JP2003303242A - System, method and program for monitoring information - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve such a problem that time is required to obtain an investigation result because an information clipping company collects and extracts information after receiving a request of a user, and such a problem that there is no effective means for acquiring a new user, in conventional service. <P>SOLUTION: In this information monitoring system by which the information clipping company collects the information according to an information extraction rule via a communication line to distribute the information desired by the user to the contracted user, information estimated to be desired by an uncontracted user is collected by an information extracting means, a name of the uncontracted user as a key is stored in a database, information including the uncontracted user name and an article handled by the user is selected out of database information by a filter means, and an advertisement document is added to the selected information by an advertisement preparing means so as to be distributed by an advertisement distributing means. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information monitoring system that collects documents existing in a wide range on a network such as the Internet, extracts information such as evaluations and rumors about products and companies, and delivers them to users. The present invention relates to an information processing technology for notifying a user who needs information in a timely manner, and particularly to a system for acquiring new users.

[0002]

2. Description of the Related Art In recent years, electronic documents such as Web documents and electronic mails on a network such as the Internet, or accumulated text data are rapidly increasing. Information such as evaluations, reputations, and complaints about companies and their products are also included in these large volumes of documents, and it is becoming important to collect and extract such information in business decision making. In particular, in crisis management, business analysis, market research, etc., it is required to promptly detect information / evaluation regarding the company and its products and respond to problems / issues.

As a conventional technique for responding to such a demand, Japanese Unexamined Patent Application Publication No. 2001-325275 "Method and apparatus for creating a WEB page search report using a plurality of search engines operated on the Internet" (hereinafter Reference 1) and “Reputation information retrieval from the Internet” of IPSJ research group material (natural language processing 144-1)
1, pp. 75-82, July 16, 2001) (hereinafter referred to as Document 2). In these systems, mechanization of information collection processing is intended to speed up the processing as compared with the case where information is collected manually.

FIG. 14 shows an outline of a main block configuration in Document 1. In FIG. 14, reference numeral 201 denotes a document search unit that applies a common search condition as a query to a plurality of search engines to execute a search, and acquires a combination of a URL and an update date and time returned from each search engine as a hit list for each engine. . 202 merges the hit list for each engine, creates an engine integrated hit list that eliminates duplicate URLs, compares the latest engine integrated hit list with the previous engine integrated hit list, and identifies the URL that has new or updated. This is a download unit that extracts a URL to be reported and acquires a WEB page corresponding to this URL. Reference numeral 203 denotes a report creating unit that creates an index set by extracting character information from a document set downloaded based on a keyword. A delivery unit 204 delivers the index set to the client as an engine integrated search report.

Further, FIG. 15 shows an outline of a main block configuration in Document 2. In FIG. 15, reference numeral 301 denotes a document search unit that performs a full-text search on the Internet using the input product name as a keyword and outputs a URL list of search results. A download unit 302 acquires each Web page corresponding to each URL in the URL list. A crawler unit 303 follows the link information of the web page and acquires the web page associated with the link. 304 is an evaluation expression dictionary 30 corresponding to a separately designated product category from the input Web page.
5 is an opinion collection unit that extracts a sentence including words of the evaluation expression. 306 is a classification / classification that classifies the sentences extracted by the opinion collection unit 304 into positive / negative classifications, and outputs the result as a result of searching the evaluation expression for the product name.
It is the analysis department.

[0006]

A service that receives a survey request from a user of a company, collects / extracts information from a network, and reports it to the user (hereinafter, referred to as an information clipping service) by using Document 1 and Document 2. Called) has been put to practical use.

However, all of these services collect and extract information after receiving a request from the user, and the information clipping company starts collecting a huge amount of documents published on the network after receiving the request. However, there is a problem that it takes time to obtain the survey results. Also, since it is not possible to know when the necessary information will be published on the network, in order to quickly detect information and evaluations about your company and your products and deal with problems and issues, always contact an information clipping company. There was a problem that it was necessary to keep issuing survey requests.

