JPH05233717A - Information retrieving device - Google Patents

Abstract

PURPOSE:To provide the information retrieving device which facilitates access to the second information group stored in a data base by a user. CONSTITUTION:An information retrieving device 10 includes an inputting device 11 which inputs information A and related range E, relevance adding circuit 12 which operates the relevance of the information A transmitted from the inputting device 11 with the second information stored in a data base 20, and adds the operated relevance to the second information, relevance information extracting circuit 13 which extracts the second information to which the relevance in the relevance range E transmitted from the inputting device 11 is added, from the second information group to which the relevance is added by the relevance adding circuit 12, and display device 14 which displays the second information extracted by the relevance information extracting circuit 13. In this case, the relevance adding circuit 12 operates the number of a keyword preliminarily added to the information A which is matched with the keyword preliminarily added to each second information, so that each relevance can be searched.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information retrieval device,
In particular, the present invention relates to an information retrieval device that retrieves and outputs second information related to the first information from the second information group stored in the database.

[0002]

2. Description of the Related Art Conventionally, this type of information retrieval apparatus retrieves the second information most relevant to the information A (first information) from the second information group stored in the database. The second information retrieved in the order of high relevance to the information A is displayed on the display device.
Therefore, the conventional information retrieval device is excellent in that the user can quickly find the second information most related to the information A.

However, it is desired to access a wide range of second information by seeing not the second information most related to the information A but the second information having a low degree of association or the second information indirectly related. Of course, there is also a user request. That is, for example, for a user who wants to purchase a personal computer, the second information in which the information about the model of the personal computer and the information about the performance of the personal computer are most related to the information A of the candidate personal computer. It is information, but the user has information A (the personal computer)
The second information to be accessed based on the above includes the following information. (1) Information about peripheral devices such as external storage devices, printers, monitors and modems (2) Information about boards that can be installed inside (3) Information about connection cables (4) Information about power supply devices (5) For placing various devices Information about racks (6) Information about available software (7) Information about phone numbers of shops that can be purchased (8) Information about good after-sales service (9) Information about delivery date (10) Information about price (11) Comparison target Therefore, the second information that the user wants to access based on the information A (the personal computer) is the information A.
From the viewpoint of the degree of relevance to the second information, the second information has a high degree of relevance to the second information having a low degree of relevance.

[0004]

However, since the conventional information retrieval apparatus displays the second information retrieved in descending order of the degree of association with the information A on the display device, the second information with the highest degree of association is displayed first. It is difficult to answer the user's request to access even the second information that is less relevant,
There is a problem that only the second information having a high degree of association can be retrieved and displayed.

Further, since the conventional information retrieval apparatus uses the degree of association created by the system creator, the degree of association does not necessarily match the degree of association seen from the user's point of view, and the degree of association with the information A is high. When retrieving the second information, the second information desired by the user does not necessarily match the second information retrieved and displayed by the system. In other words, if the personal computer described above is taken as an example, it can be considered that the information A regarding the personal computer desired to be purchased and the second information B regarding the printer, which is a product of the same company, are highly related to each other. It can be understood that the degree of association between the information A about the personal computer and the second information C about the printer having the same level of performance as a product of another company is higher. In this way, although the degree of relevance changes depending on the user's viewpoint, the conventional information retrieval device can only retrieve and display the second information determined by the system to have a high degree of relevance. Therefore, in order to search from a different perspective, the user must redefine the question (search formula) for extracting the second information desired by the user, and access to related information is not possible. There is a problem of being difficult.

[0006] An object of the present invention is to provide an information retrieval device in which a user can easily access the second information group stored in the database.

[0007]

The information retrieval apparatus of the present invention is arranged such that, from the second information group stored in the database,
In an information retrieval device for retrieving and outputting second information related to first information, an input unit for inputting the first information and a related range, and the first unit sent from the input unit. And a degree of association between the second information stored in the database and the degree of association between the second information stored in the database and the degree of association. Related information extracting means for extracting, from the second information group to which the degree of association is given, the second information to which the degree of association is given within the relevant range sent from the input means.

Here, the first keyword is previously added to the first information, and the second keyword is previously added to each of the second information. , The number of matching first keywords pre-assigned to the first information and second keywords pre-assigned to the second information is calculated to obtain the degree of association, the association range However, the normalized relevance value obtained by normalizing the relevance may be one of the first to Nth ranges divided into N stages.

