JPH03296837A - Method for retrieving document - Google Patents

Abstract

PURPOSE:To reproduce the state of a data be obtained at an optional point of past time by successively updating a data base stored in a data base control means based upon operation history information obtained between the reproduction requiring time of the data base and its reproducing time and reproducing and retrieving the data base state of the reproducing time. CONSTITUTION:When a data base reproducing requirement is inputted from an input part 1 or a communication part 2, a requiring source program 84 starts a reproducing operation control part 82 to require reproducing. The control part 82 searches and retrieves the operation history information of an operation history information storing part 81 based upon the reproduction requirement and finds out the past time for reproducing. The control part 82 retrieves the operation history information in accordance with the reproduction requirement and determines the reproducing time. In addition, the control part 80 operates a data base managing part 80 in accordance with the operation classifications of the history information and reproduces the data base state successively in the past direction up to the objective reproducing time.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a document retrieval method for user-friendly retrieval of document information in a document retrieval device that stores and retrieves document information.

(Prior art) In a document retrieval device that stores and retrieves document information, 1
Document information is usually managed in a database mechanism within one document search device using the document body and management information such as key information such as headings and link information between key information. The management information of this database mechanism is updated every time one document body or key information is added, changed, or deleted, and is always kept up-to-date.

When searching for document information, the search operation is performed based on the management information of the database mechanism at the time of the search.
Desired document information can be obtained.

(Problem to be solved by the invention) In this way, the conventional sentence @kenken old 1. In two months, the database structure changes due to additions, changes, deletions, etc.

Even though the database changes and grows over time, searches are performed based on the structure of the database at the time of the search.
There was an inconvenience that a search could not be performed if the document information accumulated in the past (including key information such as headings, not the text of the document itself) was not in the database at the time of the search. In addition, even if it exists in the database at the time of the search, if you forget which heading you put it in, you will not know the situation at the time of entry, so it will be difficult to remember and you will waste extra time searching. There was a gun between them.

(Object of the Invention) The object of the present invention is to solve the drawbacks of the conventional document retrieval methods described above, and to solve the problems of the conventional document retrieval methods described above.
Second, it provides an effective search method for users.

(First stage for solving the problem) In order to achieve the second object, the present invention aims to accumulate document information,
In the document retrieval device r that performs the search, there is a database that manages the data structure of the document information and the files on the storage device in which the document information is stored, and executes update and search processing in response to update and search requests. It has a management means, and when an update request is made to add, change, or delete the document information, 1. The operation history information is stored in chronological order according to the update operation contents, and further, when the update request is a deletion request, deletion information such as treatment status is created and registered in the database management means, Using the contents of the operation history information stored at the time of the database reproduction request as a search key, the operation history information is searched to determine four points in time for reproducing the database. The database of the database management means is sequentially updated based on the operation history information during the period, the current database state is reproduced again in step 7, and the database management means searches the reproduced database state.

(Function) The present invention stores M-history information such as operation requests to the database every time the structure of the database is updated due to addition, change, or deletion of document information, and specifically responds to the document information to be deleted. For example, delete the main document with a large amount of data, and instead store processing status information with a small amount of data (save destination, reason for deletion, etc.) as historical information, and then delete the document itself at a specified past point in time. Data amount/space status ma,
” - The feature is that it is reproduced based on historical information and searches are performed using the state of the database at that time.

Therefore, if document information accumulated in the past (including not only the document itself but also key information such as headings) does not exist in the current database, the treatment status can be restored by reverting to a past point in time and reproducing the database. The reason for deletion, storage location, etc. can be determined by attribute information such as information.

Furthermore, even if the document information exists in the current database, if you forget the filing destination, there is an advantage that the past state can be reproduced, making it easier to remember the situation at the time of filing.

Furthermore, since the method of the present invention allows past memories to be preserved without being forgotten, they can be used to look back on the past, making it more familiar.