[0008] The present invention has been made to solve the above problems, and periodically circulates and monitors the network to extract important information, and distributes information regarding the user who has issued a survey request to that user. On the other hand, regarding information about uncontracted users, by distributing a part of the extracted information to that user as an advertisement and proposing the use of the information clipping service, non-contracted users can use the information clipping service to The purpose is to be able to quickly detect information and evaluations regarding in-house products and respond to problems and issues. In this way, by collecting and extracting important information in advance and notifying non-contracted users of the existence of important information as advertisements, the information has already been gathered at the time when the user makes a contract, so the user has problems. Can quickly respond to issues. In addition, the user has the merit that the information clipping company sends the presence or absence of information to be investigated as an advertisement, and the information clipping company can obtain an effective advertising means for increasing new customers. .

[0009]

In the information monitoring system according to the present invention, an information clipping company uses a communication line to
In an information monitoring system that collects information according to the rules described in the information extraction rule database and delivers the information desired by the contract user to the contract user, the information presumed to be desired by the non-contract user is extracted as the information. Information extraction means that collects according to the rules described in the rule database and stores it in the general user unique name database using the name of the non-contract user as a key, and of the information extracted by the information extraction means, stored in the general user unique name database Filter means for selecting information including the name of the non-contracted user and the unique name of the product handled by the non-contractor user, and an advertisement creating means for general users that adds an advertisement document to the extracted information selected by the filter means , The advertisement document created by the advertisement creation method for general users is not applicable And a general user-friendly ad delivery means for delivering against Za.

The information monitoring system according to the present invention is
The advertisement distribution means for general users is provided with appearance tendency monitoring means for accumulating the extracted information about a specific uncontracted user and controlling to distribute the advertisement to the user when the number of accumulated information exceeds a threshold value.

Further, the information monitoring system according to the present invention is
The advertisement delivery means for general users holds an advertisement delivery history for a specific uncontracted user, and when the number of advertisements sent to the user within a predetermined period has exceeded a threshold value, the delivery of the advertisement to the user is stopped. The advertisement delivery history management means is provided.

Further, the information monitoring system according to the present invention is
The general user specific name database passes a synonym dictionary storing synonyms of specific names and synonyms of specific names registered in the general user specific name database to the filter means as specific names to be filtered. And a unique name expansion means.

Further, the information monitoring system according to the present invention is
The filter means includes a proper name estimating means for selecting documents including a character string that can be guessed as a proper name.

Further, the information monitoring system according to the present invention is
The contract request receiving means for receiving the contract request document returned from the non-contract user who received the advertisement document distributed by the general user advertisement distribution means, and the extracted information is distributed to the non-contract user who issued the contract request. An information distribution means for contracted users and a charging means for calculating a usage charge according to the extracted information distribution history for the contracted users are provided.

Further, in the information monitoring program according to the present invention, an information clipping company collects information via a communication line in accordance with the rules described in the information extraction rule database, and requests the contract user to request the contract user. Is an information monitoring program for causing a computer to execute the distribution of information, which collects information presumed to be desired by non-contracted users according to the rules described in the information extraction rule database, and uses the name of the non-contracted user as a key. Among the information extracted by the general user's unique name database and the information extracted by the information extraction procedure, the non-contract user name and the unique name of the product handled by the non-contract user are stored in the general user's unique name database. Filtering procedure to select included information, extraction information selected by filtering procedure , The general user-friendly ad creation process to wipe the advertisement document, the advertisement document generally user-friendly ad creation procedure has been created to execute the general user-friendly ad distribution procedure to be distributed to the corresponding non-contract users to the computer.

Further, in the information monitoring program according to the present invention, the advertisement distribution procedure for general users accumulates extracted information about a specific uncontracted user, and when the accumulated number exceeds a threshold, the advertisement is distributed to the user. An appearance tendency monitoring procedure for controlling the operation is provided.

Further, in the information monitoring program according to the present invention, the advertisement distribution procedure for general users holds an advertisement distribution history for a specific uncontracted user, and the number of advertisements transmitted to the user within a recent predetermined period is a threshold value. When the number exceeds, the advertisement distribution history management procedure for stopping the distribution of the advertisement to the user is provided.

