JP5142509B2 - Document creation support apparatus and document creation support system - Google Patents

Description

  The present invention relates to a document creation support apparatus and a document creation support system for supporting creation of a document such as a specification in a patent application, creation of a document for responding to a notification sent from the Patent Office, and the like.

  When preparing a specification in a patent application, it is necessary to search prior art documents regarding the invention of the application. As a system for supporting such document research, for example, techniques shown in Patent Documents 1 to 4 have been proposed. In addition, when creating a specification, it is necessary to satisfy predetermined system requirements. From the viewpoint of such system requirements, for example, techniques shown in Patent Document 5 and Patent Document 6 have been proposed as systems that support specification creation so as to satisfy such requirements.

However, in the conventional support system, the comparison with the prior literature and the laws and regulations are only performed, and the document asset created by the user who created the description or the invention is published in the world. Document assets could not be used for patent applications.
JP 2005-234868 A JP 2004-110271 A JP 2003-162542 A JP 2002-023650 A JP 2004-133896 A JP 2003-288407 A

  Conventional document creation support apparatuses and document creation support systems have a problem that document assets created in the past and document assets publicly disclosed cannot be used for patent applications. SUMMARY An advantage of some aspects of the invention is that it provides a document creation support apparatus, a document creation support system, and a document creation support method that can make use of past document assets.

In order to achieve the above-described object, a document creation support apparatus according to one aspect of the present invention includes information relating to a description of an application case, notification case information indicating a command content for the application case, and the notification of the application case. A storage unit that stores response case information including response information indicating response contents to case information and preceding case information including information related to final disposal of the application case for each application case, and a feature related to a new application case An input unit that receives notification information indicating a search item and a command content for a predetermined application case, and similar document information that has characteristics relating to the new application case from the storage unit using the search item received by the input unit as a search key And using the notification information received by the input unit as a search key, it is common with the notification information related to the predetermined application case from the storage unit. Extracting a search unit for searching the response case information including the notification case information volume, the information related to the final disposal of the similar document information the search unit searches, the similarity based on the information relating to the final disposal Predicting the final disposition by calculating the proportion of all document information that was determined to be rejected and determining whether the final disposition of the new application is probabilistically higher or lower based on the proportion And a display unit for displaying the prediction result of the prediction unit and the response case information searched by the search unit.

Further, the document creation support system according to one aspect of the present invention shows information related to the specification of an application case, notification case information indicating a command content for the application case, and a response content to the notification case information of the application case. A server device including a storage unit that stores response case information including response information and preceding case information including information relating to final disposal of the application case for each application case, and a search item indicating characteristics relating to the new application case And an input unit that receives notification information indicating the content of a command for a predetermined application case, and searches for similar literature information having characteristics relating to the new application case from the storage unit using a search item received by the input unit as a search key In addition, the notification information received by the input unit is used as a search key and is shared with the notification information related to the predetermined application case from the storage unit. Wherein a search unit for searching a response case information, extracts the information relating to the final disposal of the similar document information the search unit searches, the similar documents based on the information relating to the final disposal, including the notification case information A predicting unit that calculates the proportion of all information that has been determined to be rejected, and predicts the final disposition by determining whether the final disposition of the new application is probabilistic or rejected based on the proportion And a terminal device including a display unit that displays a prediction result of the prediction unit and response case information searched by the search unit.

  According to the present invention, document creation support utilizing past document assets can be realized. In addition, provision of information related to the future prediction of application projects can speed up decision-making at the time of filing and intermediate processing, etc., and can also help promote examination.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a document creation support apparatus according to an embodiment of the present invention. As shown in FIG. 1, the document creation support apparatus 1 according to this embodiment is a computer device that includes a communication unit 10, an input unit 12, a display unit 14, a storage unit 16, and a calculation unit 18. The document creation support apparatus 1 is connected to the document server 3 via the network 2.

  The network 2 is a communication network such as the Internet or a telephone network. The document server 3 is a server that mainly accumulates technical documents such as patent documents, and has a function of providing information through the network 2. The document server 3 can use, for example, a patent / utility model bulletin database provided by the Patent Office.

  The communication unit 10 is an interface unit that communicates with the document server 3 via the network 2. The communication unit 10 has a function of establishing network communication via the network 2 using, for example, the TCP / IP protocol. In addition to the interface function, the communication unit 10 may have a function as a transmitter / receiver that transmits / receives predetermined data and an electronic application function for electronically filing a patent application. In this case, the document creation support apparatus 1 functions not only as an interface unit but also as a data transmission / reception unit. That is, in addition to the document creation support function, an electronic application function for electronically applying an application using the created document can be exhibited.

  The input unit 12 is data input means such as a keyboard, a mouse, and a removable disk drive. The input unit 12 has a function of receiving a predetermined command, data, or the like by a user input operation or a read operation from a recording medium on which data is recorded. The display unit 14 is a display device such as an LCD or a CRT. The display unit 14 has a function of displaying input information, search results, prediction results, and the like in a predetermined format. In addition, the input part 12 and the display part 14 may function as a terminal device in cooperation.

