JP2003085206A - Information retrieving method and system, its program and storage medium with the same program stored - Google Patents

Abstract

PROBLEM TO BE SOLVED: To retrieve information whose utility value is high for a user without making it necessary to previously register the personal information of the user. SOLUTION: A retrieval condition from a client terminal is inputted according to the destination of a user, and the information of a data base to be retrieved is retrieved according to the retrieval condition, and the strength of the relevancy of the retrieved information retrieved from the data base to an attribute keyword is judged by referring to a history management table constituted of retrieval keywords stored for each user and the use frequency of those retrieval keywords, and the retrieved information is outputted as a retrieved result according to the strength of the relevancy.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information search method and apparatus, an information search program, and a storage medium storing the information search program, and more particularly to a search for highly useful information in an information search system on a network. The present invention relates to an enabling information search method and apparatus, an information search program, and a storage medium storing the information search program.

[0002] More specifically, an information search method and device, an information search program, and a memory that stores an information search program capable of performing information search that reflects the user's preference at the time of search without registering the personal information of the user in advance. Regarding the medium.

[0003]

2. Description of the Related Art In a conventional information retrieval system, information corresponding to a search keyword (for example, "Yakiniku") designated by a user (for example, "Yakiniku shop") is provided.
Is searched from the database, and all information for the search keyword is provided to the user. However, this method also retrieves information that the user does not need (for example, stores in different locations, stores in different price ranges, etc.),
It is not possible to immediately obtain the information that the user wants most (information with high utility value). That is, although the searched information can be personalized, the search result for each user is not reflected, and the search result that reflects the preference of each user cannot be obtained. Therefore, a desired search result cannot be obtained for a user who has a strong preference.

As an information retrieval system that solves such a problem, there is a concept of individualization. Personalization is to characterize the user (individualization) by registering the user's personal information (age, gender, hobbies, etc.) before using the information retrieval system, and to It is a search.

FIG. 10 is a diagram for explaining the operation of the conventional information retrieval system. Step 10) First, when searching for information, the user registers personal information (here, address, age, and gender) of the user himself / herself in the information search system in advance.

Step 11) The information search system uses the search keyword input by the user and the user ID which is information for determining which user has searched.
To receive.

Step 12) The DB 36 (personal information table) is searched using the received user ID as a key, and the personal information corresponding to the user ID is acquired. In the content of the personal information table shown in FIG. 11, when the user ID is user B, “address: Tokyo, age: 23, gender: female”
Information is retrieved.

Step 13) Next, the database (search information table) is searched according to the search keyword. In the search information table shown in FIG. 12, when the search keyword is “Western goods”, information A, information C, and information F are searched.

Step 14) Here, the information is narrowed down based on personal information among the three pieces of information. When narrowing down based on the user's address and age, the information F (location Tokyo, target youth) most suitable for the user is narrowed down among the three conditions.

Step 15) As a result of the narrowing down, "information F" is provided to the user.

In this way, the user can search the personalized information by using the personal information.

The above is an example of the conventional technique. In addition, by analyzing the access shape of the user whose registered information is similar to that of the user, the user can search for information that is interesting to the user's taste. You can search for valuable information.

However, in the above-mentioned conventional method of searching based on personal information, it is possible for the user to search for information having a high utility value by registering his / her own personal information in advance. However, from the user's point of view, the problem of leakage of personal information is assumed, and registering personal information is likely to be a hurdle to using the information search service.

As a search service for outputting highly accurate search results without setting personal information, search results are arranged according to the number of times other users have accessed information output as search results in the past. There is something to display instead. With this method, among the information that matches the search conditions, information that has been frequently used (popular) in the past is preferentially output, and it becomes easy to obtain high-quality information.