Further, in the information monitoring program according to the present invention, the general user proper name data storage procedure includes a synonym storage procedure for storing a synonym of a proper name in a synonym dictionary.
A unique name expansion procedure for passing a synonym of the unique name registered in the general user unique name data storing step to the filter means as a unique name to be filtered.

In the information monitoring program according to the present invention, the filtering procedure includes a proper name estimating procedure for selecting a document including a character string that can be guessed as a proper name.

Further, in the information monitoring method according to the present invention, the information clipping company collects information according to the rules described in the information extraction rule database via the communication line, and requests the contract user to request the contract user. In the information monitoring method for distributing the information, the information presumed to be desired by the non-contracted user is collected according to the rule described in the information extraction rule database, and the non-contracted user name is used as a key in the general user unique name database. From the information extraction step to be stored and the information extracted by the information extraction step, the information including the name of the non-contract user and the unique name of the product handled by the non-contract user is selected from the information of the general user's proper name database Add an advertising document to the filter step and the extracted information selected by the filter step. Provided that the general user-friendly ad creation step, the advertisement document generally user-friendly ad creation step is created, and the general user-friendly ad delivery step to be distributed to the corresponding non-contract users.

The information monitoring method according to the present invention further includes a contract request receiving step of receiving a contract request document returned from an uncontracted user who has received the advertisement document distributed by the general user advertisement distribution step, and a contract. The method includes an information distribution step for contracted users that distributes extracted information to non-contracted users who have made a request, and a charging step that calculates a usage fee according to the extracted information distribution history for contracted users.

[0022]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiment 1. FIG. 1 shows a system configuration diagram in the first embodiment of the present invention. In FIG. 1, reference numeral 1 denotes a document collecting unit, which is published on the network by using a plurality of search engines operated on the network while tracing external document references such as links embedded in documents on the network. A set of a document, the date and time of the last update of the document, and document location information is collected.

Reference numeral 2 denotes a collected document database, which stores a set of document location information of documents collected by the document collecting means 1 and the last update date and time. Reference numeral 3 denotes update determination means, which compares the last update date and time and the document location information of the document collected by the document collection means 1 with the information registered in the collected document database 2.
It outputs a set of documents that the system has not yet collected, the date and time the document was last updated, and document location information.

4 is an information extraction rule database,
A set of information fields to be extracted and extraction patterns is stored. 5
Is an information extraction unit, which extracts information from the document determined to be updated by the update determination unit 3 according to the rule described in the information extraction rule database 4, and the field name of the applied extraction rule and the final source document. Output as a set with the update date and time and document location information.

Reference numeral 6 is a unique name database for contracted users, which is the unique name of the user who has requested the information clipping vendor to extract information, the name of the user, and the name of the product sold by the user. To store. 7
Is a non-contracted general user unique name database,
Stores unique names that are not held by the contracted user unique name database 6 among general company names and the like.

Reference numeral 8 is a filter means, and the information extraction means 5
From the information output by, the unique name database 6 for contracted users and the unique name database 7 for general users
Only the information including the unique name stored in is selected and output as a set of the last update date and time and the document location information of the extraction source document for each user name.

Reference numeral 9 denotes an extracted information database for contracted users, which contains a set of extracted information including the unique name of the contracted user selected by the filtering means, the last update date and time and the document location information of the original document for extraction. Store the user name as a key. Reference numeral 10 denotes an extraction information database for general users, which stores, with the user name as a key, a set of the extraction information including the unique name of the general user selected by the filter means, the last update date and time and the document location information of the extraction source document. To do.

Reference numeral 11 denotes an information distribution means for contracted users, which distributes the extracted information stored in the contracted user extraction information database 9 to each user by e-mail or the like. Reference numeral 12 is an advertisement creating means for general users, which creates an advertisement document for promoting the information clipping service to uncontracted users based on the extracted information stored in the extracted information database 10 for general users, and Output a set of name and advertisement document.

Reference numeral 13 denotes an advertisement distribution means for general users, which sends the advertisement document created by the advertisement creation means 12 for general users to an uncontracted user corresponding to the user name by e-mail or the like.