  The storage unit 16 is a non-volatile storage device such as a hard disk drive. The storage unit 16 stores executable software and the like, and also functions as a plurality of independent databases. In the storage unit 16, a preceding case DB 20, a case DB 22, and a response case DB 24 are constructed.

  The precedent case DB 20 is information on prior art published in the world such as published patent gazettes, patent gazettes, technical papers, industry magazines, and all related documents (applications) from patent application to assessment. Specification information consisting of data such as specifications and drawings submitted at the time of application, application memo information of examination materials used at the time of application, notification information such as notification of reasons for refusal issued by the JPO after application, It is a database configured to be able to store response information of response documents, and precedent case information including response memo information of examination materials used when responding. Prior art information and prior case information stored in the precedent case DB 20 may be accumulated in the precedent case DB 20 through the input unit 12, or may be automatically collected from the document server 3 through the network 2. .

  The case DB 22 includes bibliographic information on application cases scheduled to be filed in the past or filed in the past, pending information indicating the current pending status of the application case (whether it is pending at the JPO, examination stage, etc.), the application Includes family project information including the contents and pending status of projects (family projects) that are legally related to the project, and includes the contents and pending status of projects (related projects) that have a prescribed relationship with the application project. It is a database for storing case information including related case information, specification information such as specifications and drawings submitted at the time of application, application memo information including examination materials used at the time of application. The response example DB 24 is searched when responding, such as notification information issued from the JPO after application, response information including a response document for the notification, response memo information including examination materials used for responding, etc. Response cases including accumulated history information, etc. that are configured by linking the provided prior art information, previous case information, and case information (information associated with the response subject case when responding) together with the date information of the response date A database that stores information. The response case information is linked to each other by a case number or the like, and is stored in the response case DB 24 in a state where all or a part thereof can be freely searched. In addition, the accumulated history information is not only information at the time of response information creation, but also the status of the relevant case / family case / related case at the time of response information creation (existence of pending, review progress, final disposal, etc.) It may include information indicating whether or not.

  Various pieces of information included in the case information and the response case information are linked to each other by the case number, the application number, etc., and are stored in the case DB 22 and the response case DB 24 in a state where all or a part thereof can be freely searched. . The key that associates various pieces of information included in the case information and the response case information is not limited to the case number or the application number. For example, having common family item information (belonging to the same family item group) or having common related item information (having the same business content, etc.) can be searched as an association key. Good. As a result, management and processing can be performed in any group rather than in units.

  The calculation unit 18 is a calculation unit including a CPU, a work memory, and the like. The computing unit 18 can realize the functions of the software by deploying the execution format software stored in the storage unit 16 on its work memory. The calculation unit 18 functions as an information acquisition unit 30, a preceding case search unit 32, a future prediction unit 34, a prediction information generation unit 36, a response case search unit 40, and a response information generation unit 42.

  The information acquisition unit 30 is a data processing unit having a function of acquiring the prior art information and the previous case information through the communication unit 10 and the input unit 12 and storing them in the previous case DB 20. The information acquisition unit 30 also has a function of acquiring notification information through the communication unit 10 and storing it in the response case DB 24.

  The preceding case search unit 32 is a database searching unit that searches the preceding case DB 20 for predetermined prior art information and preceding case information based on the search key received by the input unit 12. The precedent case search unit 32 is realized by a so-called knowledge database system or the like. For example, from the predetermined specification information received by the input unit 12 and the prior art information stored in the precedent case DB 20, a common point ( It has a function to search for documents having many similar points) as similar documents. Similarly, the precedent case search unit 32 permits the final disposition assessment among documents having a lot in common with the description data from the predetermined description information and the preceding case information stored in the preceding case DB 20 ( (Patent) and refusal are also searched and extracted separately.

  The future prediction unit 34 is a data processing unit having a function of predicting what the final disposal will be as a result of applying for the predetermined case information received by the input unit 12. Specifically, the future prediction unit 34 calculates a patent search constant and a rejection determination constant from the previous case information that has a lot in common with the specification information related to the case information searched and extracted by the previous case search unit 32. The final disposal of the case information is estimated from the calculation result.

  The prediction information generation unit 36 is a data generation unit that generates prediction information to be displayed on the display unit 14 based on the prediction result of the future prediction unit 34. The prediction information generation unit 36 causes the display unit 14 to display a prediction result based on an index such as a probability (numerical value) or a graph that is a rejection decision. Moreover, the prediction information generation part 36 may generate | occur | produce the correction information (suggest information which specified the point which should be corrected) of the specification data which concern on the said case information based on the said prediction result.

  The response case search unit 40 is a database search unit that searches predetermined response case information from the response case DB 24 based on the search key and notification information received by the input unit 12. Similarly to the preceding case searching unit 32, the response case searching unit 40 may search the preceding case DB 20 for predetermined preceding case information based on the search key received by the input unit 12. The response case search unit 40 can be realized by the same method as the preceding case search unit 32, but the search target and search key are different from the preceding case search unit 32. That is, the preceding case search unit 32 searches for prior art information and preceding case information having high commonality with predetermined specification information, whereas the response case search unit 40 determines predetermined specification information and notification information for the same. Search for precedent case information that is highly common. In other words, the former aims to determine patentability before application and extract improvement points, and the latter aims to extract the most appropriate cases from the past response cases for notification information issued after filing. .

