JP6925611B2 - Literature search system - Google Patents

Description

The present invention relates to a document search system, particularly a document search system for searching a document containing at least digitized textual information.

Conventionally, there is known a document search system in which a user searches for a document by inputting a keyword into a client terminal (see Patent Document 1). In the technique of Patent Document 1, the document of patent information is searched based on the character information (keyword) input by the user.

JP-A-2007-199987

In the document search system of Patent Document 1, it is necessary for the user to appropriately select the character information to be input. Therefore, the search result of the document varies depending on the character information input by the user. In addition, it takes time and effort to select character information until an appropriate search result is obtained.

An object of the present invention is to make it possible to search for an appropriate document without the trouble of inputting characters.

The document search system according to the present invention is a system for searching documents containing at least digitized textual information. The document search system includes a data sheet, a reading unit, and a conversion unit. In the data sheet, at least the item, the function of the item, the failure mode based on the function, and the countermeasure corresponding to the failure mode are input in character information. The reading unit reads the character information of the data sheet. The conversion unit converts the document into a searchable character string based on the character information read by the reading unit.

In this document search system, the reading unit automatically reads the character information based on the character information of the data sheet. Based on the character information read by the reading unit, the conversion unit converts it into a search character string. As a result, a search character string for searching a document can be automatically created based on the data sheet. As a result, in the document search system, an appropriate document can be searched without the trouble of inputting characters.

The document search system may further include a synonym search unit that searches for synonyms of character information read by the reading unit. Further, the conversion unit may convert the character information read by the reading unit and the synonyms searched by the synonym search unit into a search character string.

With such a configuration, it is possible to search for the character information input to the data sheet and the documents including the synonyms thereof. As a result, the literature can be searched with higher accuracy.

The literature may include at least problems and solutions. The conversion unit may convert the countermeasure into a search character string corresponding to the solution. The conversion unit may convert the failure mode into a search character string corresponding to the problem.

With such a configuration, a search character string is created in response to the countermeasures entered in the data sheet, so a search for documents containing character information related to the countermeasures entered in the data sheet is automatically performed. Can be done as a target. In addition, in order to create a search character string corresponding to the failure mode entered in the data sheet, a document containing character information related to the failure mode entered in the data sheet should be automatically searched. Can be done.

The search character string may be a search formula used for searching a document. With such a configuration, it is possible to search for documents using a search formula. As a result, more complex searches can be performed automatically.

The document search system may further have a risk evaluation unit that performs risk evaluation based on the result of searching with the search character string.

With such a configuration, the risk can be evaluated from the text information related to the countermeasure entered in the data sheet.

According to the present invention, it is possible to perform an appropriate document search without the trouble of inputting characters.

The figure of the document search system of this invention. An example diagram of a data sheet. The figure which arranged the character information of the data sheet 4. The figure which showed the list of the literature search result. The figure which added the judgment to the list of the literature search results. The figure which added the risk column to data sheet 4. The flowchart in this embodiment. Data sheets in other embodiments.

Hereinafter, embodiments will be described with reference to the drawings.

As shown in FIG. 1, the document search system 2 includes a data sheet 4, a reading unit 6, a synonym search unit 8, a conversion unit 10, a risk evaluation unit 12, a corporate information database 14, and a patent classification search unit. It has 16, a display unit 18, an information retrieval server interface 20, and an Internet line interface 22.

The document retrieval system 2 is connected to the information retrieval server S via the Internet line 24.

The information retrieval server S has a server-side Internet line interface 26 and an information retrieval unit 28.

The document search system 2 can be configured by any of hardware, DSP (Digital Signal Processor), and software. For example, when configured by software, the document search system 2 is actually configured to include the CPU, RAM, ROM, etc. of the computer 1, and is stored in each part stored in a recording medium such as RAM, ROM, hard disk, or semiconductor memory. It is realized by running the corresponding program.

