JPH02311958A - Data retrieving device - Google Patents

Abstract

PURPOSE:To efficiently retrieve a keyword by retrieving stored data, to which the designated keyword and a keyword in inclusive relation with the above mentioned keyword are added by using a retrieval table to control the keyword, which is added to the stored data, and a relation table to define the inclusive relation of the keyword. CONSTITUTION:A retrieval table 2 is prepared to control the keyword, which is added to the stored data, and a relation table 3 is prepared as an independent table or an integral table to define the inclusive relation between the keywords. By using these retrieval table 2 and relation table 3, the retrieval is executed by the designated keyword or the keyword in the inclusive relation. Accordingly, even when there is no consistency in the keyword and it is difficult to designate the suitable keyword or the keyword added to the data being a purpose is not clear, possibility for success to tightly close the data being the purpose, is increased only by designating the keyword with meaning to relate with the contents of the data which are the purpose. Thus, without repeatedly designating the keyword, the data, which are the purpose, can be efficiently retrieved.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a data search device that performs a keyword search for target data from a large amount of stored data.

BACKGROUND OF THE INVENTION Conventionally, this type of data retrieval device has a search table for managing one or more attribute keywords added to each piece of stored data, and by examining this search table,
It was configured to search for stored data that corresponds to a specified keyword.

Problems to be Solved by the Invention In such a keyword search, search results are obtained by pattern matching of specified keywords and keywords registered in a search table. This pattern matching uses prefix matching, middle matching, and
Since this is also done with suffix matching, it is possible to sequentially narrow down the target data without specifying the exact same keyword as the keyword added to the target data.

However, if the keywords are inconsistent, it is difficult to specify the appropriate keywords, and it is necessary to specify keywords many times to narrow down the desired data, which complicates the work and increases search time. Furthermore, in the worst case, there is a problem that it becomes impossible to narrow down the target data.

For example, when searching for map data, place names are generally included as keyword attributes. When registering a keyword for this place name, you can use "Kanto" or "Kanto" for the same map.
Keywords may be inconsistent, such as "Metropolitan area."

In this case, to search for a map with "Kanto" added, "
If you specify "Metropolitan Area", you cannot search for that map.

Free keywords that do not limit attributes are also used. This search using free keywords is important for searches based on the content of document data, but since consistency of free keywords cannot be expected,
Keyword searches are particularly susceptible to the problems described above.

The present invention has been made in view of the above-mentioned problems, and an object of the present invention is to provide a data retrieval device that can perform efficient keyword searches even when keywords are inconsistent and it is difficult to specify appropriate keywords. shall be.

Means for Solving the Problems In order to solve the above-mentioned problems, the present invention has, in addition to a search table for managing keywords added to stored data, a relation table that defines inclusive relationships between keywords. This system uses a table and a relational table to search for stored data to which a designated keyword and a keyword in an inclusive relationship therewith are added. Note that the relational table is either a table independent of the search table or a table integrated with the search table.

Effects of the present invention With the above-described structure, for example, if a superordinate concept, a subordinate concept, or a synonymous relationship between keywords is defined in a relation table in advance based on the meaning of the keyword according to the content of the data, it can be added to the target stored data. Even if you cannot specify a keyword correctly, if you specify a keyword with a meaning that corresponds to the content of the data you want to obtain, you can search using keywords that have a superordinate concept, a subordinate concept, or a synonymous meaning for this specified keyword. will be held. Therefore, even if you do not know the exact keyword added to the target data, there is a high possibility that you will succeed in narrowing down the target stored data just by specifying a keyword whose meaning is related to it, making data searches more efficient. can.

Embodiment FIG. 1 is a schematic diagram of a data retrieval device according to an embodiment of the present invention.

In FIG. 1, 1 is a data storage device that stores data, 2 is a search table that manages keywords added to each stored data in the data storage device 1, and 3 is a search table that manages keywords added to each stored data in the data storage device 1. This is a relationship table that defines inclusive relationships. 4 is a keyboard for the operator to perform input operations such as specifying keywords, and 5 is a display for displaying search results and the like. Reference numeral 6 denotes a search control device that performs keyword searches of stored data using the search table 2 and relational table 3, as well as the creation of the search table 2 and relational table 3, output control to the display 5, and the like. Search table 2 and relation table 3
is actually a table on a storage device that can be accessed by the search control device 6.

An example of the contents of the search table 2 in the case of map data is shown in FIG. Further, an example of the contents of relational table 3 is shown in FIG.

In the example of search table 2 shown in FIG.
N types of attribute keywords such as "place name" and "date" can be added to each stored data (record) to map data). The n-th keyword (n) is a keyword item for defining a subsumption relationship between keywords defined in relational table 3, and corresponds to the keyword salt shown in FIG. 4.

The contents of the relationship table 3 shown in FIG. 4 refer to the hierarchical relationship shown in FIG. 5 or 6.

FIG. 5 is a hierarchical diagram showing the subsumption relationships of the keywords defined in relational table 3 using keyword salts, and the keyword groups ``■'' to ``■'' have the following semantic subsumption relationships.

(1)■〉■,■ (2)■〉■,■,■ (3)■,■〉■ Figure 6 shows the subsumption relationship shown in Figure 5 with the subsumption relationship definition number (see Figure 4). FIG.

Search processing for the data search device configured as described above will be described below.

This data search device is capable of keyword searches using each of the keywords (]) to (n). for example,
In the case of a search using keyword (1), when the operator specifies "Japan", the search control device 6 compares the keyword 1) registered in the search table 2 with the specified "Japan" and displays the search result as a search result. Obtain the data of record (10). For keywords (2) to (n-1), the keyword search is the same as the conventional one, using only the similar search table 2.