  The response information generation unit 42 is a data generation unit that generates response information to be displayed on the display unit 14 based on the search result of the response case search unit 40. The response information generating unit 42 displays the response information on the display unit 14 and operates to assist the user in creating a response document.

  Next, the operation of the document creation support apparatus 1 will be described. 2 is a flowchart showing a new document creation support operation in the document creation support apparatus according to this embodiment, FIG. 3 is a diagram showing a display example of an input screen in the new document creation support operation, and FIG. 4 is a new document creation support FIG. 5 is a flowchart showing a response document creation support operation in the document creation support apparatus according to this embodiment.

  First, a support operation for creating a new document will be described. The information acquisition unit 30 periodically accesses the document server 3 through the communication unit 10, acquires the prior art information and the previous case information, and stores them in the previous case DB 20 (step 100, hereinafter referred to as “S100”). . In addition, the information acquisition unit 30 stores the case information received by the input unit 12 in the case DB 22, and similarly stores the response case information in the response case DB 24.

  The input unit 12 accepts case information input by the user and passes it to the information acquisition unit 30. The information acquisition unit 30 stores the received item information in the item DB 22 (S101). The input unit 12 may read the case information stored in the memory unit or the like through its own reading unit and store it in the case DB 22. The item information accepted at this time includes, in addition to bibliographic information, the contents of the proposal (invention manual), technical items (search items) as keywords indicating the technical characteristics related to the item.

  FIG. 3 shows an example of display data displayed by the display unit 14 when receiving item information. As shown in FIG. 3, the received item information includes technical fields, problems (invention problems), solution means (means for solving the problems), prior art documents, and the like as search items.

  When the case information is received, the preceding case search unit 32 searches the preceding case DB 20 for the prior art information and / or the preceding case information using the search item as a search key. As a result, the preceding case search unit 32 acquires, as similar documents, the prior art information and the like that have a lot in common with the contents of the proposal related to the received case information (S102). Here, the similar document acquired by the preceding case search unit 32 is a document that is highly likely to be cited as the prior art when the content related to the received case information is filed. Further, when searching from the precedent case information, information related to the final disposal of past application cases having high commonality with the contents related to the received case information is also acquired. The precedent case search unit 32 passes the acquired similar documents to the future prediction unit 34.

  Based on the received similar documents, the future prediction unit 34 determines what the final disposal will be when applying for the contents of the received case information, specifically, a patent assessment (including a patent trial decision). Whether or not a rejection decision (including a rejection decision) is made is predicted (S103). This prediction process is realized by extracting information related to the final disposition of a past application case that has a high degree of commonality with the contents related to the received case information, and determining whether the permission or rejection is probabilistically higher .

  When the prediction process is executed, the future prediction unit 34 associates the information related to the content of the proposal from the received case information and stores the prediction result in the preceding case DB 20. Thereby, the said prediction result is utilized also in the time of the prediction process after the next of the future prediction part 34, and what is called a learning effect can be acquired (S104).

  Subsequently, the future prediction unit 34 generates display data to be displayed on the display unit 14 based on the prediction result (S105). The display data can be displayed by a visual method such as a probability (numerical value) or a graph related to the prediction result. In addition, the future prediction unit 34 generates suggestion information for improving the prediction result and adds it to the display data. The information for improving the prediction result is to extract the search items and feature points from the precedent case information that is the basis when the prediction result indicates that the rejection is judged, for example, to improve them. Generated from the viewpoint. Specifically, the content (formal reason, substantive reason) of the notification information included in the precedent case information that led to the result of the rejection decision is extracted, and the search item corresponding to the content is linked to promote correction. It can be content. The future prediction unit 34 passes the generated display data to the display unit 14.

  The display unit 14 displays the received display data (S106). In FIG. 4, an example of the prediction result which the display part 14 displays is shown. In this example, as a display data, a similar document (similar patent document), a similarity to the similar document, a final assessment of the similar document (patent in the example shown in FIG. 4), a prediction result (in the example shown in FIG. 4, the reason for refusal) Probability of notification (99%), assumed reference (= similar literature), reason for the closest or probable final disposal in similar literature (not complete in the example shown in FIG. 4), and suggestion information (final disposal) The reverse contents for the reason) are included and provided to the user.

  As described above, in the document creation support apparatus 1 of this embodiment, the preceding case search unit 32 searches for the preceding case information in addition to the prior art information. It is possible to evaluate based on whether or not it has been done. In addition, since the document creation support apparatus 1 of this embodiment includes the future prediction unit 34, it is possible to obtain the possibility of refusal decision and the improvement method thereof before the application. In other words, provision of information related to the future prediction of the application project can speed up decision making at the time of filing or intermediate processing, and can also be used for promoting examination.

  Next, a response document creation support operation will be described.