(Embodiment) FIG. 1 is a block diagram showing an example of a document retrieval device for explaining an embodiment of the method of the present invention. J in the figure is a mouse,
2 is an input unit that has input devices such as keyboards, scanners, and tablets, and code conversion and encoding processing functions; 2 is a communication unit that exchanges document information with external devices such as information centers, personal computers, and facsimiles; and 3 is a communication unit that exchanges document information with external devices such as information centers, personal computers, and facsimiles. A display section for displaying search results, etc.; 4 a large-capacity storage section for storing document information, etc. input from the input section 1 and the communication section 2; 5, for displaying document information on the display section 3, 1 document information; An image processing section having a function of performing image processing such as enlargement, reduction, scrolling 9 encoding/decoding processing at high speed; 6 a bus for transferring document information; 7 a transfer control section controlling transfer on the bus 6; ,
Reference numeral 8 denotes a control unit that controls the entire apparatus and controls the execution of the two methods of the present invention.

FIG. 2 is a diagram showing an example of the data structure of document information managed by the database mechanism of the control unit 8 shown in FIG. do.

FIG. 3 is a diagram showing the operational flow of the method of the present invention in the database mechanism of the control section 8 shown in FIG. 1, and a detailed explanation thereof will be given later in the explanation of the operation.

FIG. 4 is an example of a functional block configuration diagram of the database mechanism of the control unit 8 for explaining the present invention in detail.
In the figure, 1 to 4 and 8 are respectively the input section 9 communication section 9 display section, storage section, and control section in FIG. 1, 80 to 84 are functional blocks in the control section 8, and 80 is the data structure of document information. This is a data pace management unit that manages files on the document information storage unit and executes update and search processing in response to update and search requests. An example of the configuration of a table for managing correspondence is shown. Reference numeral 81 denotes an operation history information storage section that stores operation history information in chronological order according to update operation contents when updating document information such as addition, modification, or deletion. FIG. 6 shows an example of the operation history information.

82 collects information stored in the operation history information storage section 81;

When requesting to reproduce the database, the operation history information storage unit 81 is searched using the contents of the operation history information as a search key to determine the current point in time of the database, and then the operation history information storage unit 8
1 and a reproduction operation control unit that operates the data pace management unit 80 to reproduce the current database again; 83 a deletion processing unit that operates the data pace management unit 80 to perform deletion processing for attribute information in a deletion request; and 84 The database management unit 80 receives the document information operation request from the input unit 122 communication 2 in FIG. The reproduction operation control unit 82 is operated and executed.

This is a request source program that returns the response result to the display section 3 or the communication section 2.

The document retrieval operation according to the method of the present invention will be explained below with reference to FIGS. 1 to 6 above.

Document information is input from the input section 1 or the communication section 2 in FIG. The data is decoded and displayed on the display section 3. Note that the types of document information may include a code format document 9 handwritten document, image data, and color images (still images, moving images).

The range to be stored is specified using a pointing device such as a mouse or tablet in the input section 1, and key information such as headings and titles and other attributes as shown in Figure 2 are specified using a document input device such as a keyboard or tablet. Information (for example, where the document information is obtained) is added and registered in the database of the control unit 8. Note that if the desired heading already exists, the document is registered under that heading.

In addition to the above-mentioned registration process in the database, the control unit 8 starts the transfer control unit 7, reads out the document information in the area specified above from the display unit 3, performs compression processing in the image processing unit 5, and stores it in the storage unit. 4 for high-speed transfer and storage.

This storage unit 4 stores not only the document itself but also various database management information for backup purposes. Document information is sequentially added to the database by repeating the above operations.

Furthermore, in response to a request from a user, not only the above-mentioned additions but also changes and deletions are made, and the structure of the database is updated. That is, in database processing in the control unit 8, operation history information is acquired in response to database update requests for additions, changes, and deletions in order to enable the database structure to be returned to any point in the past. Below, the operation history information acquisition operation in the control unit 8 will be explained in Figs.
This will be explained according to FIG.

First, a method of managing document information in a database, which is necessary for explaining the history information acquisition operation, will be explained. Document information in the database is managed using a data structure as shown in FIG.

In the example in Figure 2, document information is managed in a three-level hierarchical structure (Heading 1. Heading 2. Title), and the nodes (circles in Figure 2) corresponding to the key information in each hierarchy are It is managed using the corresponding attribute information (square marks in Figure 2) and link information between notes (links between dark layers, links within the same layer). Regarding the title node,
The attribute information consists of the document itself and other attribute information (for example, where the document was obtained).