[0015]

However, in the conventional method of outputting a search result according to the number of accesses, the popularity of the information (URL, etc.) to be searched is evaluated. Although "Yakiniku" is a restaurant that is more popular than other stores, "Yakiniku" is also offered, but "Information Y", which is very popular for "Chinese cuisine," , When the number of accesses is larger than that of "information X", "information Y" is preferentially output. As described above, even if the “information X” has a higher utility value for the user, the information Y having a high popularity as a whole is prioritized. This is because the information as a whole is evaluated by the degree of popularity.

The present invention has been made in view of the above points, and it is possible to register the personal information of the user without registering it in advance.
An object of the present invention is to provide an information search method and device capable of searching information having high utility value for a user, an information search program, and a storage medium storing the information search program.

[0017]

The present invention (Claim 1) includes:
In the information retrieval method for retrieving information in the information retrieval system on the network, the user inputs the retrieval condition from the client terminal according to the specification, retrieves the information of the retrieval target database according to the retrieval condition, and accumulates it for each user. The history management table consisting of the search keywords that are used and the frequency of use of the search keywords is referenced to determine the strength of the relationship between the attribute keywords of the search information searched from the database, and the search is performed according to the strength of the relationship. Output information as search results.

According to the present invention (claim 2), the search information arbitrarily selected by the user from the search result is input, and the attribute keyword stored in the database corresponding to the search information selected by the user is stored in the history management table. Add as a search keyword.

According to the present invention (claim 3), the correlation value calculated by matching the attribute keyword corresponding to the search information with the history information of the history management table for each user is used as the history information and the attribute keyword. Judge as the strength of the relationship with.

According to the present invention (claim 4), the first point indicating the degree of relevance of the attribute keyword corresponding to the search information and the degree of relevance between the history information and the attribute keyword corresponding to the search information. By the sum of the second points that represent the
A third point representing the strength of the relationship between the user and the search information is calculated.

According to the present invention (Claim 5), when calculating the third point, a coefficient is calculated based on the bias of the use point with respect to the search keyword in the history information, and the first point and the second point are calculated. Perform weighting.

The present invention (claim 6) is an information retrieval device for retrieving information in an information retrieval system on a network, which comprises input means for inputting retrieval conditions from a client terminal according to user designation, and retrieval conditions. In accordance with the above, the database search means for searching the information of the search target database, the history management table consisting of the search keyword and the frequency of use of the search keyword for each user, and the history management table are referenced to search the database of the search target. And a search result determining unit that determines the degree of association with the attribute keyword of the search information, and an output unit that outputs the search information as a search result according to the degree of association.

According to the present invention (Claim 7), when the search information arbitrarily selected by the user from the search result is input, the attribute keywords stored in the database corresponding to the search information selected by the user are stored in the history. The management table further includes update means for adding the search keyword.

According to the present invention (claim 8), in the search result judging means, the correlation value calculated by matching the attribute keyword corresponding to the search information with the history information of the history management table for each user is calculated. It includes means for determining the strength of the relationship between the history information and the attribute keyword.

According to the present invention (claim 9), in the search result judging means, the first point indicating the degree of relevance of the attribute keyword corresponding to the search information, the history information and the attribute keyword corresponding to the search information. And a sum total calculation means for calculating a third point indicating the degree of association between the user and the search information, based on the sum of the second points indicating the degree of association with the search information.

According to the present invention (claim 10), when calculating the third point in the total sum calculation means, a coefficient is calculated based on the bias of the use point with respect to the search keyword in the history information, and the first point and Means for weighting the second point are included.

The present invention (claim 11) is an information retrieval program for retrieving information in an information retrieval system on a network, wherein an input process for inputting a retrieval condition from a client terminal according to a user's designation and the retrieval condition. According to the database search process for searching the information of the database to be searched and the history management table consisting of the search keyword and the frequency of use of the search keyword for each user, and the relevance to the attribute keyword of the searched search information Of the search result and the output process of outputting the search information as the search result according to the strength of the relevance.

According to the present invention (claim 12), when the search information arbitrarily selected by the user from the search result is input,
It further has an update process of adding the attribute keyword stored in the database corresponding to the search information selected by the user to the history management table as a search keyword.