Further, the degree-of-association adding means adds the first keyword and the second keyword, which are added in advance to the first information.
Instead of calculating the number of matches with the second keyword given in advance to the information of 1, the degree of association may be obtained based on the distance between the first keyword and the second keyword. ..

Further, the degree-of-association giving means is the first
Of the first word described in the information and the second word described in the second information are calculated to calculate the degree of relevance, and the relevance range determines the degree of relevance. The normalized normalized relevance value may be one of the first to Nth ranges divided into N levels.

In addition, instead of the degree-of-association assigning means calculating the number of times the first word described in the first information and the second word described in the second information match. Alternatively, the degree of association may be obtained based on the distance between the first word and the second word.

[0012]

In the information retrieval apparatus of the present invention, the first information sent from the input means and the second information stored in the database are used.
From the input means, the degree of association with the information is calculated, and the degree of association is added to the second information, and the second information group to which the degree of association is added by the degree of association is input. The user inputs the first information and the related range using the input means by including the related information extracting means for extracting the second information to which the degree of association within the transmitted related range is added. Only then can the second information in the desired relevant range be accessed.

[0013]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the essential parts of a first embodiment of the information retrieval system of the present invention.

The information retrieval device 10 has information A and a related range E.
And the input device 11 to which
Information A to be received and the first stored in the database 20
Information B of 2₁~ B_NRelevance D with₁~ D_NTo calculate
Relevance D₁~ D_NThe second information B₁~ B_NTo give to
The degree-of-association circuit 12 and the degree-of-association D in the degree-of-association circuit 12 ₁
~ D_NSecond information group B to which is added₁~ B_NFrom the input device
Relevance D within the relevant range E sent from the storage 11₁~ D_N
The second information B added with₁~ B_NRelated information extraction to extract
The output circuit 13 and the second information extracted by the related information extraction circuit 13
Information B₁~ B_NIs displayed.

Here, the degree-of-association circuit 12 adds the information A to
Pre-assigned keyword and each second information B₁~ B_N
Matches the keywords assigned to each
By calculating the number, each degree of association D₁~ D_NThat
Ask for. That is, for example, the first key word is added to the information A.
K₁ And the second keyword K₂ And the third keyword K
₃ And the second information B.₁ To the first key
Word K₁ And the fourth keyword K_Four And the fifth kiwa
De K_Five If and are given in advance, the first key
K₁ Only the information A and the second information B match.₁ When
Degree D of₁ Becomes "1". Also, from the input device 11
The related range E that is input is the degree D of each relationship.₁~ D_NNormalize
Normalized relevance value ND₁~ ND_NWas divided into three stages
First to third range E₁~ E₃One of
It That is, for example, the information A is given in advance.
The number of keywords is n, and the second degree is set by the relevance assignment circuit 12.
Information B₁ Relevance D given to₁ Where p is
Continuity normalized value ND₁ Is ND₁ = P / n (1), and 0 ≦ p / n <0.3 (degree of association
Small) when the first range E ₁ Within 0.3, 0.3 ≦
The second range E when p / n <0.7 (medium degree of association) ₂ Within
And 0.7 ≦ p / n (the degree of association is high)
Range E of 3₃ It is said to be inside.

Next, the operation of the information retrieval device 10 will be described.

When the user wants to obtain the second information B _{1 to} B _N having a low degree of association with the information A, the user must input the input device 11
Using, input the information A desired to be searched and the first range E ₁ (small degree of relevance) which is the desired related range. In the degree-of-association circuit 12, the information A sent from the input device 11 is sent.
Pre-assigned to the keyword and each second information B ₁ ~
The keyword assigned to B _N in advance is compared, and the number of matching both is calculated. As a result, the relevance D ₁ to D _N is determined, each of which is assigned to each of the second information B ₁ ~B _N. A related information extracting circuit 13, the relevant degree normalized value ND ₁ to ND _N of each of the second information B ₁ .about.B _N is determined according to the above (1), in conjunction range E sent from the input device 11 second information B ₁ .about.B _N included in a range E ₁ is the first is extracted. Extracted second information B ₁
˜B _N are displayed by the display device 14. Information A
About when it is desired to obtain a second information B ₁ .about.B _N in the degree of association, the user using the input device 11, the second range E is related the desired range and the desired information A search ₂
You can enter (medium relevance). Further, when it is desired to obtain the second information B _{1 to} B _N having a high degree of association with the information A,
The user may use the input device 11 to input the information A desired to be searched and the desired _third related range E ₃ (high degree of association).