Next, the processing of the information extraction system executed according to the first embodiment of the present invention shown in FIG. 1 will be described with reference to the flowchart of FIG. 2 and the explanatory views of FIGS.

In FIG. 2, first, the document collecting means 1 collects documents on the network (step S20).
1). As a method of collecting documents, it is realized by acquiring documents published on the network while tracing external document references such as links embedded in the documents. At this time, a set of the latest update date and time of the collected document and the document location information is collected. In general, a document provided on the network is provided with a means for obtaining the date and time when the document was last updated.

For example, on the Internet, "Last-Modify" in the header information returned by the Web server is used.
The date and time when the document was updated is written in the part beginning with "d". In addition, the document location information is URL (Uni
This is a character string uniquely indicating the location of the document represented by "form Resource Locator". Besides this,
If it is a document on a network composed of Windows (registered trademark) machines, UNC (Universal)
Information called Naming Convention can be used as document location information. Document collection means 1
Outputs a set of the collected document, the last update date and time of the document, and document location information.

Next, the update determination means 3 extracts only the documents that the system has not yet collected from the documents collected by the document collection means 1 (step S202). The update determining means 3 searches the collected document database 2 based on the document location information of the documents collected by the document collecting means 1. As shown in FIG. 3, the collection document database 2 stores a set of document location information and the last update date / time of the document. When the item corresponding to the document location information of the collected document does not exist in the collected document database 2, the update determination means 3 adds the document location information of the collected document and the last update date and time to the collected document database 2 and collects it. It outputs a set of the document, the last update date and time of the document, and document location information. If the relevant item exists in the database 2, the last updated date and time recorded in the database is compared with the last updated date and time of the collected document. If the collected document is newer, the record in the database is rewritten. , A set of the document, document location information, and last updated date and time are output.

Next, the information extraction means 5 extracts information from the document output by the update determination means 3 (step S2).
03). The information extraction unit 5 sequentially applies each information extraction rule stored in the information extraction rule database 4 to the document to be processed, and extracts information that matches the rule. Figure 4
Shows an example of information extraction rule. The information extraction rule consists of a set of information fields to be extracted and an extraction pattern. The field is a label for classifying information to be extracted, such as general consumer's evaluation of products, complaint information, and image of a company. Further, in the extraction pattern, the symbols "[" and "]" are symbols that indicate that the character string enclosed between them is matched even if it is not included in the character string to be extracted.

For example, the extraction pattern "[very] difficult to use" matches both "the product A is very difficult to use" and "there is such a difficult product to use".
Further, the portion enclosed by the symbols "{" and "}" is a symbol that indicates matching when any of the character strings delimited by "|" between them matches. For example,
The extraction pattern "design is good / bad" is
It matches the expressions "good design" and "bad design", but not "excellent design". In addition, the symbol “*” is a symbol that represents matching with an arbitrary character string of 0 or more characters. Information extraction means 5
Scans the document to be processed from the beginning, extracts all the paragraphs containing the character strings that match the extraction pattern from the document, and combines the field name of the applied extraction rule with the last update date and time and document location information of the extraction source document. And output.

Next, the filter means 8 extracts only the information including the proper names registered in the contracted user proper name database 6 and the general user proper name database 7 from the information extracted by the information extracting means 5. Selection (step S204). The contracted user unique name database 6 stores a set of a user name and a unique name as shown in FIG. The user name is a label for uniquely identifying the contracted user, and generally, the contracted company name or personal name is used. As the unique name, a general name of the user or a product name sold by the user is used. As the method of determining the unique name, a method specified by the user at the time of contract or a method automatically determined based on the product list sold by the user is used.

The general user unique name database 7 also stores a set of user names and unique names as shown in FIG. The user name is registered in advance by extracting from a list in which general company names are listed, such as a list of listed company names and a company yearbook. In addition to the general name of each user, a unique name such as a product name extracted from the product list that the user sells is registered in advance.

The filter means 8 outputs a set of the selected extraction information, the user name corresponding to the unique name used for the selection, the document location information of the extraction source document and the last update date and time, and the extraction information concerning the contracted user is It is stored in the contracted user extraction information database 9, and the extraction information related to the general user name is stored in the general user extraction information database 10. The extracted information database 9 for contracted users and the extracted information database 10 for general users are shown in FIG.
User name, extraction information, document location information,
The set of the last update date and time is stored.