According to the present invention (claim 13), in the search result determination process, the correlation value calculated by matching the attribute keyword corresponding to the search information and the history information of the history management table for each user is calculated. It includes a process of judging the strength of the relationship between the history information and the attribute keyword.

According to the present invention (claim 14), in the search result determination process, the first point indicating the degree of relevance of the attribute keyword corresponding to the search information, the history information and the attribute keyword corresponding to the search information. And a summation process of calculating a third point representing the strength of the relationship between the user and the search information by the sum of the second points representing the strength of the relationship with.

According to the present invention (claim 15), when calculating the third point in the total sum calculation process, the coefficient is calculated based on the bias of the use point with respect to the search keyword in the history information, and the first point and The process includes weighting the second point.

The present invention (Claim 16) is a storage medium storing an information search program for searching information in an information search system on a network, wherein an input for inputting search conditions from a client terminal is designated by a user. The process and the database search process for searching the information of the database to be searched according to the search condition, and the history management table consisting of the search keyword and the frequency of use of the search keyword for each user, and the attributes of the searched search information It has a search result determination process for determining the strength of relevance to a keyword, and an output process for outputting search information as a search result according to the strength of relevance.

According to the present invention (claim 17), when the search information arbitrarily selected by the user from the search result is input,
It further has an update process of adding the attribute keyword stored in the database corresponding to the search information selected by the user to the history management table as a search keyword.

According to the present invention (claim 18), in the search result determining process, the correlation value calculated by matching the attribute keyword corresponding to the search information with the history information of the history management table for each user is calculated. It includes a process of judging the strength of the relationship between the history information and the attribute keyword.

According to the present invention (claim 19), in the search result judging process, the first point indicating the degree of relevance of the attribute keyword corresponding to the search information, the history information and the attribute keyword corresponding to the search information. And a summation process of calculating a third point representing the strength of the relationship between the user and the search information by the sum of the second points representing the strength of the relationship with.

According to the present invention (claim 20), when calculating the third point in the sum calculation process, the coefficient is calculated based on the bias of the used point with respect to the search keyword in the history information, and the first point and The process includes weighting the second point.

As described above, according to the present invention, the history of the search keywords used by the user is stored in the history management table for each user, so that the relationship between the attribute keyword corresponding to the search information and the tendency of the search behavior of the user is stored. The search result can be provided to the user depending on the strength of the sex.

Further, by adding the attribute keyword stored in the database corresponding to the search information selected by the user as history information to the history management table, the tendency of the search information preferred by the user can be obtained by using the histogram of the attribute keywords. It becomes possible to express.

Further, the correlation value calculated by matching the attribute keyword corresponding to the search information and the history information of the history management table for each user is determined as the strength of the relationship between the history information and the attribute keyword, It is possible to extract the search action history of the user, that is, the search information having the attribute keyword that matches the user's preference.

Further, the sum of the first point indicating the degree of association between the search keyword and the search information and the second point indicating the degree of association between the history information and the attribute keyword is calculated, It is possible to extract the search information close to the user's preference by determining the strength of the relationship between the user preference and the search information via the keyword.

Further, when the total sum is obtained in the above, the count based on the spread of the history information is calculated and used as the weight of the first point and the second point to reflect the characteristics of the individual search by the user. It is possible to extract the retrieved search information.

[0042]

DESCRIPTION OF THE PREFERRED EMBODIMENTS [First Embodiment] FIG. 1 shows a system configuration according to a first embodiment of the present invention.

In the system shown in the figure, a client terminal 10 and an information retrieval system 30 are connected via a network 20.

The information retrieval system 30 is composed of a CPU, a memory, and a hard disk in terms of hardware. Functionally, the search keyword specified by the user and the search information selected by the user from the search results are input from the client terminal 10, the input unit 31, and the DB 3 according to the search condition.
6, the database search unit 32 for searching 6, the database update unit 34 for updating the information in the DB 36 corresponding to the search information selected by the user from the search results, the search result determination unit 33 for determining the relevance of the search results, and the client for the search results. It is composed of an output unit 35 for outputting to the terminal 10.