Therefore, the information retrieval apparatus of this embodiment is
It is possible to search and display the same operation to large information relevance from a small information of relevance, user access a second information B ₁ .about.B _N be easily related from the information A to be searched based can do.

Next, a second embodiment of the information retrieval system of the present invention will be described.

The information retrieval apparatus of the present embodiment, relevance imparting circuit, based on the distance between keywords that are previously assigned to the keyword and the second information B ₁ .about.B _N which is previously given to the information A The _first point shown in FIG. 1 is to obtain the degrees of association D _{1 to DN} .
This is different from the information search device 10 of the embodiment. That is, for example, the first keyword K ₁ , the second keyword K _2, and the third keyword K ₃ are given to the information A in advance, and the second keyword B ₁ and the first keyword K ₁ are given to the second information B ₁ . If the fourth keyword K ₄ and the fifth keyword K ₅ are given in advance, the degree of association D between the information A and the second information B _1
1 is calculated as follows.

In the degree-of-association circuit, each of the keywords K
₁~ K_FiveDistance d₁₁~ D₁₅, D_{twenty two}~ D_{twenty five}, D₃₃~ D₃₅，
d₄₄~ D₄₅, D₅₅Are stored in advance. here
And the distance d_ij(= Distance d_ji) Is the i-th keyword K_i When
Jth keyword K_j Represents the distance from. First, the emotion
The first keyword K assigned to the information A in advance₁ And the second
Information B₁ The first keyword K that has been previously assigned to
₁ , The fourth keyword K _Four And the fifth keyword K_Five 
And each distance d₁₁, D₁₄, D₁₅Are required respectively. Continued
And the second keyword K that is previously assigned to the information A
₂ And the second information B ₁ First keyer pre-assigned to
K₁ , The fourth keyword K_Four And the fifth kiwi
De K_Five And each distance d_{twenty one}, D_{twenty four}, D_{twenty five}Are required respectively
It Then, the third keyer previously assigned to the information A
K₃ And the second information B₁ Is assigned to the first
Keyword K₁ , The fourth keyword K_Four And the fifth key
Word K_Five And each distance d₃₁, D₃₄, D₃₅Each wanted
To be After that, each distance d obtained first₁₁, D₁₄, D
₁₅Minimum value of min (d₁₁, D₁₄, D₁₅), And second
Each distance d_{twenty one}, D_{twenty four}, D_{twenty five}Minimum value of min (d_{twenty one}, D
_{twenty four}, D_{twenty five}), And each distance d obtained last₃₁, D₃₄, D₃₅of
Minimum value min (d₃₁, D₃₄, D₃₅) And are calculated respectively
D₁ = Min (d₁₁, D₁₄, D₁₅) + Min (d_{twenty one}, D
_{twenty four}, D_{twenty five}) + Min (d₃₁, D₃₄, D₃₅) (2) gives information A and second information B₁ Relevance D with₁ Wanted
Be done. Information A and remaining second information B₂~ B_NDegree of association with
D₂~ D_NAre similarly calculated respectively. Reference
Therefore, in FIG.₂~ D_NOne of the specific steps for seeking
An example is shown in a flow chart.

Next, an example of the definition of the distance between the keywords K _{1 to} K ₅ will be described with reference to FIG.

This definition example is based on the Kadokawa Thesaurus (Shinju Ohno,
(Hamano Masato; Kadokawa Shoten) is used to define the distance between the keywords K _{1 to} K ₅ . That is, in the Kadokawa Thesaurus, each vocabulary is classified into a tree shape as shown in FIG. For example, the vocabulary "computer" is a major category "culture."
It is located under the subcategory "Electrical equipment" under the subcategory "Machine" under the subcategory "Merchandise" under. Here, the distance between each vocabulary can be defined by the length of the path on the classification tree shown in FIG. For example, the distance between the vocabulary "computer" and the vocabulary "TV" may be defined by the vocabulary "computer" -a more detailed classification "electrical equipment" -the vocabulary "TV", "2". Further, the distance between the vocabulary “computer” and the vocabulary “cable” is calculated as follows. It can be defined by the fine classification "wire"-"6" in the vocabulary "cable". Thus, by using the respective word shown in Figure 2 as a keyword K ₁ ~K _5, each keyword K ₁
The distance between K _{5 and} K ₅ can be defined. Note that instead of the Kadokawa Thesaurus, a distance between each keyword K _{1 to} K ₅ is defined using a classification vocabulary table (National Institute for Japanese Language; Hideei Publishing) in which each vocabulary is classified in a tree shape. You may.

