JPH0612457A - Data base device - Google Patents

Abstract

PURPOSE:To reduce the burden on a user to perform retrieval at a high speed by adopting the access history of a data base device as relativity of each data by the user and describing the degree of relativity as the position in a multi- dimensional virtual space of data. CONSTITUTION:A retrieval means 108 receives character data transferred through an information transfer path 107 and requests write or read of objective data to an auxiliary storage means 105 through information transfer paths 109 and 111. At the time of new data generation of this write, a multidimensional attribute is added by a describing means 110. That is, default data in the same dimensions as already prepared data is added when an instruction is not especially given, but attribute data obtained by shifting contents of the attribute of similar data out of already prepared data by one quantum stage is added when the instruction is given.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to improving the search performance of a database device, and more particularly to a database device which is convenient when performing a search based on the relationship between the data.

[0002]

2. Description of the Related Art As disclosed in Japanese Patent Application Laid-Open No. 2-202674, there is known a database system for automatically performing a search or a classification according to a database access and sequentially displaying the results of the search or the classification. The database
It is considered that the system improves the easiness of database operation, that is, retrieval of target data, by the automatic update.

[0003]

In the above conventional example, when the database is accessed, the database structure is adapted by performing a search or a classification as needed to improve the ease of data search. There is.

However, the search or classification is the presence or absence of the target data or the fitting of the target data to the matching classification, and there is a possibility that the past access history is not fully utilized for the adaptation.

Therefore, even if the database is added, changed, or deleted by the access, all the data in the target classification in the database have the same priority, and when the user's memory is ambiguous, etc. The usability may not be improved by the random search that is supposed to be performed and the subsequent neighborhood search.

An object of the present invention is to improve the problems of the above-mentioned conventional example and to improve the ease of ambiguous search of target data.
By storing parameters indicating the relationship between the multidimensional data so that they are arranged in the distance and the direction of the defined difference in the multidimensional virtual space according to the approximate correlation degree with the known data of contents. It is possible to perform an order search and a plurality of in-condition searches (hereinafter abbreviated as grasp search) according to the degree of approximate correlation with known data of the above contents.

[0007]

In order to solve the above problems and improve the ease of information retrieval by the database device, a descriptor expressing the arrangement of each data in a multidimensional space is added to the data. Described in the association auxiliary storage device,
A description means is provided.

[0008]

The description means associates the classification of the respective data and the relation between the respective data with the arrangement in the multidimensional space.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment of the present invention will be described below with reference to FIGS. 1, 2, 3 and 4. 1 is a block diagram of the database device of the present invention, FIG. 2 is a diagram showing a data image of the database device of the present invention, FIG. 3 is a diagram showing an image of a virtual data arrangement space of the database device, and FIG. 4 is a diagram of the present invention. It is a flowchart which shows operation | movement of a database apparatus.

First, the configuration and operation outline of the database device of the present invention will be described with reference to FIG. As shown in FIG. 1, the database device according to the present invention is a keyboard 10
6, search means 108, description means 110, auxiliary storage means 1
05, a video RAM (hereinafter abbreviated as VRAM) 103, and a display 101 as main components.

The keyboard 106 is an input man-machine interface capable of converting character input into character data and outputting the character data.

The retrieving means 108 has a function of reading and accessing suitable data in accordance with a keyword retrieval instruction by inputting character data, and adding / changing / data.
It is an information control system having a function of writing access to target data according to a deletion instruction.

The description means 110 is an information processing system for adding multidimensional attributes to the write data. When there is no particular instruction, the description means 110 adds default data of the same dimension as the existing data, and when there is an instruction, it already exists. Attribute data obtained by shifting the attribute content of similar data among data by one quantum step is added.

The auxiliary storage means 105 is a memory capable of outputting data in accordance with access to specify data, and has a capacity necessary for constructing a target database.

The VRAM 103 is a memory that includes a display control unit that expands an image of a character string indicated by character data input into a memory as a bit image, and generates a display signal for display of the contents of the expansion. It has a capacity capable of storing one or more surfaces of a binary bit image of a size.