Next, the structure of the DB 36 will be described.

FIG. 2 shows the structure of the database according to the first embodiment of the present invention.

The DB 36 includes search target information 37 to be searched by a search keyword, a history information management table 38 for managing a history for each user, and an information management table 39 for managing a plurality of attribute keywords for each information. Composed.

FIG. 3 shows the field structure of the history information management table in the first embodiment of the present invention.

The history information management table 38 manages the search keywords used by the user in the past, and the user ID, the search keyword field used by the user in the past, and the use corresponding to the search keyword. It consists of a point-of-use field that stores information that represents the frequency of numbers. The history information management table 38
The above configuration is an example, and any configuration may be used as long as the usage point field corresponding to the search keyword field is configured.

FIG. 4 shows an example of setting contents of the information management table in the first embodiment of the present invention. The information management table 39 shown in the figure consists of search information and attribute keywords. In the example of the figure, it is assumed that search information and attribute keywords are already stored in each field. A circle mark in the figure indicates that the attribute is set, and a minus sign indicates that the attribute is not set.

Next, the operation of the above configuration will be described.

FIG. 5 is a flowchart of the first stage in the first embodiment of the present invention. The flowchart shows the operation of performing information search according to the relevance to the search keyword specified by the user.

Step 110) Client terminal 10
The search keyword specified by the user is input from the input unit 31, and the search keyword is transferred to the DB search unit 32. Here, the search keyword is temporarily stored in a memory (not shown). In the following description, it is assumed that “attribute D” is input as the search keyword.

Step 120) In the DB search section 32,
The attribute keywords including the character string of the search keyword are sequentially searched from the attribute keywords of the information management table 39. Here, in addition to the case where the attribute keyword including the character string of the search keyword is a perfect match such as the search keyword “Yakiniku”, the attribute keyword “Yakiniku specialty store”
As described above, it means that those including the character string “Yakiniku” are to be searched. When the attribute keyword matching the search keyword exists in the information management table 39, the corresponding record is stored in the memory.

Step 130) Information management table 39
When the search is completed and there is search information corresponding to the search keyword, the search result determination unit 33 determines the search result. Therefore, the process proceeds to step 140. Without performing, the process proceeds to step 150.

Step 140) The search result judging section 33 judges the strength of the relevance of the search information to the attribute keyword based on the use points of the search keyword set in the history information management table 38. Here, an example of a calculation method for determining the strength of relevance is shown below,
The determination method is not limited to this example.

The search result judgment unit 33 searches the searched information for each user for the same keyword set as the attribute keyword from the search keywords set in the history information management table 38 for each user. It is assumed that the sum total of the points of use set in the history information management table 38 corresponding to the keyword is obtained and the search result is output according to the sum total.

For example, the information management table 39 shown in FIG.
If the information A hits when the user A searches the contents of the above, the point of the information A is the point of the user A of the history information management table 38 in the setting example of the history information management table 38 shown in FIG. Historical information management record,
The value is masked by the attribute record of the information A, and the usage points of the remaining keywords are summed. That is, “attribute A”, “attribute D”, and “attribute E” are set in “information A”, and user A has previously performed “attribute A” twice.
Since "attribute D" is used once and "attribute E" is used twice, 2 + 1 + 2 = 5 is the point of "information A".
Similarly, if the information B is hit, the point of the "information B" is 2, and the value of the "attribute C" is 2 + 1 = 3.

Since the contents of the history information management table 38 are different for each user, the search result varies depending on the individual characteristics of each user, and it is possible to perform a search that makes the best use of the individuality of the user. For example, if user B searches in the same way and "information A", "information B", and "information C" are hit,
The point of "information A" is 0, the point of "information B" is 10, and the point of "information C" is 0, and the search information is displayed in an order different from the result of the user A. That is, the search history information accumulated for each user is reflected, and the individualized search becomes possible.

