that is updated at any time in a data processing system at high speed or in real time.

[0002]

2. Description of the Related Art A conventionally known general knowledge base system is configured as shown in a block diagram of FIG. In the figure, the inference unit 22 makes an inference based on the data stored in the knowledge base 24 in response to an instruction or inquiry for using the knowledge base, which is input from the input / output unit 21, and returns the result again. Output via the input / output unit 21.

[0003] The inference unit 22 stores knowledge data necessary for inference,
The data is read from a database (knowledge base) placed on a secondary (external) storage through the knowledge base operation unit 23 by a load operation and used. When the knowledge base 24 needs to be updated as a result of the inference processing or the like, the update processing is similarly performed on the knowledge base on the secondary storage via the knowledge base operation unit 23.

[0004] As described above, the operation of the knowledge data is performed on the data on the external storage (secondary storage) as in the general database operation. This is useful in that a knowledge base can be constructed in accordance with a database structure based on a file structure on an external storage and a method of updating the database structure, which has been used for many years.

[0005]

However, in this conventional knowledge base system, since the external storage device is accessed and used at any time during use of the system, there is a physical limit to the processing speed. However, it is insufficient as a method for configuring a system that requires a high-speed or real-time knowledge base update.

[0006] The present invention has been made in view of the above points,
It is an object of the present invention to provide a knowledge base system that can withstand input / output specifications that require updating and using a high-speed or real-time knowledge base in the construction of a knowledge base.

[0007]

According to the present invention, in order to attain the above object, an inference unit responds to an instruction or inquiry input from an input / output unit by storing knowledge data stored in a knowledge base on a secondary storage. In the knowledge base system that performs inference based on and outputs the result again via the input / output unit, of the knowledge data stored in the knowledge base, the necessary knowledge data according to the input from the input / output unit is A storage means for reading and saving from the knowledge base is provided on the main memory,
The inference unit performs an inference process using the knowledge data stored in the storage means, and updates the knowledge data with respect to the knowledge data stored in the storage means.

According to the present invention, during the use of the system, the inference processing is performed by using the knowledge data to be loaded in the storage means on the main memory, and the knowledge data is updated. Knowledge data can be used at higher speed or in real time than when using knowledge data in a knowledge base on the secondary storage.

In the present invention, since the storage means stores knowledge data described as a variable structure, knowledge can be expressed by the variable structure.

Further, in the present invention, the target knowledge update processing unit loads necessary knowledge data from the knowledge base to the target knowledge operating unit via the knowledge base operation unit at the start of use of the system, and at the end of use of the system, Since the knowledge data stored in the knowledge operating unit is stored in the knowledge base via the knowledge base operation unit, the knowledge data updated by the inference processing can be returned to the knowledge base.

[0011]

Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram of an embodiment of a knowledge base system according to the present invention. As shown in the figure, this embodiment is characterized in that a target knowledge update processing unit 13, a target knowledge operating unit 14, and a knowledge base operation unit 15 are provided on a main memory.

The target knowledge operating unit 14 stores knowledge data of a knowledge base 16 described as a variable structure according to an instruction of the target knowledge update processing unit 13. The knowledge base operating unit 15 loads the necessary knowledge data read from the knowledge base 16 into the target knowledge operating unit 14 and stores the knowledge data of the target knowledge operating unit 14 in the knowledge base 16.

Next, the operation of this embodiment will be described. At the start of the system, the inference unit 12 responds to an instruction or an inquiry for using the knowledge base 16 input from the input / output unit 11 so that the target knowledge update processing unit 13
, The knowledge base operation unit 15 is instructed to read necessary knowledge data in the knowledge base 16. As a result, a necessary data set of the knowledge base 16 existing in the external storage is read by the knowledge base operating unit 15 and loaded into the target knowledge operating unit 14 as knowledge data described as a variable structure.

Subsequently, when an instruction or inquiry for using the knowledge base 16 is input from the input / output unit 11, the inference unit 12 uses the variable structure knowledge data stored in the target knowledge operating unit 14. Then, processing such as inference is performed, and the result is output again via the input / output unit 11. The inference unit 12 performs only the change to the knowledge data, the insertion, and the addition of the knowledge data by the inference processing, and performs only the change to the target knowledge operation unit 14.

Thereafter, a system use end instruction is sent from the input / output unit 11 to the target knowledge update processing unit 13 via the inference unit 12.
In response to the instruction from the target knowledge update processing unit 13, the knowledge base operation unit 15 stores a set of knowledge data placed in the target knowledge operating unit 14 in the knowledge base 16.