The display 101 is an output man-machine interface for reading a display signal indicating display information and generating a target image.

Next, an outline of the operation of the database device according to the present invention will be described with reference to FIG.

Assuming that the user starts the database device according to the present invention and starts inputting information, the input from the keyboard 106 is passed as character data through the information transfer path 107 to the searching means 108. .

The search means 108 uses the information transfer path 1
Receives character data passed through 07,
The information transfer path 10 is used for writing or reading target data.
Request is made to the auxiliary storage means 105 via 9, 111.

Here, in the writing, when the new data is created, the description means 110 adds the multidimensional attribute.

When it is necessary to read the target data, the target data in the auxiliary storage device 105 is loaded into the VRAM 103 via the information transfer path 104 and then displayed via the information transfer path 102.

Next, the structure of the data image and the functional outline of the database device of the present invention will be described with reference to FIG.

As shown in FIG. 2, the data image is
A header 201 and data 202.

The header 201 is a part for supplementarily defining the content indicated by the data.

The data 202 is a data stream in which data is continuously read out for a certain length by a decoder set by the contents of the header 201.

The header 201 has the X-coordinate value 203, the Y-coordinate value 204, and the Y-coordinate value 204 of the multidimensional attribute, that is, the virtual space.
A region in which the z coordinate value 205 can be set is provided, and the attributes, that is, the X coordinate value 203, the Y coordinate value 204, and the z coordinate value 20 are set.
5 can be set to multiple values by securing areas each having a plurality of bits or more.

Next, with reference to FIG. 3, an image configuration of the virtual data arrangement space of the database device of the present invention and an outline of functions will be described.

As shown in FIG. 3, the image of the virtual data arrangement space is a finite space of a three-dimensional Cartesian coordinate system here, and the arrangement position is expressed by X coordinate, Y coordinate and Z coordinate.
The six rectangular parallelepipeds scattered in the finite space represent data.

Here, if it is assumed that the hatched objects are accessed during a certain period of access, it is determined that they are related information, and the positions of the hatched objects are respectively indicated by the arrows, that is, Parallel movement is performed for a certain distance in the position direction of the center position of each coordinate of the shaded data.

The X-coordinate, the Y-coordinate, and the Z-coordinate are used to describe the arrangement position again. To explain again, the average value for each coordinate value of the header 201 of the shaded data accessed during a certain period, and the respective data. The middle point of the arrangement position and the arrangement position is obtained, and the arrangement position is replaced with the position of the middle point.

Next, the operation of the database device of the present invention will be outlined with reference to FIG.

As shown in the flow chart of FIG. 4, the database device of the present invention first branches the operation depending on whether it is a search or maintenance after starting the operation, and further branches according to the search type in the case of the above search. If the search type uses the relation between the data, GRASP of area designation
Or a search in which the designated coordinates or the data closest to the designated data is read in order, and in the case of GRASP, the center position and radius of the virtual sphere indicating the range are read in order from the closest data. Specify the position or data.

The GRASP collates the arrangement positions in the header of all the data with the arrangement positions in the virtual sphere, and outputs and displays the satisfied data.

In the retrieval in which the closest data is read in order, the distances between the positions of all the data in the header and the distances between the positions or the positions of the data are calculated, and the distances are sorted in ascending order, and the sorting is performed if necessary. Output and display the corresponding data from the beginning of the result.

Here, the arrangement position of the data accessed during the series of searches is moved to the center position of the arrangement position of the data by a certain distance.

When the branch is maintenance, the target data is designated, and the target data is added or changed or deleted.

If the operation is continued after the target access is completed, the process is repeated from the beginning of the flowchart.

A second embodiment of the present invention will be described below with reference to FIG. FIG. 5 is a diagram showing an image of the virtual data arrangement space of the database device.

The configuration of the image of the virtual data arrangement space of the database device of the present invention and the functional outline will be described with reference to FIG.

As shown in FIG. 5, the image of the virtual data arrangement space is a finite space of a three-dimensional Cartesian coordinate system here, and the arrangement position is expressed by X coordinate, Y coordinate and Z coordinate.
The six rectangular parallelepipeds scattered in the finite space represent data.