Step 150) The output unit 35 outputs the search result to the client terminal 10 according to the strength of the relevance, and ends the processing. To output according to the strength of relevance means to sort and output the search information in order of strong relevance to the search keyword, and to express the strength of the relevance with symbols.
There is a method of outputting the information by adding it to the search information, but any method that can output the information according to the strength of relevance may be used.

If the search result is 0, it is notified that there is no information corresponding to the search keyword. At this time, the search keyword saved in the memory necessary for the operation of step 120 is also output.

Next, as the second stage of the operation, the operation of adding points to the search keyword of the information selected by the user from the search result will be described.

FIG. 7 is a flowchart of the second stage in the first embodiment of the present invention.

Step 210) Step 150 described above
In step 150, the search information arbitrarily selected by the user from the search results (when there is one or more) output to the client terminal 10 and the information for specifying the user (for example, the user ID), The search keyword output to the client terminal 10 is input from the input unit 31 and transferred to the DB updating unit 34.

Step 220) The DB updating unit 34 identifies the record to be updated from the history information management table 38 based on the user ID, and the same as the search keyword used by the user among the search keywords of the corresponding record. Add points to the points used corresponding to the keyword.

Assuming that the search keyword used by the user A for the search is search A, the history information management table 3 shown in FIG.
In the case of 8, the point of the search keyword A is “2”, so it is updated to “3”. The table after updating is shown in FIG.

In the above method, it is necessary to perform a large number of searches until the number of search histories is sufficiently accumulated, so it is expected that it will take time to realize a search that reflects the user's preference. Therefore, all the attribute keywords set in the search information arbitrarily selected by the user may be used. That is, assuming that the search information selected by the user A is "information A" in FIG. 4, "information B".
“Attribute A”, which is the attribute keyword set in
History information management table 3 of "attribute D" and "attribute E"
Add to the corresponding record of 8. Since “search A”, “search D”, and “search E” in the history information management table 38 of the user A are “2”, “1”, and “2”, respectively,
After updating, as shown in FIG. 9, "3",
It becomes "2" and "3". The points to update are
It is not limited to “1”, and can be set arbitrarily.

In addition, when there are a wide variety of attributes that the user can use (for example, arbitrary words), the history field of the user may become long. Therefore, the history may be stored according to the top n search keywords in descending order of points.

[Second Embodiment] In the present embodiment,
The points described in the first embodiment,
By utilizing the point disclosed in Japanese Patent Application No. 2001-86192, which determines the strength of the relationship between the search keyword specified by the user and the search information, a search result that is more accurate and suits the user's preference can be obtained. A method for acquiring will be described.

Here, the point acquisition method according to Japanese Patent Application No. 2001-86192 is to add points to the points corresponding to the search information and the attribute keyword stored in the database corresponding to the search information selected by the user. Is.

The points acquired by the method described in Japanese Patent Application No. 2001-86192 are the first points, and the first points
The point obtained by the embodiment will be described as a second point.

Hereinafter, two methods for obtaining the third point from the above-mentioned first point and second point will be described.

First, the first method will be described.

In this method, the sum of the first point and the second point is the third point. It is considered that the third point directly represents the strength of the relevance between the user's preference and the search information through the search keyword. However, in this case, the following problems are considered to occur.

The first problem is that a user who makes many searches has a large number of pieces of history information, so that the value of the second point becomes large and the influence of the first point becomes small. The second problem is that the search information (for example, a famous store) accessed by many users has a large point corresponding to the attribute of the search information, resulting in a large first point and a second point. The effect of is reduced.

In order to solve the above problems, it is possible to standardize the first point and the second point with the maximum value. By using the value obtained by adding the standardized points as the third point, it is possible to perform a search in which the individual characteristics of the user and the search information are treated equally.

Next, the second method will be described.

In the above-mentioned first method, the method in which the personality is treated equally for each of the user and the search information has been described.
A method of giving weight to the personality information of the user may be considered depending on the distribution of points corresponding to the user's search history keyword.