As described above, in this embodiment, the use and update of the knowledge data during the use of the system is not performed in the secondary storage (external storage) but in the target knowledge operating unit existing in the main storage. 14, the processing speed is remarkably improved as compared with the conventional knowledge base system, and it is possible to sufficiently cope with a request that requires a high-speed or real-time knowledge base update as compared with the conventional one. it can.

A logical example of the knowledge data loaded into the target knowledge operating unit 14 will be described with reference to FIG. The example of FIG. 2 expresses a description of a company called Kokusai Service Co., Ltd. in a frame format. The update process of the knowledge data performed on the initial description includes an insert process, an add process, and a change process.

Assuming that the initial description is a slot description of “Headquarters Chiyoda-ku”, “Kansai Sales Office Osaka City”, and “Tohoku Sales Office Sendai” as shown in FIG. This means that the slot description “Laboratory Yokohama” is inserted at the position shown in FIG. 2, and the above additional processing means that the slot description “Kyushu Sales Office Fukuoka” is added to the end of the slot description group. Further, in the above change processing, for example, the value of the Kansai office slot is changed from "Osaka city" to "Kobe city", and the "Kansai office Kobe city"
Means to rewrite the slot description as follows.

In this embodiment, since such a free update process can be performed, not only the speed but also the description order of the slots in the frame can be given a certain "meaning". Various processes of the unit 12 can be realized.

FIG. 3 shows a detailed description of the variable structure for realizing the logical structure of the knowledge data. The figure shows a structure as an array. However, the array description does not necessarily realize a variable structure in a general computer language processing system. Here, this is merely a description means for explanation.

It can be considered that an array surrounded by "[" at the left end and "]" at the right end of each row indicates a unit of knowledge description, for example, a name and a value of a slot in a frame. FIG. 3 shows how the variable structure adds, changes, and inserts an array. Further, it can be understood from FIG. 3 that, for the variable structure, for example, the length of each structure is not fixed, and the structure can be recursively described.

[0022]

As described above, according to the present invention,
During use of the system, inference processing is performed using only the target knowledge data loaded in the storage means on the main memory, and updating of the knowledge data is performed, so that the knowledge base on the secondary storage is updated. A system that can withstand input / output specifications that require high-speed or real-time updating and use of knowledge base, because knowledge data can be used faster or in real time than when using knowledge data. Can be constructed.

Also, according to the present invention, the knowledge data is described by the variable structure by storing the knowledge data described as the variable structure in the storage means.
Update processing such as addition, change, and insertion can be easily expressed, and a more flexible and complicated structure such as a recursive structure can be realized.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of the present invention.

FIG. 2 is a diagram illustrating a frame format logical description example of knowledge data.

FIG. 3 is a diagram illustrating a description example of a variable structure using an array expression.

FIG. 4 is a block diagram of an example of the related art.

[Description of Signs] 11 Input / output unit 12 Inference unit 13 Target knowledge update processing unit 14 Target knowledge operation unit 15 Knowledge base operation unit 16 Knowledge base

Claims (4)

[Claims] 1. An inference unit performs inference based on knowledge data stored in a knowledge base on a secondary storage in response to an instruction or inquiry input from an input / output unit, and returns the result to the input / output unit again. A knowledge base system that outputs necessary knowledge data in accordance with an input from the input / output unit from the knowledge base and stores the knowledge data among the knowledge data stored in the knowledge base. Provided on a main memory, the inference unit performs inference processing using the knowledge data stored in the storage means, and updates the knowledge data with respect to the knowledge data stored in the storage means. A knowledge base system characterized by: 2. The storage device according to claim 1, wherein the storage unit stores knowledge data described as a variable structure.
The described knowledge base system. 3. The knowledge base operation unit for reading the knowledge data stored in the knowledge base and writing the input knowledge data to the knowledge base, wherein the knowledge base operation unit reads the knowledge data from the knowledge base. The read knowledge data is loaded as knowledge data described as a variable structure, and the knowledge data stored based on the inference result of the inference unit is updated. 3. The knowledge base according to claim 1, further comprising: an operating unit; and a target knowledge update processing unit that controls an operation of the target knowledge operating unit in accordance with an inference result of the inference unit. 4. system. 4. The target knowledge update processing unit loads necessary knowledge data from the knowledge base to the target knowledge operating unit via the knowledge base operation unit at the start of use of the system, and at the end of use of the system. The knowledge base system according to claim 3, wherein the stored knowledge data of the target knowledge operating unit is stored in the knowledge base via the knowledge base operation unit.