Here, if an object with diagonal lines is accessed during the access for a certain period of time, it is judged that the white-painted object is irrelevant information, and the arrangement position of the white-painted object is indicated by the arrow. Translate to a certain distance.

The X-coordinate, the Y-coordinate, and the Z-coordinate will be described again in accordance with the arrangement position expression. The starting point is the average value of each coordinate value of the header 201 of the unpainted white-painted data during a certain period. A vector having an arrangement position of each data as an end point is obtained, and when the start point of the vector is fixed and the length is doubled, the arrangement position is replaced by the position indicated by the vector end point.

A third embodiment of the present invention will be described below with reference to FIG. FIG. 6 is a diagram showing an image of the virtual data arrangement space of the database device.

The configuration of the image of the virtual data arrangement space of the database device of the present invention and the functional outline will be described with reference to FIG.

As shown in FIG. 6, the image of the virtual data arrangement space is a finite space of a three-dimensional Cartesian coordinate system here, and the arrangement position is expressed by X coordinate, Y coordinate and Z coordinate.
The six rectangular parallelepipeds scattered in the finite space represent data.

Here, if an object with diagonal lines is accessed during a certain period of access, it is determined that the objects with diagonal lines are related to each other, and the white-painted objects are considered to be unrelated information. Judgment is made and the positions of the hatched and white-painted objects are moved in parallel in the directions indicated by the arrows by a certain distance.

The X-coordinate, Y-coordinate, and Z-coordinate are used to describe the arrangement position again. To explain again, the average value of each coordinate value of the header 201 of the shaded data accessed during a certain period is used as a starting point, and The middle point of the arrangement position of the data is obtained, and the arrangement position is replaced by the position of the middle point, and the average value for each coordinate value of the header 201 of the unpainted white-painted data during a certain period is A vector having the starting point as the starting point and the ending point of the arrangement position of each data is obtained, and when the starting point of the vector is fixed and the length is doubled, the arrangement position is replaced by the position indicated by the vector end point.

The fourth embodiment of the present invention will be described below with reference to FIG. FIG. 7 is a diagram showing an image of the virtual data arrangement space of the database device. Here, the upper diagram is shown in FIG.
a, the lower left figure is FIG. 7b, and the lower right figure is FIG. 7c.

The configuration of the image of the virtual data arrangement space of the database device of the present invention and the functional outline will be described with reference to FIG.

As shown in FIG. 7a, the image of the virtual data arrangement space is a finite space of a two-dimensional Cartesian coordinate system, and the arrangement position is expressed by X and Y coordinates. And three rectangular parallelepipeds scattered in the finite space represent data.

Here, if an object with hatching is accessed during a certain period of access, it is determined that the object with hatching is related to each other, and a white-painted object is considered to be unrelated information. Judgment is made and the positions of the hatched and white-painted objects are moved in parallel in the directions indicated by the arrows by a certain distance.

The X-coordinate and the Y-coordinate are used to describe the arrangement position again. To explain again, the average value of each coordinate value of the header 201 of the shaded data accessed during a certain period is used as a starting point, and each data has it. The middle point of the arrangement position is obtained respectively, and the arrangement position is replaced by the position of the middle point, and the white-painted data that has not been accessed during a certain period is the average value for each coordinate value of the header 201 of the previous data. Is obtained as a starting point, and a vector having an arrangement position of each data as an end point is obtained, and when the starting point of the vector is fixed and the length is doubled, the arrangement position is replaced by the position indicated by the vector end point.

Here, in the two-dimensional Cartesian coordinate system as shown in FIG. 7A, when the data with the slanted lines are brought close to each other in the virtual space, the data of the white paint may be brought close to the data. This is because the position description in the finite n-dimensional space cannot independently describe parameters with a finite scalar number n or more. Therefore, in that case, the dimension of the space is increased by one dimension, the space description is changed as shown in FIGS. 7a to 7b, and a new Z coordinate is defined. Here, the Z coordinate is divided by 2 when the number of unrelated sets, here the shaded data and the white-painted data, is divided into two, and the center point of each unrelated set is aligned with the midpoint of each region. Deploy. FIG. 7c shows an arrangement state in the YZ coordinate plane when the one-dimensional increase is made.