Each node in each hierarchy is given a node number that corresponds one-to-one with the node 'A1. Ru. In this way, the data structure is as follows: - For boat fishing, once three pieces of information are determined for each node: the node number, the corresponding attribute information, and the ring information (node number and link type linked to the node) Established and added document information.

Data structure modification operations such as modification and deletion are performed based on the above three pieces of information.

In addition, attribute information (including the document itself) is normally stored in the human-image storage device to prevent damage caused by power outages to the document retrieval system.Securing the storage area on the storage device 9
File management for M broadcasts, etc. is done in the same way as on ordinary computers, workstations, etc.
Management is performed using file numbers assigned to each file on a one-to-one basis.

- As mentioned above, attribute information needs to be associated with a storage location on a storage device, and this is done by associating a node number with a file number. The node number and the file number are not necessarily one pair] - For example, in the case of a title node, the file number of one document body and the file number of other attribute information correspond to one node number.

In this way, multiple file numbers correspond to one note number, so in addition to the node number 50 and file number 51, the type 52 of attribute information (document body, other The database management described above, which simultaneously manages attributes, etc.), is performed by the database management unit 80 shown in FIG.

Hereinafter, based on the database management method described above, an operation for acquiring history information necessary for reproducing a past database will be explained in accordance with FIGS. 3 to 6.

When an update request for adding, changing, or deleting document information or a search request is input from the input section shown in FIG. 300 in the figure, the request source program 84 of the control unit 8 starts the database management unit 80 and adds or deletes files when it is necessary to add or delete files in accordance with attribute information due to an update request. .

In particular, in the case of a deletion request, the request source program 84 starts the deletion processing unit 831, creates selection 1 of attribute information to be retained as history information, starts the database management unit 80, and creates selection 1 of attribute information. The request source program 84, which adds and deletes files that occur with processing, starts up the database management unit 80 again and performs database processing according to the request type. The operation will be explained in detail for each request type below.

In the case of a search request (301 to 303 in FIG. 3), the request source program 84 starts the database management unit 80, searches the database (301 in FIG. 3), and returns search results according to the search type. obtain. Furthermore, the database management unit 80 is started again after the document information file number 51 of the search result is read out from the storage unit 4 (302 in FIG. 3) and output to the display unit 3 and the communication unit 2. (303 in Figure 3).

For example, in the case of an inspection request to retrieve the document information of all nodes linked to the lower hierarchy of the node specified by the user, the database management unit 80 selects the original node number 50 corresponding to the node. Search for all nodes that are linked in the hierarchy below the node number, and find the file number 51. corresponding to the node number (there may be more than one) of the corresponding node. The attribute information type 52 is obtained and sent back to the request source program 84.The request source program 84 then receives the file number.

The database management section 80 is activated again based on the attribute information type, and the attribute information is read out from the storage section 4 and output to the display section 3 and the communication section 2.

The example described above is an example of the search type, and there are various types of search requests, and the database management unit 80 performs various searches in accordance with each request. In the case of the search request described above, the structure of the database does not change, so history information acquisition processing is not performed.

In the case of an addition request (311 to 313 in FIG. 3), there are three types of addition: addition of a node, addition of new attribute information, and addition of link information, and the operations are different for each.

First, in the case of adding a node, the requesting program 84 uses the database management unit 8 before starting database processing.
0 to store the attribute information (including the document body) corresponding to the additional node in the storage unit 4 (311 in FIG. 3), and the file number 51 and attribute information type 52 of the attribute information shown in FIG. obtain. In addition, the request source program 84 obtains the node number and link type of the addition destination based on the instruction information from the input unit 1 or the communication unit 2, and sends it to the database management unit 80 as input information along with the file number and attribute information type. Request to add a node to update the data structure and update the fifth node.
The correspondence table between the node number 50 and file number 51 shown in the figure is updated, that is, added to the database (312 in FIG. 3).

When the update of the current database is completed as described above, the request source program 84 starts the reproduction operation control unit 82 and collects operation history information necessary for restoring the database before addition or returning to the state after addition.

The reproduction operation control unit 82 stores the collected operation history information in chronological order in the operation history information storage unit 81 (313 in FIG. 3). Operation history information in the case of node addition includes the type of addition (node addition), node number, attribute information file number, and attribute information type.

The destination node number and link type are required.