As shown in FIG. 2, the data sheet 4 is a sheet using a format for FMEA (Failure Mode and Effect Analysis), which is used for design review when developing a product, for example. The data sheet 4 contains at least item A, function F1 of item A, failure mode F2 based on the function, influence F3 of failure, influence degree F3A obtained by quantifying them, frequency F3B, and detection difficulty level F3C. And, the column of the countermeasure content F4 corresponding to the failure mode and the failure cause F5 is provided. Character information is entered in each field of the data sheet 4. The data sheet 4 is data created by commercially available spreadsheet software or a word processor. Textual information means words, phrases, and sentences consisting of letters. Here, the letters include numbers and symbols.

As shown in FIG. 2, the data sheet 4 is used, for example, in a design review of crankshaft bearings. As shown in FIG. 2, the character information "crankshaft bearing" is input in the item A column. In the column of function F1, the text information of "holding the crankshaft rotatably", "receiving the explosive load transmitted from the crankshaft", and "lubrication between the crankshaft and the crankshaft bearing" is input. .. In the failure mode F2, the text information "bearing galling" and "bearing corrosion" are input separately in the fields.

As shown in FIG. 1, the reading unit 6 is a program that executes text mining and the like. The reading unit 6 reads the character information input in each field of the data sheet 4. The reading unit 6 reads the character information in each column of the data sheet 4, extracts each word such as a verb, a noun, and an adjective from the character information, and arranges the characters.

The synonym search unit 8 is a program that searches for synonyms of words extracted by the reading unit 6.
A database of synonyms is stored in advance in the recording medium of the computer 1. The synonym search unit 8 searches for synonyms from the synonym database based on the words extracted by the reading unit 6, and arranges them together with the character information read by the reading unit 6.

The conversion unit 10 converts the character information read by the reading unit 6 and the synonyms searched by the synonym search unit 8 into a search character string that can be searched by the information search server S. The search character string may be a search expression. As shown in FIG. 3, the conversion unit 10 uses the character information read by the reading unit 6 and the synonyms searched by the synonym search unit 8 as an item group, a problem group P, a problem group Q, and a solution group. Divide into and align. More specifically, the conversion unit 10 arranges the character information in the column of the countermeasure content F4 of the data sheet 4 in the solution means group as related character information. The conversion unit 10 also arranges synonyms of the character information in the column of the countermeasure content F4 in the solution group. The conversion unit 10 assigns an identification code such as F41 to the character information divided into each group and arranged. The conversion unit 10 assigns an identification code such as F411 to a synonym of F41, for example.

The conversion unit 10 creates a search character string (also referred to as a search formula) based on the identification code. For example, the conversion unit 10 can search for documents including character information of both A1 "crankshaft" and A2 "bearing" for the item group A1 "crankshaft" and A2 "bearing". Create a search string or search expression in. Further, the conversion unit 10 includes a document including character information of either A1 "crankshaft" or A11 "crank mechanism" for the item group A1 "crankshaft" and its synonym A11 "crank mechanism". Create a search string or search expression so that can be searched. Further, the conversion unit 10 creates, for example, a search character string or a search formula that can be searched so that the character information of the item group corresponds to the title of the invention of the patent document or the patent classification stored in the patent classification search unit 16. do. Similarly, the conversion unit 10 can search the search characters so that the character information of the problem group P or the problem group Q can search the background technology of the patent document or the patent document described in the column of the problem to be solved by the invention. Create a column or search expression. The conversion unit 10 creates a search character string or a search formula so that the character information of the solution group can search the patent document described in the column of the problem solving means of the invention in the patent document. The conversion unit 10 transmits the search character string or the search formula to the information search server interface 20. The search character string or search expression input to the information search server interface 20 is sent to the information search server S via the Internet line interface 22 and the server-side Internet line interface 26.

The information retrieval server S can be configured by any of hardware, DSP (Digital Signal Processor), and software. For example, when configured by software, the information retrieval server S is actually configured to include a CPU, RAM, ROM, etc., and a program corresponding to each part stored in a recording medium such as RAM, ROM, hard disk, or semiconductor memory is provided. It is realized by working. The information retrieval server S sends a search character string or a search expression via the server-side Internet line interface 26 to the information retrieval unit 28.