Next, the contracted user information distribution means 11 distributes the extracted information stored in the contracted user extraction information database 9 to each contracted user by e-mail or the like (step S205). First, the extracted information stored in the extracted information database 9 for contracted users is totaled for each user name, and then converted into a report format as shown in FIG. 7, for example, and then emailed to each user. To deliver.

Next, an advertisement creating means 12 for general users
Thus, based on the extracted information stored in the extracted information database 10 for general users, an advertisement for promoting the information clipping service to uncontracted users is created (step S206). First, the extracted information stored in the general user extracted information database 10 is totaled for each user name. Then, any extracted information is selected from the plurality of extracted information divided for each user name. In this embodiment, the selection method at this time is not limited.

For example, as shown in FIG. 8, when four pieces of information are extracted as the extraction information related to the user name "Corporation XX", the selection method is to select the first extraction information having the latest date. A method or a method of selecting the third extracted information which seems to be most important can be considered. Based on the extracted information thus selected, the general user advertisement creating means 12 creates an advertisement document for each user. An example of the advertisement document generated at this time is shown in FIG. The advertisement creating means 12 for general users outputs a set of an advertisement document and a user name corresponding to the advertisement document, that is, a set of the user name selected by the filter means 8, the last update date and time of the original document and the document location information. To do.

Next, the advertisement distribution means 13 for general users
Thus, the advertisement created by the advertisement creating means 12 for general users is distributed to each user (step S207).
The set of the advertisement document and the user name output by the advertisement creating means 12 for general users is acquired and delivered to each user by e-mail or the like. At this time, it is possible for an operator of the information clipping company to add processing such as manually checking before distribution.

As described above, in the present embodiment, the information is first extracted from the documents collected by the information extracting means 5 and is classified by the user name by the filter means 8. Therefore, the information is extracted to the uncontracted user. By distributing part of the information as an advertisement and proposing the use of the information clipping service, the information is already gathered at the time this user makes a contract, so the information clipping service can send the survey result to the user quickly. There is an effect. Therefore, the user can quickly respond to the problem / issue.

Further, the user has the merit that the information clipping trader can send the presence or absence of information to be investigated as an advertisement, and the information clipping trader can obtain an effective advertising means for increasing the number of customers. There is an effect that you can.

In the above embodiment, the system using the network has been described, but the system may be a system for collecting documents from an external database directly connected via a line.

Further, in the above embodiment, the document collecting means 1 for collecting the documents on the network by using the plurality of search engines operated on the network is provided, but the document collecting means 1 may be omitted. it can. In that case, the update determination means 3 is provided at the subsequent stage of the information extraction means 5.
The information extraction means 5 extracts information according to the rules described in the information extraction rule database 4, and outputs a set of the field name of the applied extraction rule, the last update date and time of the extraction source document, and the document location information. Extraction information database 9 for contracted users and extraction information database 1 for general users
When the extracted information is stored in 0, the last update date and time and the document location information of the document collected by the update determination means 3 are compared with the information registered in the respective databases 9, and it is determined that the document has not been collected by the system. The set of the last update date and time of the document and the document location information is stored using the user name as a key.

Embodiment 2 FIG. 10 is a system configuration diagram according to Embodiment 2 of the present invention. The same components as those described above (see FIG. 1) are designated by the same reference numerals and their detailed description will be omitted.

Reference numeral 14 is a synonym dictionary, which stores synonyms of proper names. For example, as shown in FIG. 11, as a synonym of the proper name “Mitsubishi Electric”, in addition to notation variations such as “Mitsubishi Electric” and “Mitsubishi Electric”, abbreviations such as “Mitsubishi” are stored.

Reference numeral 15 is a proper name expansion means, which acquires an item corresponding to a proper name registered in the proper user proper name database 7 from the synonym dictionary 14, and causes the filter means 8 to be a proper user proper to be filtered. Pass as a name.