For example, since the user A in FIG. 6 uses many search keywords evenly, it is considered that the preference of the user A is wide. On the other hand, the user B is prominent only in the “search B”, and the preference is considered to be extremely limited. A user with a strong commitment, such as the user B, may present the search information that matches his / her preference, and the individuality of the search information does not have to be emphasized.

Therefore, when the distribution of points corresponding to the search history keyword is widened as in the case of the user A in FIG. 6, the second point representing the user's individuality is lowered, and in the case of the user B, , Adjust the second point to be higher.

In such a case, the third point is
It may be calculated by the following formula. Third point = (1−α) × first point + α × second point (1) where α is a coefficient representing the spread (bias) of the distribution of points corresponding to the search history key warpo. For example, it may be calculated as follows using the standard deviation σ. α = 1−exp (−σ) (2) According to the above formula (2), when the spread of points is large (when the user's preference is wide), α is close to zero, and the points are biased. In the case (when the user's preference is limited), since α is close to 1, the influence of the first point is large in the former case, and the influence of the second point is large in the latter case. Therefore, it is possible to adjust the balance between the personalization of the user and the personalization of the search information according to the distribution of the preference of the user.

Regarding the operation in this embodiment, the calculation method for obtaining the strength of the relationship between the search information and the user has been described. The calculation for obtaining the strength of the above-mentioned relevance corresponds to step 140 in FIG.

Note that, in the above embodiment, FIG. 5 and FIG.
However, it is possible to construct these operations as a program and install them in the CPU of a computer used as an information retrieval device, or distribute them via a network.

Further, a hard disk drive connected to a computer used as an information retrieval device using the constructed program, a floppy (registered trademark) disk, a CD
-The present invention can be easily realized by storing it in a portable storage medium such as a ROM and installing it when carrying out the present invention.

The present invention is not limited to the above embodiments, and various modifications and applications are possible within the scope of the claims.

[0086]

As described above, according to the present invention, the preference of the user is represented by the search keyword as the search history and the frequency of use thereof, and the strength of the relevance between the preference and the search information is determined. , By enabling the information search according to the strength of relevance, it is possible to obtain the information having the attribute keyword that best matches the user's preference, and it is highly useful even without inputting personal information. You can get information.

Furthermore, by determining the strength of the relationship between the search keyword and the search concession and the strength of the relationship between the user's directivity and the search information, the search keyword and the attribute keyword are used. It is possible to determine the strength of the relevance between the user's preference and the search information, and it is possible to obtain highly useful information without inputting personal information.

[Brief description of drawings]

FIG. 1 is a system configuration diagram according to a first embodiment of the present invention.

FIG. 2 is a configuration diagram of a database according to the first embodiment of the present invention.

FIG. 3 is a diagram showing a field configuration of a history information management table according to the first embodiment of the present invention.

FIG. 4 is an example of information management table setting contents in the first embodiment of the present invention.

FIG. 5 is a flowchart of a first stage according to the first embodiment of the present invention.

FIG. 6 is a setting content of a history information management table in the first embodiment of the present invention.

FIG. 7 is a flowchart of a second stage in the first embodiment of the present invention.

FIG. 8 shows updated history information management table setting contents in the first embodiment of the present invention.

FIG. 9 shows the updated history information management table setting contents (when there are a plurality of keywords to be added) according to the first embodiment of this invention.

FIG. 10 is a flowchart of the operation of the conventional information search system.

FIG. 11 is an example of a personal information table in a conventional information search system.

FIG. 12 is an example of a search information table in a conventional information search system.