Based on the search character string or the information of the search formula, the information retrieval unit 28 searches for a document including at least the digitized character information stored in the information retrieval server S, or at least the digitized character information on the Internet. Search for literature that contains. In the present embodiment, the information retrieval unit 28 is a program for searching patent documents. The information retrieval unit 28 searches for a patent document stored in the information retrieval server S or a patent document on the Internet from a search character string or a search formula, and then, for example, as shown in FIG. 4, the patent number and the filing date. And make a list of applicants. The information retrieval unit 28 transmits the patent number, the filing date, and the list of applicants to the document retrieval system 2 via the server-side Internet line interface 26 and the Internet line interface 22.

The risk evaluation unit 12 is a program that performs risk evaluation based on the results searched by the information search server S. The risk evaluation unit 12 determines the risk of the result searched by the information search server S based on the result searched by the information search server S, the character information read by the reading unit 6, and the evaluation method programmed in advance. evaluate.

The company information database 14 is a database in which company information is stored. The company information database 14 stores at least the company name, information on the bargaining power of the company, and information on the number of patents held.

The patent classification search unit 16 is a program that searches for related patent classifications based on the character information in the item A column read by the reading unit 6. The patent classification search unit 16 stores patent classifications such as CPC and IPC and data of corresponding items. For example, the information "IPC = F16C9 / 00: crank bearing or connecting rod or similar link" is stored. The patent classification search unit 16 searches for the corresponding patent classification based on the character information in the item A column and the synonyms of the character information in the item A column according to the instruction from the conversion unit 10. For example, the patent classification search unit 16 matches or one character from the character information of "A1 crankshaft A2 bearing" of the item group shown in FIG. 3 and the character information of the item group synonym "A11 crank mechanism, A21 bearing". Search for the corresponding patent classification of 16C9 / 00 that matches the part. When the patent classification search unit 16 finds the corresponding patent classification, the patent classification search unit 16 transmits the patent classification to the conversion unit 10.

The display unit 18 is a program that displays the result of evaluation by the risk evaluation unit 12 and the character information read by the reading unit 6 on the computer 1. For example, as shown in FIG. 5, the display unit 18 adds a comment field and a determination to the search result of the information retrieval server S and displays the result. Further, as shown in FIG. 6, the display unit 18 also adds a risk column to the data sheet 4 to display the degree of risk. Further, for example, as shown in FIG. 3, the display unit 18 arranges and displays the character information read by the reading unit 6 and the synonyms of the character information read by the reading unit 6.

Next, the control flow of the document search system 2 will be described with reference to the flowchart of FIG. 7.

The document search system 2 starts the program when the icon displayed on the screen of the computer 1 is selected by the user. Further, in the present embodiment, a case where the data sheet 4 relating to the “crankshaft bearing” shown in FIG. 2 is created by commercially available spreadsheet software will be described. Further, it is assumed that the data sheet 4 is stored in the computer 1.

In step S1, the document search system 2 activates the reading unit 6 to read the character information of the data sheet 4. When the reading unit 6 is activated, the document search system 2 displays the data sheet 4 stored in the computer 1 via the display unit 18 of the computer 1 so that the user can select it. When the user selects the data sheet 4, the document search system 2 reads the character information of the data sheet 4 by the reading unit 6. The reading unit 6 has characters in the item A column of the data sheet 4, the function F1, the failure mode F2 column, the failure effect F3 column, the countermeasure content F4 column, and the failure cause F5 column. Read all the information. When the reading unit 6 completes reading the character information of the data sheet 4, the process proceeds to step S2.

In step S2, the synonym search unit 8 searches for synonyms from the synonym database stored in the computer 1 based on the character information read in step S1. A database of synonyms, for example, for the text information "crankshaft bearing", is "crank mechanism", "crankshaft bearing", "C / S" which is an abbreviation for crankshaft, and "C / S" which is an abbreviation for bearing. "BRG" and the like are stored in association with the character information "crankshaft bearing". For example, when the reading unit 6 reads the character information of "crankshaft bearing", the synonym search unit 8 searches for a synonym of "crankshaft bearing" from the database of synonyms stored in the computer 1. When the synonym search unit 8 has searched for all the synonyms of the character information in the data sheet 4, the process proceeds to step S3.