Of these, if you only want to reproduce the state before addition, you can memorize only the node number and issue a deletion request to the database management section 80 using this node number, but - once you have reproduced the state before addition, In order to reproduce the state after addition again, it is necessary to store the information required in the above-mentioned addition process, and the above-mentioned information is required.

When adding attribute information, 5 as in the case of adding a node.
Before database processing is started, the request source program 84 starts the database management unit 80, stores the attribute information to be added in the storage unit 4 (311 in FIG. 3), and stores the attribute information file number and attribute information type. get.

In addition, the request source program 84 obtains the node number to which attribute information is added based on the instruction information from the input unit 1 or the communication unit 2, and receives the attribute information as input information along with the attribute information file number and attribute information type. request to be added to the database management unit 80, add the node number, attribute information file number, and attribute information type to the node number and file number correspondence table, and update the database (third
312 in the figure).

The reproduction operation control unit 82 collects additional type (addition of attribute information), node number, attribute information file number, and attribute information type as operation history information, and stores them in the operation history information storage unit 81 (third 313 in the figure).

In the case of adding link information, the request source program 84 performs the following steps.
Based on the instruction information from the input unit 1 or the communication unit 2, obtain the two node numbers and link types to which links are to be added, request the database management unit 80 to add link information, and update the data structure ( 312 in Figure 3).

In the reproduction operation control unit 82, as operation history information.

Addition type (add link information), Add link 2
The node number and ring type are collected and stored in the operation history information storage unit 8 (313 in FIG. 3).

In the case of a deletion request (321 to 324 in FIG. 3), there are deletion types such as node deletion, attribute information deletion, and link information deletion, and the operation is different for each, as in the case of an addition request.

Further, in the case of a deletion request, the requesting program 84 activates the deletion processing unit 83 and creates a selection 9 of attribute information to be retained as history information. In the case of deleting a node, the requesting program 84 selects the node number 50. to be deleted based on the instruction information from the input unit 1 or the communication unit 2. Attribute information file number 51. Obtaining the attribute information type 52, the deletion processing unit 8
Start 3.

The deletion processing unit 83 determines which attribute information should be retained as historical information based on the amount of data, importance, etc., and creates deletion information (321 in FIG. 3).
.

In addition, for example, in order to effectively use the storage unit 4, attribute information with a large amount of data such as the document body is deleted, and instead of processing status information (storage of document information to be deleted). light information, reason for deletion, etc.)
To create information that is effective as historical information and requires a small amount of data.

As a result of the determination, the database management unit 80 is activated to delete the horse information group file to be deleted, and the database management unit 80 is activated to store the newly created deletion information in the storage unit 4. 322 in FIG. 3), the file number is obtained.

Then, in the deletion processing unit 83, the node number 50. File number 51 of deletion information to be kept as history information. Attribute information type 57. Enter Yuwado L2 and go to database management section 8 again.
A node deletion request is made to 0, the data structure is updated, and the response information is used to obtain link information (link destination node number, link type) for the node connected to the deleted node.

In addition, all the file numbers 51 corresponding to the node number 50 are deleted (323 in Figure 3), and the node number, file number and attribute information type 5z of the deletion information to be left as ff history information are added to the node number. and file number mapping shell.

When the deletion processing for the current database is completed as described above, the request source program 84 starts the reproduction operation control unit 82 and collects operation history information necessary for restoring the database before deletion or returning to the state after deletion.

The reproduction operation control unit 82 stores the collected operation history information in 5 chronological order in the operation order layer information storage unit 81 as in the case of an addition request.
(324 in FIG. 3).

Here, as operation history information in the case of node deletion, the type of deletion (seed deletion), node number, file number of deletion information to be left as history information, and link information are stored.

In addition, in the case of deleting attribute information, one request source program 84 uses instruction information from the input section 1 or the communication section 2 to specify the node number 50.50 corresponding to the attribute information to be deleted. File number 51. Obtain the attribute information type 52.

The deletion processing unit 83 is activated. In this deletion processing section 83,
Delete the file number 51 corresponding to the attribute information to be deleted.
- Delete the file by giving it to the database management section 80.

Note that the attribute information to be deleted does not remain as historical information.
If there is a new attribute information shrine, new deletion information is created, the database management section 80 is activated, the deletion information is stored in the storage section 4, and a file number is obtained (321 in FIG. 3).
.