  The input unit 12 receives notification information input by the user and passes it to the information acquisition unit 30. The communication unit 10 may directly access a server provided in the patent office or the like to automatically acquire notification information and pass it to the information acquisition unit 30. The information acquisition unit 30 stores the received notification information in the case DB 22 and the response case DB 24 (S200).

  When the notification information is stored, the response case search unit 40 searches and reads out the case information corresponding to the notification information from the case DB 22 (S201).

  The response case search unit 40 analyzes the content of the notification information (S202), and searches the response case DB 24 for response case information that has a lot in common with the combination of the content of the notification information and the read case information ( 203). Specifically, the response case search unit 40 first extracts bibliographic information, text information, comment information of the examiner, and the like from the notification information as search items. Subsequently, the response case search unit 40 searches the response case information stored in the response case DB 24 for items whose common points with the search item are equal to or higher than a predetermined reference. By such processing, it is possible to acquire notification examples received from the Patent Office and response cases having many similarities to the case information.

  When the response case information whose common point is equal to or greater than the predetermined reference is acquired, the response case search unit 40 selects response case information to be provided to the user from among them (S204). The criteria for selecting the response case information by the response case search unit 40 include the commonalities between the notification information and the case information, the commonality of the technical fields, the commonality of the text information, and the like. The response case search unit 40 selects one or more response case information satisfying these criteria and passes it to the response information generation unit 42. The response information generation unit 42 generates display data based on the received response case information and passes it to the display unit 14.

  The display unit 14 displays the received display data (205). The user creates a response document for the notification information with reference to the display unit 14. The response document may be created by the input unit 12, the display unit 14, and a word processor unit (not shown).

  The input unit 12 receives response document information including a response document created by the user, and passes the response document information to the information acquisition unit 30 (S206). When a response document is created by a word processor (not shown) provided in the document creation support apparatus 1, the word processor directly passes response document information to the information acquisition unit 30.

  The information acquisition unit 30 stores the received response document information in the case DB 22 and the response case DB 24 (S207). More specifically, the information acquisition unit 30 configures response information, response memo information, accumulated history information, and the like and stores them in the response case DB 24. By storing such response document information in the response case DB 24 in addition to the case DB 22, it is possible to improve the accuracy in the next response case search. That is, the response information created at the time of response and the response memo information and the accumulated history information given or referred to at the time of creating the response information are stored as a “response state” as a backup. Therefore, it is possible to search afterwards what kind of information and state the determination at the time of creating the response information is made, and the learning function for the response case information is enhanced.

  As described above, according to the document creation support apparatus 1 of this embodiment, the past response case information is searched and displayed based on the notification information received for the case related to the predetermined case information. Can support creation. In particular, since response memo information is included in the response case information, it is possible to provide so-called know-how information as to what kind of response should be made in any case.

  Next, the preceding case search unit in the document creation support apparatus 1 will be described in detail. FIG. 6 is a block diagram showing the configuration of the preceding case search unit in the document creation support apparatus according to this embodiment, and FIG. 7 is a flowchart showing the operation of the preceding case search unit shown in FIG. As shown in FIG. 6, the preceding case search unit 32 of this embodiment includes a table storage unit 320, an area setting unit 321, a clustering processing unit 322, a keyword extraction unit 323, a search expression creation unit 324, and a search execution unit 325. I have.

  The table storage unit 320 is a storage unit that temporarily stores data of various processes. In the table storage unit 320, a clustering result table (clustering result storage area), a keyword table (keyword storage area), a search expression storage area, and a search result table (search result storage area) are provided.

  The area setting unit 321 is a data processing unit having a function of setting an item to be searched (document area to be searched) received by the input unit 12. The area setting unit 321 passes information regarding the set document area to the clustering processing unit 322. Here, the item (document area) to be searched means an area (paragraph, item, page, etc.) to be searched in the prior art information to be searched. For example, in the specification of a patent application, description areas such as “Claims” and “Problem of Invention” are items to be searched.

  The clustering processing unit 322 is data processing means that executes clustering (classification) processing. The clustering process is a process for grouping documents similar to each other in a document aggregate including a plurality of documents. This clustering process may be either hierarchical clustering or non-hierarchical clustering. As a result of the clustering process, documents (similar documents) having similar descriptions in the set document areas are classified as the same cluster.

  The keyword extraction unit 323 is data processing means having a function of extracting one or a plurality of keywords (search items) that characterize the classified clusters based on the result of the clustering process. For this keyword extraction processing, a known method such as TF-IDF (term frequency-inverse document frequency) can be used. Usually, in the case of a case to be filed, the input unit 12 accepts a keyword indicating a technical feature from a proposal or the like as a search item. However, the keyword extraction unit 323 extracts a common keyword separately from the search item. Enable.

  The search formula creation unit 324 uses a search item that is a keyword received by the input unit 12 and a keyword table stored in the table storage unit 320 to combine a plurality of keywords that belong to each cluster in each area to be searched. Is a data processing means having a function of creating a search expression. The search formula creation unit 324 writes the created search formula into the search formula storage area of the table storage unit 320.