[Explanation of symbols]

10 client terminal 20 networks 30 Information retrieval system 31 Input section 32 DB search section 33 Search Result Judgment Section 34 DB Update Department 35 Output section 36 DB 37 Search target information 38 History information management table 39 Information management table

[Procedure amendment]

[Submission date] September 26, 2001 (2001.9.2)
6)

[Procedure Amendment 1]

[Document name to be amended] Statement

[Name of item to be amended] Title of invention

[Correction method] Change

[Correction content]

Title: Information retrieval method and system, information retrieval program, and storage medium storing information retrieval program

[Procedure Amendment 2]

[Document name to be amended] Statement

[Name of item to be amended] Claims

[Correction method] Change

[Correction content]

[Claims]

[Procedure 3]

[Document name to be amended] Statement

[Correction target item name] 0001

[Correction method] Change

[Correction content]

[0001]

TECHNICAL FIELD The present invention relates to an information search method and
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system, an information search program, and a storage medium storing the information search program, and in particular, in an information search system on a network, an information search method, system, information search program, and information search that enable search of highly useful information. The present invention relates to a storage medium that stores a program.

[Procedure amendment 4]

[Document name to be amended] Statement

[Correction target item name] 0016

[Correction method] Change

[Correction content]

The present invention has been made in view of the above points, and it is possible to register the personal information of the user without registering it in advance.
An object of the present invention is to provide an information retrieval method and system capable of retrieving information of high utility value for a user, an information retrieval program, and a storage medium storing the information retrieval program.

[Procedure Amendment 5]

[Document name to be amended] Statement

[Name of item to be corrected] 0022

[Correction method] Change

[Correction content]

The present invention (claim 6) is based on a network.
An information retrieval system, an input means for inputting a retrieval condition from a client terminal by a user's designation, a database retrieval means for retrieving information on a database to be retrieved according to the retrieval condition, a retrieval keyword for each user, and the retrieval A history management table that includes the frequency of use of keywords, a search result determination unit that refers to the history management table and determines the strength of association with the attribute keyword of the search information retrieved from the database to be searched, and the association Output means for outputting the search information as a search result in accordance with the strength of.

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Yasufumi Kikuchi             2-3-1, Otemachi, Chiyoda-ku, Tokyo             Inside Telegraph and Telephone Corporation F-term (reference) 5B075 KK07 ND16 PR04 UU40

Claims (20)