[0054]

Since the database device using the present invention is constructed and functions as described above, it has the following effects.

In addition to the conventional keyword search, the user describes the access history of the database device as the relevance of each data and describes the degree of the relevance as the position of the data in the multidimensional virtual space. It becomes possible to search based on the data relevance, which reduces the burden on the user and enables high-speed search.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a database device of the present invention.

FIG. 2 is a diagram showing a data image of the database device of the present invention.

FIG. 3 is a diagram showing an image of a virtual data arrangement space of the database device of the present invention.

FIG. 4 is a flowchart showing the operation of the database device of the present invention.

FIG. 5 is a diagram showing an image of a virtual data arrangement space of the database device of the present invention.

FIG. 6 is a diagram showing an image of a virtual data arrangement space of the database device of the present invention.

FIG. 7 is a diagram showing an image of a virtual data arrangement space in the same manner.

[Explanation of Codes] 101 ... Display, 102, 104, 107, 109, 111 ... Information transfer path, 103 ... VRAM, 105 ... Auxiliary storage means, 106 ... Keyboard, 108 ... Search means, 110 ... Storage means, 201 ... Header , 202 ... data, 203 ... X coordinate value, 204 ... Y coordinate value, 205 ... z coordinate value.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Tomohisa Kobinoyama 292 Yoshida-cho, Totsuka-ku, Yokohama, Kanagawa Stock, Ltd., Microelectronics Device Development Laboratory, Hitachi, Ltd. (72) Hiroyuki Sakai 292 Yoshida-cho, Totsuka-ku, Yokohama, Kanagawa House number Hitachi Ltd. Microelectronics equipment development laboratory (72) Inventor Junji Nakata 292 Yoshida-cho, Totsuka-ku, Yokohama, Kanagawa Prefecture Household number Hitachi Ltd. Microelectronics equipment development laboratory (72) Inventor Masami Yamagishi Yokohama, Kanagawa Prefecture Hitachi, Ltd. Microelectronics Device Development Laboratory, 292 Yoshida-cho, Totsuka-ku (72) Inventor Yuji Kimura 292 Yoshida-cho, Totsuka-ku, Yokohama, Kanagawa Inside the Device Development Laboratory (72) Inventor Taminori Tomita 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Ltd. Microelectronics Device Development Laboratory, Hitachi, Ltd.

Claims (7)

[Claims] 1. In a database device, a description means for associating a descriptor representing a classification of each data and a relation between the data with an arrangement in a multidimensional space with the data in an auxiliary storage device, A database device characterized by being provided. 2. The database device according to claim 1, wherein the arrangement position in the multidimensional space indicating the classification of each of the data and the relation between the data has a multilevel value. 3. The database apparatus according to claim 1, further comprising a learning unit that rewrites the descriptor according to access. 4. A converging / dispersing means for rewriting the descriptor so that the arrangement positions of the data accessed in a fixed period or in a series of operations in the multidimensional space are close to each other. The database device according to item 1 or 2. 5. An arrangement position in the multidimensional space of the data accessed within a fixed period of time or a series of movements, and a position in the multidimensional space of the data accessed outside of the fixed period of time or outside the series of movements. 3. The database device according to claim 1 or 2, further comprising a concentrating unit that rewrites the descriptor so that the arrangement positions of are separated from each other. 6. A converging / dispersing means for rewriting the descriptor so that the arrangement positions in the multidimensional space of the data accessed outside a fixed period or outside a series of operations are mutually discrete. The database device according to item 1 or 2. 7. The approach of the arrangement position in the multidimensional space of the data accessed within the certain period of time or the series of operations, or the data of the data accessed outside of the certain period of time or outside the series of operations. 7. The expansion means for increasing the parameter of the descriptor when it is difficult or impossible to disperse the arrangement positions in the multidimensional space, wherein the expansion means is provided.
The database device according to the item.