  The search execution unit 325 uses the search formulas stored in the search formula storage area in the table storage unit 320 to search the prior art information and the precedent case information stored in the precedent case DB 20, and sets each search formula. This is a data processing unit having a function of acquiring a corresponding search result (similar document) and writing it in the search result table of the information acquisition unit 30. Here, the search result written in the information acquisition unit 30 is indicated by the document number of the document that satisfies the search expression. As a result, the search execution unit 325 acquires a similar document by a document number.

  Next, the operation of the preceding case search unit 32 will be described with reference to FIG.

  In the precedent case DB 20, for example, information related to a survey target such as a product to be newly developed is described, and a plurality of survey target documents (prior art information and / or precedent case information) in which one or a plurality of items are assigned. Accumulated. In the following description, it is assumed that document numbers A1, A2, A3, A4,..., An such as a project number and an application number are attached to the survey target document. Each investigation target document includes one or a plurality of items corresponding to each description content, and this is a document area to be searched. This document area indicates a concept representing the description contents of the document such as issues, techniques, effects, etc., and a portion indicating the contents of the document area is referred to as a partial document.

  The region setting unit 321 receives one or a plurality of document regions input by the user via the input unit 12, and passes them to the clustering processing unit 322 (S300). Here, description will be made assuming that two paragraphs of “technology” and “issue” are designated as a document area to be searched.

  The clustering processing unit 322 applies the partial document to which the designated “technology” and “issue” document areas in each investigation target document accumulated in the precedent case DB 20 are added to all the investigation target documents for each document area. Collected and clustered for all partial documents 30 for each document area (S301).

  When the clustering process is executed, the clustering processing unit 322 writes the clustering result in the clustering result table in the table storage unit 320 (S302). Here, the partial documents are collected for each document area, but the partial documents are extracted over all the documents to be investigated, and the extracted partial documents are classified for each document area to obtain the partial documents for each document area. It is also possible to perform clustering for each set.

  In this example, the cluster number and the document number included in the cluster are written in the clustering result table of the “technology” document area. Similarly, the clustering result regarding the “issue” document area is also written in the clustering result table. For example, the cluster with cluster number 1 includes document numbers A1, A2, and A5. As a result of clustering, documents with similar descriptions in the document areas of “technology” and “issue” are collected as the same cluster.

  The keyword extraction unit 323 extracts one or a plurality of keywords characterizing each cluster based on the clustering result (S303), and writes it in the keyword table of the preceding case DB 20 (S304). This keyword extraction process uses a known method such as TF-IDF as described above. For example, keywords such as “display”, “liquid crystal display”, “luminance”,... Are written for the cluster with cluster number 1 in the “technical” document area. Further, keywords such as “speed”, “CPU frequency”, “transmission speed”,... Are written for the cluster having the cluster number 101 in the “issue” document area. That is, a plurality of keywords are extracted for each document area. When the keywords are extracted using an index such as TF-IDF, the order of the arrangement in the keyword table of each keyword for each cluster may be regarded as the importance, and the order of the importance may be high. Other orders may be used. Depending on the clustering technique, the clustering process and the keyword extraction can be performed simultaneously.

  The search formula creation unit 324 uses the keyword table in the table storage unit 320 to create a plurality of search formulas that combine keywords belonging to each cluster of each document area (S305), and searches the preceding case DB 20 Writing into the formula storage area (S306). For example, the search expression corresponding to the keyword in the cluster with the cluster number 101 in the document area “issue” and the keyword in the cluster with the cluster number 1 in the item “technology” is “issue: A and A2 and A3” and “technology: α. and α2 and α3 ”. This retrieval formula is to search for documents including A, A2, and A3 from the document area corresponding to the “issue” item, and including α, α2, and α3 from the document area corresponding to the “technology” item. That means. A search expression using “or” instead of “and” may be used. Further, instead of using the three keywords extracted for each cluster, for example, it is possible to create a search expression using the upper two keywords. Alternatively, a search expression may be created using any one or a plurality of keywords.

  The search execution unit 325 uses the search formulas stored in the search formula storage area of the table storage unit 320 to search the precedent case DB 20 in which various documents are stored, and obtains search results corresponding to the search formulas. Obtained and written in the search result table of the table storage unit 320 (S307). This search result is indicated by the document number of the document that satisfies the search expression. This search may be performed on the entire document registered in the precedent case DB 20, but for example, only a part of the documents registered in the precedent case DB 20 is targeted based on the IPC or publication date in the patent document. It is good. When the search is completed, the search execution unit 325 reads the search result from the search result table in the table storage unit 320 and passes it to the prediction information generation unit 36.

  As described above, since the document creation support apparatus 1 according to the present embodiment performs a search using the clustering process, a plurality of information on products to be developed in the future and information on survey targets such as research to be tackled from now on are described. It is possible to search for prior art information and the like having high commonality with a plurality of search items by accumulating the search target documents and specifying the document area to be searched. In addition, if the specification information collected to some extent is input as a search item instead of the search item input from the input unit 12, it is possible to search for prior art information or the like having higher correlation. Furthermore, in the document creation support apparatus 1 according to the present embodiment, in addition to the search item input from the input unit 12, the search item can be extracted from the prior art information, so that a highly accurate search is possible.