[Claims] 1. An information retrieval method for retrieving information in an information retrieval system on a network, wherein retrieval conditions are input from a client terminal by a user's designation, and information in a retrieval target database is retrieved according to the retrieval conditions. , Referring to a history management table consisting of search keywords accumulated for each user and the frequency of use of the search keywords, determining the strength of association with the attribute keywords of the search information searched from the database, An information search method, wherein the search information is output as a search result according to the strength of relevance. 2. The search information arbitrarily selected by the user from the search result is input, and the attribute keywords stored in the database corresponding to the search information selected by the user are stored in the history management table. Claim 1 to add as a search keyword
Information retrieval method of the description. 3. A correlation value calculated by matching the attribute keyword corresponding to the search information with the history information of the history management table for each user is a correlation value between the history information and the attribute keyword. The information retrieval method according to claim 1, wherein the information retrieval method is determined as strength. 4. A first point indicating the degree of relevance of the attribute keyword corresponding to the search information, and a first point indicating the degree of relevance between the history information and the attribute keyword corresponding to the search information. The information search method according to claim 1, wherein the third point representing the strength of the relationship between the user and the search information is calculated by the sum of the points of 2. 5. When calculating the third point, a coefficient is calculated based on the bias of the use point with respect to the search keyword in the history information, and the first point and the second point are weighted. The information search method according to claim 4. 6. An information retrieval device for retrieving information in an information retrieval system on a network, comprising input means for inputting retrieval conditions from a client terminal according to user designation, and a database to be retrieved according to the retrieval conditions. Database retrieval means for retrieving information, a history management table consisting of a search keyword and the frequency of use of the search keyword for each user, and referring to the history management table, of the search information retrieved from the database to be searched. An information retrieval apparatus comprising: a search result determination unit that determines the strength of association with an attribute keyword; and an output unit that outputs the search information as a search result according to the strength of association. 7. When the search information arbitrarily selected by the user from the search result is input, the attribute keywords stored in the database corresponding to the search information selected by the user are set to the history management table. 7. The information search device according to claim 6, further comprising an updating unit for adding a search keyword to the. 8. The search result determination means sets a correlation value calculated by matching an attribute keyword corresponding to the search information and history information of the history management table for each user with the history information. 7. The information search device according to claim 6, further comprising means for determining the strength of association with the attribute keyword. 9. The search result determination means, the first point indicating the strength of relevance of the attribute keyword corresponding to the search information, the relationship between the history information and the attribute keyword corresponding to the search information 7. The information search device according to claim 6, further comprising a sum total calculation means for calculating a third point representing the strength of relevance between the user and the search information, based on the sum of the second points representing the strength of sex. 10. The sum total calculation means calculates a coefficient based on the bias of the use point with respect to the search keyword in the history information when calculating the third point, and the first point and the second point. 10. The information search device according to claim 9, further comprising means for weighting the points. 11. An information retrieval program for retrieving information in an information retrieval system on a network, wherein an input process for inputting retrieval conditions from a client terminal according to user's designation, and a database to be retrieved according to the retrieval conditions Database retrieval process for retrieving information of the above, and the history management table consisting of the retrieval keyword and the frequency of use of the retrieval keyword for each user is referred to determine the strength of association with the attribute keyword of the retrieved retrieval information. An information search program, comprising: a search result determination process; and an output process for outputting the search information as a search result according to the strength of the relevance. 12. When the search information arbitrarily selected by the user from the search result is input, the attribute keyword stored in the database corresponding to the search information selected by the user is set as the history management table. 12. The information search program according to claim 11, further comprising an update process for adding as a search keyword to. 13. The search result determination process includes a correlation value calculated by matching an attribute keyword corresponding to the search information and history information of a history management table for each user, with the history information. The information search program according to claim 11, including a process of determining the degree of association with an attribute keyword. 14. The search result determination process, wherein the first point indicating the degree of relevance of the attribute keyword corresponding to the search information, the relationship between the history information and the attribute keyword corresponding to the search information. The information search program according to claim 11, further comprising a sum total calculation process of calculating a third point indicating the strength of the relationship between the user and the search information, based on the sum of the second points indicating the strength of sex. 15. The sum total calculation process calculates a coefficient based on a bias of usage points with respect to a search keyword in the history information when calculating the third point, and calculates the first point and the second point. 15. The information search program according to claim 14, further comprising a process of weighting the points. 16. In an information retrieval system on a network, a storage medium storing an information retrieval program for retrieving information, an input process for inputting retrieval conditions from a client terminal according to user's designation, , Referring to the database search process for searching the information of the database to be searched and the history management table consisting of the search keyword and the frequency of use of the search keyword for each user, and checking the relevance of the attribute keyword of the searched search information. A storage medium storing an information search program, comprising: a search result determination process of determining strength; and an output process of outputting the search information as a search result according to the strength of relevance. 17. When the search information arbitrarily selected by the user from the search result is input, the attribute keyword stored in the database corresponding to the search information selected by the user is set to the history management table. 17. The storage medium storing the information search program according to claim 16, further comprising an update process for adding as a search keyword to. 18. The correlation result calculated by matching the attribute keyword corresponding to the search information and the history information of the history management table for each user with the history information, The storage medium storing the information search program according to claim 16, including a process of determining the degree of association with the attribute keyword. 19. The search result determination process, wherein the first point indicating the degree of relevance of the attribute keyword corresponding to the search information, the relationship between the history information and the attribute keyword corresponding to the search information. The information search program according to claim 16, further comprising a sum total calculation process for calculating a third point indicating the strength of association between the user and the search information according to the sum of the second points indicating the strength of sex. Storage medium. 20. The sum total calculation process calculates a coefficient based on a bias of a use point with respect to a search keyword in the history information when calculating the third point, and calculates the first point and the second point. 20. The storage medium storing the information retrieval program according to claim 19, further comprising a process of weighting the points.