JPS63205730A - Information retrieving device - Google Patents

Abstract

PURPOSE:To make it unnecessary to execute a special processing even if an index word is repeated and used, and to handle it unitedly in the same way as the case of non-repetition, by generating a non-repeated internal index word from an index word and an identifier and generating an index, executing a comparison of a prescribed condition at the time of retrieval, and executing the retrieval by the internal index word. CONSTITUTION:An information managing means 1 manages information by an identifier 202, and an index generating means 2 generates an index 4 by using a slot 200 containing an internal index word 201 and the identifier 202 as a fundamental unit. A retrieving means 3 brings a retrieval index word given as a condition of retrieval and the slot 200 in the index 4 to a condition comparison successively, and when the slot 200 conforming with the condition is detected, the identifier 202 recorded in its slot 200 is derived. In such a way, by using the non-repeated internal index word 201 irrespective of repetition and non-repetition of the index word, an index can be handled by the same algorithm, and a special processing such as an overflow list processing, etc. can be made unnecessary.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to an information retrieval device generally referred to as a database.

BACKGROUND ART FIG. 3 is an example of an index in a conventional information retrieval device.
In the figure a, 1oO is a slot for managing index words in association with information identifiers, 101 is an index word, 102 is an information identifier, 103 is an overflow pointer when the same index word is assigned to multiple pieces of information, and 11o is the overflow area.

Figure 3 shows a part of the index when information #1, #2, #4, and #9 are given the index word aaa, and information #3 is given the index word abc. be. As can be seen from the figure, when index terms overlap (index term = aa
The structure of the index changes depending on a) and the case where there is no overlap (index word=abc). Special processing is performed regarding the overflow area 110 when creating an index or searching.

Problems to be Solved by the Invention However, with the above configuration, it is necessary to change the processing depending on the presence or absence of overlapping index words or overflow areas, which complicates the handling of the index.

In view of this, it is an object of the present invention to provide an information retrieval device that eliminates the need to perform special overflow processing even when index words are used overlappingly, and can be handled in a unified manner as when the index words are not overlapping.

Means for Solving Problems The present invention provides an information management means for managing information using identifiers, an index creation means for generating internal index words from index words and identifiers, and creating an index using the internal index words, and a search method. This information retrieval device includes a search means for obtaining target information by searching an index word using an internal index word.

Effects of the present invention With the above-described configuration, by using non-overlapping internal index words regardless of whether the index words are overlapping or non-overlapping, the index can be handled by the same algorithm, and special processing such as overflow list processing is not required. Can be done.

Embodiment FIG. 1 shows a block diagram of an information retrieval device in an embodiment of the present invention. In FIG. 1, 1 is an information management means, 2 is an index creation means, 3 is a search means, and 4 is an index.

The information management means 1 manages information using identifiers.

The index creation means 2 creates an index 4 using a slot containing an internal index word and an identifier as a basic unit. The search means 3 sequentially compares the search index given as a search condition with the slots in the index 4, and when a slot that meets the conditions is found, it determines the identifier recorded in that slot.

The operation of the information retrieval device of this embodiment configured as described above will be described below.

In FIG. 2, 200 is a slot, 2o1 is an internal index word, and 202 is an information identifier. For the sake of explanation, the identifier will be represented by a number and referred to as an identification number. Although things with different identification numbers 202 may have the same index word, things with the same identification number 202 never have the same index word. Therefore, if an index word concatenated with the identification number 202 is set as a new internal index word 201, the internal index word 2o1 becomes non-overlapping.

The index creation means 2 thus creates the internal index word 201, and creates the index 4 using this internal index word 201. Third
If an index equivalent to the one shown in Figure 2 is created according to this embodiment, it will look like the one shown in Figure 2. As is clear from the figure, when creating the index 4, there is no need to consider overflow processing due to duplication of index words □.

The search means 3 executes a search operation using this internal index word 201, but it is not possible to directly compare the internal index word 201 and the search index word. If the maximum value of the identification number 202 is N, a certain index word x has the value x of the internal index word 201.
It can correspond to i=X#0 to X#N. Therefore, (1) For a size comparison search, R(X is Ri(X#0 R)X id Ri)X#N (2) For a match search, R=X is Ri >
All you have to do is convert N and the conditional expression. Search method 3
performs a search according to this conversion formula.

In addition, in this example, as a generation method of the internal index word 2o1,
Although the connection between the index word and the identification number 202 has been described, the invention is not limited to this. In addition, when the information identifier is obtained from the internal index word 201, the identification number 20 is obtained from the slot 200.
2 can also be omitted.

As described in detail, according to the present invention, an index is created by generating non-overlapping internal index terms from index terms and identifiers, and when searching, a conditional expression is converted and then a search is performed using the internal index terms. This eliminates the need for special processing even when index words are used overlappingly, and can be handled in the same way as when they are not overlapping, which has a great practical effect.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an information retrieval device according to an embodiment of the present invention, FIG. 2 is a structural diagram of an index according to the same embodiment, and FIG. 3 is a structural diagram of a conventional index. 1... Information management means, 2... Index creation means, 3... Search means, 4... Index,
200...Slot, 201...Internal index word, 202...Identifier.

Claims (1)

[Claims] an information management means for managing information using identifiers; an index creation means for synthesizing an internal index word using an index word related to the information and the identifier; and then creating an index using the internal index word; 1. An information retrieval device comprising: a retrieval means for retrieving target information by retrieving an index word using a retrieval index word.