The deletion processing unit 83 sends a request to the database management unit 80 to delete the attribute information using the 5 node number, the file number for the attribute information to be deleted, the file number of the deletion information to be kept as history information, and the attribute information type as input information. Then, delete the correspondence between the node number and the file number for the attribute information to be deleted (323 in Figure 3), and if there is newly created attribute information, the node number 6H history information and 17
Add the file number and attribute information type of the deletion information to be left to the node number and file number correspondence table.

When the above deletion process is completed, as in the case of node deletion, the request source program 84 starts the reproduction operation control unit 82, collects operation history information [J, and stores it in the operation history information storage unit 81 ( 324 in FIG. 3), as operation history information in the case of attribute information deletion, the deletion type (attribute information deletion), node number, file number of the deletion information to be left as history information, and attribute information type are stored. . In the case of deleting link information, the requesting program 84 obtains the node numbers of the two nodes to be engraved on the link to be deleted based on the instruction information from the input unit 1 or the communication unit 2, and then sends the node numbers to the deletion processing unit 8.
Start 3.

Since the deletion processing unit 83 does not need to delete attribute information, it does not perform the deletion processing as in the case of deleting a node or deleting attribute information described above, but responds to the link to be deleted given by the request source program 84. The database processing unit 80 is requested to delete the link information using the node numbers of the two nodes as input information, and the link is deleted (323 in FIG. 3).
).

When the above deletion processing is completed, as in the case of node deletion, the request source program 84 starts the reproduction operation control section 82, collects operation history information, and collects operation history information from the operation history information storage section 8.
1 (324 in FIG. 3).

As the operation history information in the case of link information deletion, the type of deletion (deletion of link information), two node numbers corresponding to the link to be deleted, and the link type are stored (324 in FIG. 3).

In the case of a change request (331 and 332 in Figure 3), the change types include attribute information changes and link information changes, but changes can be made by successively executing the above-mentioned deletion request operation and addition request operation. processing can be realized.

That is, in the case of the former deletion request operation, in the request source program 84, by sequentially executing the operation of requesting deletion of the attribute information before the change and requesting addition of the attribute information after the change (331 in FIG. 3), For the latter additional request behavior,
The change operation is performed by successively executing operations for requesting deletion of the link information before change and requesting addition of link information after change (332 in FIG. 3). Therefore, the operation history information in the case of a change request is not specially collected, but the operation history information collected by each of the deletion request operation and the addition request operation is used.

As described above, the operation history information required to reproduce the database differs depending on the type of update request and the target of the update operation. In particular, change requests can be realized by combining deletion requests and addition requests.

As shown in FIG. 6, the operation history information includes - For boat fishing, operation type 61. Information related to data structure (node number, link information) 62, information related to attribute information 63
(file number, attribute information type) and other information 6
The required information differs depending on the type of update request and the target of the update operation.

In addition, other information such as the update date and mark (a special mark is given to the update operation) will be used in order to search for operation history information, as described later, in addition to the above information necessary to reproduce the database. This information is used as key information.

As described above, the operation history information acquisition operation is performed. Next, the reproduction operation of the database will be explained.

In FIG. 1, when a user wants to reproduce and search a past database, he inputs search information of the past point in time he wishes to reproduce from the input section 1 or the communication section 2, and requests reproduction of the database.

In the control unit 8, based on the input search information.

The above-mentioned operation history information (61 to 64 in FIG. 6) is searched to find the past time point to be reproduced, and the operation history information is sequentially traced in the past direction to reproduce the database.

When a user requests a normal menu search, keyword search, etc. through the input unit 1 or the communication unit 2, the control unit 8 performs a search operation in the reproduced database state, reads the results from the storage unit 4, Output to display section 3 or communication section 2.

The database reproduction operation in the control section 8 described above will be explained in detail with reference to FIGS. 3, 4, and 6.

When a database reproduction request is input from the input unit 1 or the communication unit 2, the request source program 84 starts the reproduction operation control unit 82 and issues a reproduction request (340 in FIG. 3).
. The reproduction operation control unit 82 searches the operation history information (FIG. 6) in the operation history information storage unit 81 based on the reproduction request to find the past point in time to be reproduced (FIG. 3). 341).