In the above description, the relevance levels D _{1 to} D _N are obtained using the keywords, but the information retrieval apparatus of the present invention is not limited to this. For example, when the first information group and the second information group are texts, the words described in the texts may be used. That is, in the information retrieval device 10 shown in FIG. 1, the number of matching keywords pre-assigned to the information A and the keywords pre-assigned to the second information B _{1 to} B _N is calculated and the relations are calculated. Instead of obtaining the degrees D _{1 to DN} ,
And calculates the number of the words listed in has been a word has a second information B ₁ .about.B _N according to information A matches relevance D
_{1 to DN} may be obtained.

[0026]

Since the present invention is configured as described above, it has the following effects.

According to the first aspect of the present invention, the degree of association between the first information sent from the input means and the second information stored in the database is calculated, and the calculated degree of association is changed to the second.
Second information to which the relevance degree within the relevance range sent from the input means is given, from the degree-of-relevance giving means given to the information By including the related information extracting means for extracting the second information, the user can access the second information in the desired related range only by inputting the first information and the related range using the input means. Therefore, the user can easily access the second information group stored in the database.

[0028] In the invention according to claims 2 to 3, the degree-of-association assigning means has a first keyword assigned to the first information in advance and a second keyword assigned to the second information in advance. By calculating the number of coincidences with each other or by obtaining each degree of relevance based on the distance between the first keyword and the second keyword, the relation can be obtained regardless of the viewpoint of the system provider. You can search for information.

In the inventions according to claims 4 to 5, the degree-of-association means sets the first word described in the first information and the second word described in the second information. By calculating the number of coincidences or obtaining each degree of relevance based on the distance between the first word and the second word, the related information can be obtained regardless of the viewpoint of the system provider. You can search.

[Brief description of drawings]

FIG. 1 is a block diagram showing a main part of a first embodiment of an information search device of the invention.

FIG. 2 is a diagram for explaining an example of a definition of a distance between keywords in the second embodiment of the information search system of the invention.

FIG. 3 is a flowchart showing an example of a specific procedure for obtaining each degree of association in the second embodiment of the information retrieval system of the present invention.

[Explanation of symbols]

10 Information Retrieval Device 11 Input Device 12 Relevance Addition Circuit 13 Related Information Extraction Circuit 14 Display Device 20 Database A Information E Related Range d _ij Distance D _i Relevance

Claims (5)

[Claims] 1. An information search device for searching for and outputting second information related to the first information from a second information group stored in a database, wherein the first information and the related range are The degree of association between the input means to be input and the second information stored in the database and the first information sent from the input means is calculated, and the calculated degree of association is calculated as the first degree. From the second information group to which the degree of association is given by the degree-of-association giving means, the degree-of-association given to the information of No. And a related information extracting means for extracting the extracted second information. 2. A first keyword is pre-assigned to the first information, a second keyword is pre-assigned to each of the second information, and the degree-of-association giving unit is: The degree of association is calculated by calculating the number of times the first keyword pre-assigned to the first information and the second keyword pre-assigned to the second information match, The information retrieving apparatus according to claim 1, wherein the normalized relevance degree value obtained by normalizing the relevance degree is one of N to N ranges. .. 3. The degree-of-association unit calculates the number of matching first keywords pre-assigned to the first information and second keywords pre-assigned to the second information. The information retrieving apparatus according to claim 2, wherein the degree of association is obtained based on a distance between the first keyword and the second keyword instead. 4. The degree-of-association means calculates the number of coincidences between the first word described in the first information and the second word described in the second information. The degree of association is obtained, and the association range is any one of a first to Nth range in which an association degree normalized value obtained by normalizing the degree of association is divided into N levels. The information search device according to claim 1. 5. The relevance degree assigning means, instead of calculating the number of coincidences between the first word described in the first information and the second word described in the second information. The information retrieval apparatus according to claim 4, wherein the degree of association is obtained based on the distance between the first word and the second word.