In step S3, as shown in FIG. 3, the conversion unit 10 uses the character information read by the reading unit 6 and the synonyms searched by the synonym search unit 8 as an item group, a task group P, and a task group Q. , Divide into commentary means groups and arrange them. The conversion unit 10 arranges the character information "crankshaft bearing" described in the item A column of the data sheet 4 as an item group. Further, the character information of "crank mechanism" and "C / S", which are synonyms of "crankshaft bearing", is arranged as a synonym of item group. The conversion unit 10 assigns an identification code to the aligned character information. That is, the conversion unit 10 assigns the identification code A1 to the character information "crankshaft". The conversion unit 10 assigns the identification code A2 to the character information "bearing". Similarly, the conversion unit 10 assigns identification symbols such as A11 to the "crank mechanism" and A12 to the "C / S" for the item group synonyms. Here, the identification code is set according to the name of each column of the data sheet 4, such as "A" for the item A and "F1" for the function F1 in the first layer. Further, the identification code of the second layer is set according to the character information in the same column of each column. In the present embodiment, for example, two nouns "crankshaft" and "bearing" are input as character information in the item A column. The conversion unit 10 divides these nouns into two parts, assigns the identification code A1 to the "crankshaft", and assigns the identification code A2 to the "bearing". Further, the identification code of the third layer is given as a synonym corresponding to the character information up to the second layer. That is, the "crank mechanism", which is a synonym for the "crankshaft" A1, is given the identification code A11 in common up to the second layer. Similarly, the conversion unit 10 has problems with the character information and synonyms of the function F1 of the data sheet 4, the column of the failure mode F2, the column of the failure effect F3, the column of the countermeasure content F4, and the column of the failure cause F5. Align with group P, problem group Q, and solution group. The conversion unit 10 assigns an identification code to the task group P, the task group Q, and the solution means group in the same manner as the item group, and proceeds to step S4.

In step S4, the conversion unit 10 searches the item group, the task group P, the task group Q, and the solution group using a search character string or a program for creating a search formula stored in advance. Create a string or search expression. In the program stored in advance, a search character string or a search formula is created based on the identification code. In the present embodiment, for example, for A1 "crankshaft" and A2 "bearing", the first layer is "A" in the item group, which is common. However, it differs between "1" and "2" in the second layer. Therefore, the conversion unit 10 creates a search character string or a search formula so that the information search server S can search a document including both the "crankshaft" and the "bearing". Further, the A1 "crankshaft" and the A11 "crank mechanism" are common to the second layer in the item group A. Therefore, the conversion unit 10 creates a search character string or a search formula so that the information search server S can search a document including either A1 "crankshaft" or A11 "crank mechanism". That is, a program that creates a search character string or a search formula stored in advance searches for a document containing both character information by the hierarchy of the common part and the common part of the identification code, or either. It is decided whether to search the literature containing the character information of either one. In the present embodiment, for example, in the "crankshaft" A1 and the "crank mechanism" A11, the item group up to the second layer "A1" is common, and the third layer is different. In this case, the program for creating the search character string or the search formula is programmed so that the documents including the character information of either one of "A1" and "A11" are searched. Further, the identification code of the second layer is different between the "crankshaft" A1 and the "bearing" A2. In this case, "A1" and "A2" are programmed to create a search character string or a search formula for searching a document containing both character information. Similarly, a search character string or a search formula is created for the problem group P, the problem group Q, and the solution group.