The types of reproduction requests are (a) when reproducing operation history information sequentially like flipping through the pages, (b) when specifying a specific operation type and reproducing the point in time, (c) when reproducing past dates. (d) to reproduce the point in time when a specific mark was added when acquiring the operation history information, etc. By using this information as key information, various reproduction requests can be made.

The reproduction operation control unit 82 searches the operation history information in response to the reproduction request and determines the re-current point (341 in FIG. 3).

Furthermore, the operation history information from the present time to the next time is sequentially read from the operation history information storage section 81, and the database management section 80 is operated according to the operation type of the history information, so that the state of the database is sequentially pasted up to the desired return time. (342 in FIG. 3).

If the operation type in the operation history information is an addition operation, a deletion request is sent to the database management unit 80 to reproduce the state before the addition, and if it is a deletion operation, an addition request is sent to the database management unit 80. to reproduce the state before deletion.

When the current state of the database can be reproduced by repeating such operations, the reproduction operation control unit 82 notifies the request source program 84 of the result, and the request source program 84 displays the completion of database reproduction on the display section. 3 or the communication unit 2 to inform the user (34 in Fig. 3).
4). The user performs a search operation in the state of the database at the time of migration and obtains desired information (345 in FIG. 3).

Furthermore, when a request from the user to reproduce another point in time, including a return to the current point, is received from the input unit 1 or the communication unit 2,
The request source program 84 starts up the reproduction operation control section 82 again.

Based on the reproduction request, the reproduction operation control unit 82 searches the operation history information in the operation history information storage unit 81 to find the point in time at which reproduction should be performed (341 in FIG. 3).

Note that if the reproduction request is to return to the current point, no search processing is necessary. If the re-current point is in the past than the point at which it has already been reproduced, the transition to the re-current point is performed in the same manner as in the two-part method (342 in FIG. 3).

If the current point again is closer to the current point than the point at which it has already been reproduced, the operation history information is sequentially read in the current direction up to the point in time again, and the reproduction operation is performed according to the operation type (3rd step).
343 in the figure). That is, if the operation type is an addition operation, an addition request is sent to the database management unit 80 to reproduce the state after the addition, and if the operation type is a deletion operation, a deletion request is sent to the database management unit 80. When the reproduction operation is completed again by reproducing the state after the deletion, the user is notified of the process and the search operation is performed in the same way as described in the first reproduction operation (344 in Figure 3).゜3
45).

(Effects of the Invention) As explained above, according to the present invention, even if the database is updated daily by repeatedly adding, changing, and deleting document information (including headings), Since the state of the delta base can be reproduced, past memories such as the location of previously stored document information can be easily searched and remembered, and there is an advantage that a document retrieval method that is more familiar to people can be provided.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an example of a document retrieval device for explaining an embodiment of the method of the present invention, and FIG. 2 is a diagram showing an example of the data structure of document information managed by the control unit 8 in FIG. ,
FIG. 3 is a diagram showing an example of the operational flow of the method of the present invention in the control section 8 of FIG. 1, and FIG. Figure 5 is a diagram showing an example of the configuration of a correspondence table between node numbers and file numbers, and Figure 6 is the operation history information of Figure 4! ! Il! It is a figure showing an example of operation history information stored in storage part 8. 1...Input section, 2...Communication section, 3...Display section, 4...Storage section, 5...Image processing section,
6... Bus, 7... Transfer control unit, 8... Control unit, 80... Database management unit 581... Operation history information storage unit, 82... Reproduction operation control unit, 83... - Deletion processing unit, 84...Request source program.

Claims (1)

[Claims] In a document search device that stores and searches document information,
It has a database management means that manages the data structure of the document information and the files on the storage device in which the document information is stored, and executes update and search processing in response to update and search requests, and adds the document information. When an update request for , change, or deletion is made, operation history information is stored in chronological order according to the update operation contents, and further, when the update request is a deletion request, deletion information such as treatment status is created and the database management is performed. The content of the operation history information stored at the time of the database reproduction request is used as a search key to search for the operation history information to determine the current point in time of the database. A document characterized in that the database of the database management means is sequentially updated based on operation history information between the present time and the present time, the database state at the present time is reproduced, and the database management means searches the reproduced database state. retrieval method.