Reference numeral 16 denotes a proper name estimating means, which inspects the extracted information from the beginning where neither the proper name of the contracted user nor the proper name of the general user is detected by the filter means 8 to identify a proper name such as a company name. If there is a character string that can be estimated, the extraction information and the document location information of the extraction source document,
The set of the last update date and time is stored in the general user extraction information database 10 using the unique name as a key.

Reference numeral 17 is an appearance tendency monitoring means, which regularly monitors the extraction information database 10 for general users,
When the extracted information regarding a specific user starts to appear frequently, the advertisement is controlled to be delivered to the user. In the appearance tendency monitoring means 17, first, the general user extraction information database 10 is totaled for each user name. Next, the extracted information for each user name is sorted by the last update date and time of the original document, and if the number of extracted information items updated within a certain period of time has exceeded a predetermined threshold, the extracted information is The updated extracted information is output as a set of document location information, last update date and time, and user name. On the contrary, when the number of pieces of extracted information is less than or equal to the threshold value, the extracted information is not output and the advertisement is not sent.

Reference numeral 18 denotes an advertisement distribution history management means, which is shown in FIG.
If the number of advertisements sent to a user within a certain period of time exceeds a predetermined threshold, keep the delivery history management table as shown in 2 and do not send advertisements or reduce the number of deliveries. To control. The delivery history management table stores a set of a user name and the number of advertisements transmitted to the user within a certain period of time recently. The advertisement delivery history management means 18 acquires from the delivery history management table an item corresponding to the user name paired with the advertisement document output by the advertisement creating means for general users.

When there is no corresponding item in the distribution history management table, a new user name entry is created, the number of advertisements transmitted within a certain period of time is set to 1, and the advertisement distribution means 13 for general users is set. Outputs a set of advertisement to be delivered to the user name of the delivery destination. When there is a corresponding item in the distribution history management table, only when the number of transmissions of the item is less than or equal to a predetermined threshold value, a set of the advertisement to be distributed to the general user advertisement distribution means 13 and the user name of the distribution destination is output. To do.

Reference numeral 19 is a contract request receiving means, which receives the contract request document sent from the user by e-mail or the like, and registers the user name of the user who desires the contract in the charging means 20. Reference numeral 20 denotes a billing means, and when the contracted user information distribution means 11 distributes the extracted information to the user, if it is a registered user name, the current time is added to the distribution record, and the number of past distributions is increased. Calculate the usage fee by. As shown in FIG. 13, the distribution record holds the contracted user name and the distribution history of the extracted information for that user.

As described above, the synonym dictionary 14 and the proper name expansion means 15 shown in FIG. 10 are provided, and when the extracted information is subjected to the proper name filter, the general user's proper name database 7 is registered for the general user. By filtering the synonyms of proper names, it is possible to obtain the extracted information that includes notational fluctuations and erroneous expressions without leaking. Further, by providing the proper name estimating means 16 and guessing whether or not the proper name is used when applying the proper name filter to the extracted information, the extracted information including the user name not registered in the proper user unique name database 7 for general users. You can get it without leaking.

Further, by providing the appearance tendency monitoring means 17, it is possible to provide an advertisement to the user only when the extracted information is rapidly increasing, which has an effect of suppressing unnecessary advertisement to the user.

Further, an advertisement distribution history management means 18 is provided,
By not delivering the advertisement to the user who has sent the advertisement within a certain period in the past, or by reducing the number of deliveries, there is an effect of suppressing unnecessary advertisement to the user.

Further, the contract request receiving means 19 and the charging means 2
By setting 0 and recording the distribution history regarding the contracted user, it becomes possible to charge the service according to the distributed amount.

[0059]

As described above, according to the present invention, the documents collected by the information extracting means are classified by the user name by the filtering means, and a part of the extracted information is distributed to an uncontracted user as an advertisement and the information clipping is performed. By proposing to use the service, the information clipping service has the effect that the survey results can be sent to the user quickly because the information has already been gathered at the time this user makes a contract. Therefore, the user can quickly respond to the problem / issue.

Further, the user has the advantage that the information clipping company can send the presence or absence of information to be investigated as an advertisement, and the information clipping company can obtain an effective advertising means for increasing the number of customers. There is an effect that you can.