  Next, the future prediction unit in the document creation support apparatus 1 will be described in detail. FIG. 8 is a block diagram showing a configuration example of the future prediction unit in the document creation support apparatus according to this embodiment, and FIG. 9 is a flowchart showing the operation of the future prediction unit shown in FIG. The future prediction unit 34a shown in FIG. 8 is an example of simple prediction using a so-called statistical model.

  The future prediction unit 34 a includes an extraction unit 341 and a determination unit 342. The extraction unit 341 is a data processing unit having a function of extracting information related to final disposal included in similar documents from the search result searched by the preceding case search unit 32. The determination unit 342 is a data processing unit having a function of determining a rebate when applying for a search target case based on the extracted information relating to final disposal.

  Next, the operation of the future prediction unit 34a will be described. When the preceding case search unit 32 finishes the search, the extraction unit 341 receives the search result (similar document) (S400).

  Upon receiving the search result, the extraction unit 341 extracts information relating to the final disposal included in the similar document, and identifies the similar document that has been rejected, for example (S401). When the similar document is specified, the extraction unit 341 passes the extraction result to the determination unit 342.

  Upon receiving the extraction result, the determination unit 342 calculates the ratio of similar documents that have been rejected among all similar documents (S402). The determination unit 342 passes the calculated ratio to the prediction information generation unit 36.

  The prediction information generation unit 36 generates display data based on the received ratio and displays it on the display unit 14 (S403).

  As described above, the future prediction unit 34a extracts information related to the final disposal from the searched similar documents, and calculates, for example, the ratio of similar documents that have been rejected from the extraction result. Therefore, it is possible to give a return of a case having similar specification information as a probability. Instead of simply calculating the ratio, the weighted average may be performed in consideration of the similarity obtained in the preceding case search unit 32.

  Next, another example of the future prediction unit will be described. FIG. 10 is a block diagram showing another configuration example of the future prediction unit in the document creation support apparatus according to this embodiment, and FIG. 11 is a flowchart showing the operation of the future prediction unit shown in FIG. The future prediction unit 34b illustrated in FIG. 10 is an example of performing prediction using a so-called decision tree method.

  The future prediction unit 34b includes an extraction unit 341, a determination unit 342, a classification reference generation unit 343, and a classification processing unit 344. Since the extraction unit 341 and the determination unit 342 have the same configuration as the future prediction unit 34a illustrated in FIG. In addition, elements common to the future prediction unit 34a are denoted by the same reference numerals.

  The classification reference generation unit 343 is a data processing unit that generates a condition (reference) for classifying the similar documents searched by the preceding case search unit 32. Specifically, the classification reference generation unit 343 organizes various factors (terms, formal elements, substantive elements, number of characters, etc.) of similar documents searched by the preceding case search unit 32, and the factors and the similar documents It has a function of comparing the information indicating the final disposal corresponding to and calculating the degree of influence of the factor on the final disposal and determining the classification condition. The classification standard generation unit 343 constructs a decision tree (a text processing script indicating what kind of disposal will eventually be reached by repeating the process of whether or not the classification standard applies) based on the determined classification standard. As the decision tree construction algorithm, ID3 / C4.5 / C5.0 or the like can be used.

  The classification processing unit 344 classifies the proposal information (or the detailed description information) to be examined based on the classification standard generated by the classification standard generation unit 343, and finally the final disposal is performed. This is a data processing means for estimating the above. A proposal is a document in which an inventor wrote an invention before making a specification at a patent office or the like. Based on this proposal, a specification as an application document to the JPO is created.

  Next, the operation of the future prediction unit 34b will be described. When the preceding case search unit 32 finishes the search, the extraction unit 341 receives the search result (similar document) (S400).

  Upon receiving the search result, the extraction unit 341 extracts information relating to the final disposal included in the similar document, and identifies the similar document that has been rejected, for example (S401). When the similar document is specified, the extraction unit 341 passes the extraction result to the classification reference generation unit 343.

  Upon receiving the extraction result, the classification criterion generation unit 343 generates a classification factor (term, formal element, substantive element, number of characters, length of sentence, etc.) from all similar documents, and the classification factor gives the final disposal. (Such as the amount of similar literature that was rejected as a result, including the factor). Then, a classification factor having a large influence on final disposal is generated as classification standard information (S410). The classification reference generation unit 343 constructs a decision tree based on the calculated classification reference information and passes it to the classification processing unit 344.

  Based on the received classification reference information and decision tree, the classification processing unit 344 classifies the proposal information and the specification information received by the input unit 12 (S411). As a result, the classification processing unit 344 acquires prediction information indicating what happens as a result of applying for the proposal information and the specification information, and passes the prediction information to the prediction information generation unit 36.

  The prediction information generation part 36 produces | generates display data based on prediction information, and displays it on the display part 14 (S403).

  As described above, according to the future prediction unit 34b, the future prediction is performed based on the retrieved similar documents and the consequences thereof, so that the prediction in consideration of the technical field can be realized.

  Subsequently, another example of the future prediction unit will be further described. FIG. 12 is a block diagram showing another configuration example of the future prediction unit in the document creation support apparatus according to this embodiment, and FIG. 13 is a flowchart showing the operation of the future prediction unit shown in FIG. The future prediction unit 34c illustrated in FIG. 12 is an example of performing prediction using a Bayesian network.