On the other hand, the search process using the keyword (to) also uses the relationship table 3 and becomes a keyword search that also includes the subsumption relationship.

FIG. 2 is a schematic flowchart showing an example of the process. The processing details of this inclusive keyword search are as follows.

In step STI, the operator specifies the keyword (ω) from the keyboard 4. Here, it is assumed that rAAJ is specified. Also, the search table 2 and relation table 3 have the contents shown in FIGS. 3 and 4, respectively. Suppose that

In step ST2, the search control device 6 searches the relationship table 3 for the subsumption relationship definition number corresponding to the specified keyword rAAJ. In this case, rloooJ is found as the relevant subsumption relation definition number.

In step ST3, the search control device 6 stores the relationship table 3
Perform a prefix match search for the inclusion relationship definition number "i*motoichi" for . "Book" is the default. In this case, the search results include rAAJ, rAABJ, rAAcJ, rAAB
The rA of the keyword group ■ in Figure 5, such as DJ.
All keyword salts below AJ are obtained.

In step ST4, the search control device 6 performs step S
Each keyword salt obtained in T3 is compared with the keyword (b) registered in the search table 2, and a list of keywords for each record with a match is displayed on the display 5 together with the record number.

The keyword displayed at this time may be only the keyword ■ used in the search, or may be all or part of the keywords (1) to (n).

In step ST5, the operator specifies the target record number from the contents displayed on the screen or instructs to end the process. In the case of a termination instruction, the search control device 6 terminates the process.

If a record number is specified, the search control device 6 reads the data of the specified record number from the data storage device l and displays it on the display 5 in step ST6.

In this way, in this example, you can easily search not only records (lO) with the keyword "Japan" but also records with keywords such as "Hikari", "1 Honshu", "Kanto", etc. .

In other words, in conventional keyword searches, search results are obtained by matching keys of specified keywords.
It was not possible to perform a conceptual search according to the meaning system of the contents of the memory data. On the other hand, according to the present invention, it is possible to perform a systematic search based on the inclusive relationship of the meaning of the target data, such as, for example, Japan → Honshu 1> Kanto → Tokyo.

In addition, rAABDJ and rAADB shown in FIG.
Search using synonymous keywords is also possible by assigning the same subsumption relation definition number to keywords having a synonymous relation, such as j.

In addition, in the subsumption-related keyword search process explained above, a search is performed from the beginning using a specified keyword and a keyword with a meaning that has a subsumption relationship with it, but after performing a search using only the specified keyword, instructions from the operator Depending on the situation, a keyword search including subsumption relationships may be performed.

In addition, in the inclusion-related keyword search, the keyword (
If any keyword from 1) to keyword (n-1) is specified, search table 2 for keyword (n) for a record that matches that keyword, add this keyword (rl) to the specified keyword group, and perform step ST
Two or less processes may be performed.

Furthermore, the search table and the relationship table may be combined into one table by adding the subsumption relationship definition number as a keyword (n +1) to the search table shown in FIG.

The search table 2 and the relation table 3 are created by the search control device 6 through interactive processing with an operator during the work of adding keywords to stored data. For example, in the case of searching map data as explained in the embodiment, when keyword (1) of the place name attribute is specified, a hierarchical diagram as shown in FIG. It is generated from the content and displayed on the display 5. When a suitable keyword salt exists on the hierarchical diagram, the operator is made to specify the keyword salt, and when there is no suitable keyword salt, the operator is made to specify the subsumption relation and the keyword salt. Based on this designation, the search control device 6 registers the designated keyword salt in the keyword ■ of the search table 2. Also,
When a new keyword salt is specified, it is added to relation table 3, its inclusion relation definition number is generated, and relation table 3 is added.
Add to.

Effects of the Invention As is clear from the above description, the present invention provides a relational table that defines subsumption relationships between keywords as an independent table or an integrated table with a search table for managing keywords added to stored data. By preparing and using this search table and relation table, searches are performed using specified keywords and keywords that have a subsumption relationship, so it is difficult to specify appropriate keywords due to inconsistent keywords, etc.
Even in cases where the keyword added to the target data is unknown, if you specify a keyword with a meaning related to the content of the target data, the possibility of success in narrowing down the target data will increase, and it will be easier to narrow down the target data than before. This has the effect that the desired data can be efficiently searched without having to repeatedly specify keywords and example sentences.

[Brief explanation of drawings]

FIG. 1 is a schematic configuration diagram of a data search device according to an embodiment of the present invention, FIG. 2 is a schematic flowchart of keyword search processing including subsumption relationships, FIG. 3 is a diagram showing an example of the contents of a search table, and FIG. 4 Figures 5 and 6 are diagrams showing examples of the contents of relational tables.
Each figure is a hierarchical diagram showing the contents of the relationship table shown in Figure 4 using keywords and inclusion relationship definition numbers.
be. l...Data storage device, 2...Search table, 3...Relationship table, 4...Keyboard, 5...Display, 6
...Search control device. Name of agent: Patent attorney Shigetaka Awano and one other person Figure 1 Σ 8 Figure 4

Claims (1)

[Claims] A search table that manages keywords added to stored data, a relationship table that defines an inclusive relationship between keywords, and a specified keyword and keywords that have an inclusive relationship with it using the search table and the relationship table. 1. A data search device comprising: a search control device that searches for stored data to which is added.