Further, a synonym dictionary and a proper name expansion means are provided, and when the proper name filter is applied to the extracted information, the synonyms of the proper name for general users registered in the proper name database for general users are also filtered. By applying, it is possible to obtain the extracted information including the typographical fluctuation and the erroneous notation without leaking.

Further, by providing a proper name estimating means and estimating whether the extracted information is a proper name or not when applying the proper name filter, the user name which is not registered in the general user proper name database 7 is included. Extraction information can be acquired without leaking.

Further, by providing the appearance tendency monitoring means, it is possible to place an advertisement to the user when the extracted information is rapidly increasing, which has an effect of suppressing unnecessary advertisement to the user.

By providing an advertisement delivery history management means and not delivering the advertisement to the users who have sent the advertisement within a certain period in the past, or by reducing the number of delivery,
It has the effect of suppressing unnecessary advertisements to users.

Further, by providing the contract request receiving means and the charging means and recording the distribution history concerning the contracted users, it becomes possible to provide a service for charging according to the distributed amount.

[Brief description of drawings]

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

FIG. 2 is a flowchart according to the first embodiment.

FIG. 3 is an explanatory diagram of a collected document database.

FIG. 4 is an explanatory diagram of an information extraction rule.

FIG. 5 is an explanatory diagram of a unique name database for contracted users.

FIG. 6 is an explanatory diagram of a contracted user / general user extraction information database.

FIG. 7 is an explanatory diagram of a report for contracted users.

FIG. 8 is an explanatory diagram summarizing the extracted information for each general user name.

FIG. 9 is an explanatory diagram of an advertisement document for general users.

FIG. 10 is a system configuration diagram according to the second embodiment of the present invention.

FIG. 11 is an explanatory diagram of a synonym dictionary.

FIG. 12 is an explanatory diagram of a distribution history management table included in the advertisement distribution history management means.

FIG. 13 is an explanatory diagram of a distribution record in the charging means.

FIG. 14 is a main block configuration diagram in prior art 1.

FIG. 15 is a main block configuration diagram in Prior Art 2.

[Explanation of symbols]

1: Document collection means, 2: Collected document database, 3: Update determination means, 4: Information extraction rule database, 5: Information extraction means, 6: Contracted user unique name database, 7: General user unique name database, 8: filter means, 9: extracted information database for contracted users, 10: extracted information database for general users, 1
1 is information distribution means for contracted users, 12 is advertisement creating means for general users, 13 is advertising distribution means for general users, 14 is synonym dictionary, 15 is unique name expanding means, 16 is unique name estimating means, 17 : Appearance trend monitoring means,
18: advertisement distribution history managing means, 19: contract request receiving means, 20: charging means.

Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) G06F 17/30 170 G06F 17/30 170Z 340 340A (72) Inventor Katsushi Suzuki 2-3-2 Marunouchi, Chiyoda-ku, Tokyo Sanryishi Electric Co., Ltd. F term (reference) 5B075 ND20 PR08 UU40 5B085 AC04 AC11 AC14 BA06 BG03 BG04 BG07

Claims (13)