  The future prediction unit 34c includes an extraction unit 341, a determination unit 342, a probability generation unit 345, and a structure application unit 346. Since the extraction unit 341 and the determination unit 342 have the same configuration as the future prediction unit 34a illustrated in FIG. In addition, elements common to the future prediction unit 34a are denoted by the same reference numerals.

  The probability generation unit 345 is a data processing unit that calculates a probability between edges based on a similar literature searched for a processing structure of a Bayesian network prepared in advance. For example, if the term “application time” is used as a classification element, the final disposal included in similar documents will be aggregated according to the end of period, mid-term, end of year, etc. Probability is calculated. If the classification element is “patent classification”, the final disposal included in the similar document is tabulated based on the patent classification of IPC, and the probability of rejection is similarly calculated. In this classification method, the probability is obtained with respect to the relationship between the objective fact and the final disposal, and it is possible to obtain reference information on what kind of description information should be used from the objective fact.

  The structure application unit 346 is a data processing unit that generates a processing script by applying the probability generated by the probability generation unit 345 to the prepared processing structure of the Bayesian network.

  Next, the operation of the future prediction unit 34c will be described. When the preceding case search unit 32 finishes the search, the extraction unit 341 receives the search result (similar document) (S400).

  Upon receiving the search result, the extraction unit 341 extracts information relating to the final disposal included in the similar document, and identifies the similar document that has been rejected, for example (S401). When the similar document is specified, the extraction unit 341 passes the extraction result to the probability generation unit 345.

  Upon receiving the extraction result, the probability generation unit 345 calculates the probability between edges for the processing structure of the Bayesian network prepared in advance based on the extraction result (S420). The probability generation unit 345 passes the obtained probability to the structure application unit 346.

  The structure application unit 346 generates a processing script by applying the received probability to the processing structure of the Bayesian network, and passes the processing script to the determination unit 342 (S421).

  The determination unit 342 applies the processing script that has received the proposal information and the specification information received by the input unit 12, calculates the probability of being finally rejected, and passes the result to the prediction information generation unit 36. The prediction information generation part 36 produces | generates the data for a display based on the received result, and displays it on the display part 14 (S403).

  Thus, according to the future prediction unit 34c, it is possible to predict the outcome of the application as a probability, not as to whether or not to be rejected.

  Subsequently, another example of the future prediction unit will be further described. FIG. 14 is a block diagram showing another configuration example of the future prediction unit in the document creation support apparatus according to this embodiment, and FIG. 15 is a flowchart showing the operation of the future prediction unit shown in FIG. The future prediction unit 34d shown in FIG. 14 is an example of prediction using SVM (Support Vector Machine).

  The future prediction unit 34d includes an extraction unit 341, a determination unit 342, a feature vector generation unit 347, and a boundary calculation unit 348. Since the extraction unit 341 and the determination unit 342 have the same configuration as the future prediction unit 34a illustrated in FIG. In addition, elements common to the future prediction unit 34a are denoted by the same reference numerals.

  The feature vector generation unit 347 is a data processing unit that generates, from various features of the retrieved similar documents, a feature vector using the feature as a parameter. For example, a technical field, a method feature, a feature of substantial contents, etc. are parameterized and defined as a feature vector. This feature vector is regarded as a vector representing the feature of the similar document.

  The boundary calculation unit 348 calculates the boundary between the group rejected in the similar document and the patented group by counting the feature vectors of each generated similar document and comparing it with the final disposal of each similar document. Data processing means. That is, the boundary calculation unit 348 constructs a feature space from a set of feature vectors of similar documents, and calculates where the boundary between rejection and patent exists in the feature space.

  Next, the operation of the future prediction unit 34d will be described. When the preceding case search unit 32 finishes the search, the extraction unit 341 receives the search result (similar document) (S400).

  Upon receiving the search result, the extraction unit 341 extracts information relating to the final disposal included in the similar document, and identifies the similar document that has been rejected, for example (S401). When the similar document is specified, the extraction unit 341 passes the extraction result to the feature vector generation unit 347.

  Upon receiving the extraction result, the feature vector generation unit 347 generates a feature vector for each extracted similar document using a predetermined feature point as a parameter based on the extraction result (S430). The feature vector generation unit 347 passes the obtained feature vector to the boundary calculation unit 348.

  The boundary calculation unit 348 forms a feature space from the received set of feature vectors, and classifies the group into rejected groups and patented groups in comparison with the final disposal result corresponding to each feature vector. Then, the boundary calculation unit 348 specifies the boundary between the rejection and the patent in the feature space (S431). The boundary calculation unit 348 passes information on the feature space and the boundary surface to the determination unit 342.

  The determination unit 342 identifies where in the feature space the proposal information and the specification information received by the input unit 12 exist, and whether or not to reject from the positional relationship between the position and the received boundary surface. Is determined (S432). The determination unit 342 passes the determination result to the prediction information generation unit 36.

  The prediction information generation part 36 produces | generates the data for a display based on the received determination result, and displays it on the display part 14 (S403).

  Thus, according to the future prediction unit 34d, the possibility of rejection can be determined from the feature points of the proposal information and the specification information.

  The present invention is not limited to the above-described embodiment, and various modifications can be made within the scope of the technical idea of the present invention. For example, in the above embodiment, a patent application is taken as an example of the application, but the present invention can also be applied to a utility model registration application, a design registration application, and a trademark registration application. Further, the present invention can be applied not only to document creation related to so-called industrial property rights, but also to document creation that can utilize past document assets.

  Furthermore, the present invention allows various modifications, additions, substitutions, enlargements, reductions, etc. within the scope of the same and equivalent technical ideas. Moreover, even if the apparatus, method, software, and system produced using the present invention are mounted on the secondary product and commercialized, the value of the present invention is not reduced at all.

It is a block diagram which shows the structure of the document preparation assistance apparatus which concerns on embodiment of this invention. It is a flowchart which shows the new document creation assistance operation | movement in the document creation assistance apparatus which concerns on this embodiment. It is a figure which shows the example of a display of the input screen in new document preparation assistance operation | movement. It is a figure which shows the example of a display of the result screen in new document preparation assistance operation | movement. It is a flowchart which shows the response document creation assistance operation | movement in the document creation assistance apparatus which concerns on this embodiment. It is a block diagram which shows the structure of the precedent case search part in the document preparation assistance apparatus which concerns on this embodiment. It is a flowchart which shows operation | movement of the precedent case search part shown in FIG. It is a block diagram which shows the structural example of the future prediction part in the document preparation assistance apparatus which concerns on this embodiment. It is a flowchart which shows operation | movement of the future estimation part shown in FIG. It is a block diagram which shows the other structural example of the future estimation part in the document preparation assistance apparatus which concerns on this embodiment. It is a flowchart which shows operation | movement of the future estimation part shown in FIG. It is a block diagram which shows the other structural example of the future estimation part in the document preparation assistance apparatus which concerns on this embodiment. It is a flowchart which shows operation | movement of the future estimation part shown in FIG. It is a block diagram which shows the other structural example of the future estimation part in the document preparation assistance apparatus which concerns on this embodiment. It is a flowchart which shows operation | movement of the future estimation part shown in FIG.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Document creation assistance apparatus, 2 ... Network, 3 ... Document server, 10 ... Communication part, 12 ... Input part, 14 ... Display part, 16 ... Memory | storage part, 18 ... Calculation part, 20 ... Precedent case database, 22 ... Case Database, 24 ... Response case database, 30 ... Information acquisition unit, 32 ... Preceding case search unit, 34 ... Future prediction unit, 36 ... Prediction information generation unit, 40 ... Response case search unit, 42 ... Response information generation unit.

Claims (4)

Information related to the description of the application case, notification case information indicating the content of the command for the application case, response case information including response information indicating the response content to the notification case information of the application case, and final disposal of the application case A storage unit that stores preceding case information including information for each application case;
An input unit for receiving a search item indicating characteristics relating to a new application case and notification information indicating a command content for a predetermined application case;
The search item received by the input unit is used as a search key, the similar document information having characteristics relating to the new application case is searched from the storage unit, and the notification information received by the input unit is used as the search key. A search unit for searching the response case information including the notification case information of the content common to the notification information related to the predetermined application case from a unit;
The information related to the final disposal is extracted from the similar document information searched by the search unit, and the ratio of rejection judgment among the entire similar document information is calculated based on the information related to the final disposal. Based on the ratio A predicting unit for predicting the final disposal by determining whether the final disposal of the new application is probabilistically higher than permission or rejection;
A document creation support apparatus, comprising: a display unit that displays a prediction result of the prediction unit and response case information searched by the search unit. Information related to the description of the application case, notification case information indicating the content of the command for the application case, response case information including response information indicating the response content to the notification case information of the application case, and final disposal of the application case A server device comprising a storage unit for storing the preceding case information including information for each application case;
An input unit for receiving a search item indicating characteristics relating to a new application case and notification information indicating a command content for a predetermined application case;
The search item received by the input unit is used as a search key, the similar document information having characteristics relating to the new application case is searched from the storage unit, and the notification information received by the input unit is used as the search key. A search unit for searching the response case information including the notification case information of the content common to the notification information related to the predetermined application case from a unit;
The information related to the final disposal is extracted from the similar document information searched by the search unit, and the ratio of rejection judgment among the entire similar document information is calculated based on the information related to the final disposal. Based on the ratio A predicting unit for predicting the final disposal by determining whether the final disposal of the new application is probabilistically higher than permission or rejection;
A document creation support system comprising: a terminal device including a display unit that displays a prediction result of the prediction unit and response case information searched by the search unit.   The document according to claim 1, wherein the response case information stored in the storage unit further includes memo information including a review material for creating a response content corresponding to the notification case information of the application case. Creation support device.   The response case information stored in the storage unit further includes, as history information, information associated with the application case when creating a response content corresponding to the notification case information of the application case. Item 1. The document creation support device according to Item 1.

Images