[Claims] 1. An information clipping company collects information according to a rule described in an information extraction rule database via a communication line, and contracts the contracting user with respect to the contracting user.
In the information monitoring system that distributes the desired information, the information that is supposed to be desired by the non-contracted user is collected according to the rules described in the information extraction rule database, and the non-contracted user's name is used as a key to identify the general user. Information extraction means to be stored in the database and information that includes the names of non-contracted users and the unique names of products handled by non-contracted users stored in the general user's unique name database among the information extracted by the information extraction means Filter means, and the extracted information selected by the filter means, the advertisement creating means for general users that adds an advertisement document, and the advertisement document created by the advertisement creating means for general users,
An information monitoring system comprising: an advertisement distribution means for general users, which distributes to corresponding non-contracted users. 2. The advertisement distribution means for general users comprises:
2. The information according to claim 1, further comprising appearance tendency monitoring means for accumulating the extracted information regarding a specific uncontracted user and controlling the distribution of the advertisement to the user when the accumulated number exceeds a threshold value. Monitoring system. 3. The advertisement distribution means for general users comprises:
Keeping an ad delivery history for certain uncontracted users,
The information monitoring system according to claim 1, further comprising: an advertisement distribution history management unit that stops distribution of advertisements to the user when the number of advertisements transmitted to the user within a predetermined period exceeds a threshold value. . 4. The general user proper name database is a synonym dictionary that stores synonyms of proper names, and a synonym of proper names registered in the general user proper name database is subject to filtering by the filter means. The information monitoring system according to claim 1, further comprising: a unique name expansion unit that is passed as a unique name of the. 5. The information monitoring system according to claim 2, wherein the filter unit includes a proper name estimating unit that selects documents including a character string that can be guessed as a proper name. 6. A contract request receiving means for receiving a contract request document returned from an uncontracted user who has received an advertisement document distributed by an advertisement distribution means for general users, and an uncontracted user who has issued a contract request. The information monitoring system according to claim 1, further comprising: an information distribution unit for contracted users that distributes the extracted information; and a charging unit that calculates a usage fee according to the extracted information distribution history for the contracted users. 7. An information clipping company collects information according to a rule described in an information extraction rule database via a communication line, and the contract user-to the contract user-
Is an information monitoring program for causing a computer to execute the distribution of desired information, and collects information presumed to be desired by non-contracted users in accordance with the rules described in the information extraction rule database, and names the non-contracted users. The information extraction procedure that stores the unique name database for general users with the key as the key, and the information extracted by the information extraction procedure that is unique to the name of the non-contract user and the product handled by the non-contract user stored in the general name database for general users. A filter procedure for selecting information including names, an ad creation procedure for general users that adds an advertisement document to the extracted information selected by the filter procedure, and an advertisement document created by the advertisement creation procedure for general users.
An information monitoring program that allows a computer to execute the advertisement distribution procedure for general users that is distributed to applicable non-contract users. 8. The advertisement distribution procedure for general users includes an appearance tendency monitoring procedure for accumulating extracted information about a specific uncontracted user and controlling an advertisement distribution to the user when the number of accumulated information exceeds a threshold value. The information monitoring program according to claim 7, comprising: 9. The advertisement distribution procedure for general users comprises:
Keeping an ad delivery history for certain uncontracted users,
The information monitoring program according to claim 7, further comprising: an advertisement delivery history management procedure for stopping delivery of advertisements to the user when the number of advertisements transmitted to the user within a predetermined period exceeds a threshold value. . 10. The general user proper name data storage procedure includes a synonym storage procedure of storing a synonym of a proper name in a synonym dictionary and a synonym of a proper name registered in the general user proper name data storage step. 8. The information monitoring program according to claim 7, further comprising: a unique name expansion procedure for passing the specified name to the filter means as a unique name of a filter target. 11. The information monitoring program according to claim 8, wherein the filtering procedure includes a unique name estimating procedure for selecting a document including a character string that can be guessed as a unique name. 12. An information monitoring method in which an information clipping company collects information according to a rule described in an information extraction rule database via a communication line and distributes information desired by the contract user to the contract user. An information extraction step of collecting information presumed to be desired by non-contracted users according to the rules described in the information extraction rule database, and storing it in the general user unique name database using the name of the non-contracted user as a key; Of the information extracted by the step, the filter step that selects the information that includes the name of the non-contractor user and the unique name of the product handled by the non-contractor user that is stored in the general name unique name database An advertisement creation page for general users that adds an advertisement document to the extracted information. -Up and, in general the advertisement document a user-friendly ad creation step has been created, the relevant information monitoring method, characterized in that it comprises a general user-friendly ad delivery step, the to be delivered to the non-contract users. 13. A contract request receiving step of receiving a contract request document returned from an uncontracted user who has received the advertisement document distributed by the general user advertisement distribution step, and for the uncontracted user who has issued the contract request. The information monitoring method according to claim 7, further comprising: an information distribution step for contracted users for distributing extracted information according to a contract, and a charging step for calculating a usage fee according to an extraction information distribution history for contracted users. .