Next, the conversion unit 10 determines which item of the literature information to be searched corresponds to each group of the item group, the problem group P, the problem group Q, and the solution means group. The conversion unit 10 creates a search character string or a search formula so as to correspond to the determined item. That is, a search character string or a search formula is created in which the search condition defined by the identification code corresponds to which item of the document to be searched. In this embodiment, the character information and synonyms of the item group are pre-programmed so as to correspond to the names of the inventions in the patent document. In this case, the converter should search for documents in which both letters of the synonym "crankshaft" or "crankshaft" and the synonym "bearing" or "bearing" of the item group are included in the title of the invention. , Create a search string or search expression. Similarly, the conversion unit 10 is programmed so that the character information and synonyms of the problem group P or the problem group Q correspond to the background technology of the patent document and the patent document described in the column of the problem to be solved by the invention. .. Further, the conversion unit 10 is pre-programmed so that the character information and synonyms of the solution means correspond to the patent documents described in the column of the problem-solving means of the invention of the patent document. The conversion unit 10 creates a search character string and a search formula so that the problem group P, the problem group Q, and the solution group correspond to each item of the patent document as in the item group. The conversion unit 10 transmits the search character string or the search formula to the information search server interface 20. The search character string or search expression input to the information retrieval server interface 20 is sent to the information retrieval server S via the Internet line interface 22 and the server-side Internet line interface 26, and the process proceeds to step S5.

In step S5, the information search unit 28 of the information search server S searches for documents using the search character string or search formula created by the conversion unit 10. In the present embodiment, the patent document corresponding to the search character string or the search formula created in step S4 is searched. The information retrieval unit 28 creates a list of patent numbers, filing dates, and applicants, for example, as shown in FIG. The information retrieval unit 28 transmits the patent number, the filing date, and the list of applicants to the document retrieval system 2 via the server-side Internet line interface 26 and the Internet line interface 22, and proceeds to step S6.

In step S6, the risk evaluation unit 12 receives the search result performed by the information retrieval server S, and when the reception is completed, proceeds to step S7.

In step S7, the risk evaluation unit 12 performs a risk evaluation based on the result of the search on the information search server S. The risk evaluation unit 12 refers to the result searched by the information search server S, the character information read by the reading unit 6, and the result searched by the information search server S based on the evaluation method programmed in advance. Assess risk. In the present embodiment, the risk evaluation unit 12 bases the character information of each group on the result of the search on the information search server S (see FIG. 4) based on the degree of matching of the character information described in the patent document. Assess risk. For example, more textual information or synonyms arranged in the solution group are included in the corresponding item of the solution means of the patent document searched by the information search server S (in the present embodiment, the column of the problem solving means of the invention). The more the literature is described, the higher the risk is evaluated. Further, the character information or synonyms arranged in the task group P and the task group Q correspond to the problem group P or the task group Q of the patent document searched by the information retrieval server S (in the present embodiment, the background technique, The more documents described in (Problems to be Solved by the Invention), the higher the risk. Further, the risk evaluation unit 12 compares the item of the applicant as a result of the search on the information search server S with the company information recorded in the company information database 14, and states that the risk is higher for the company with a larger number of patents. evaluate. Further, the risk evaluation unit 12 refers to the numerical values of the influence degree F3A, the occurrence frequency F3B, and the detection difficulty level F3C read by the reading unit 6. The risk evaluation unit 12 determines that the higher the value of the degree of influence F3A, the higher the risk. After performing the risk evaluation, the risk evaluation unit 12 proceeds to step S8.

In step S8, as shown in FIG. 5, the display unit 18 creates and displays a determination column after performing a relative risk evaluation on the result searched by the information search server S. As shown in FIG. 6, the display unit 18 further creates a risk column in the data sheet 4 and displays the risk corresponding to the countermeasure content F4. At this time, the display unit 18 displays the determination that the information search server S has evaluated as having the highest risk with respect to the search result in the risk column shown in FIG.

As described above, in the document search system 2 of the present invention, the reading unit 6 can automatically read the character information based on the character information of the data sheet 4. Further, the conversion unit 10 can automatically create a search character string based on the character information read by the reading unit 6. As a result, it is possible to create a search character string that automatically searches the literature based on the data sheet. Therefore, the trouble of inputting characters by the user can be saved. Further, in the data sheet 4, the item A, the function F1, the failure mode F2, and the countermeasure content F4 are input as character information. Therefore, it is possible to create a search character string for automatically searching a document from the data sheet 4 used for design review and the like. As a result, not only the time and effort of the user can be saved, but also an appropriate document can be searched without relying on the work of selecting the character to be input by the user by using a tool for ensuring design quality such as a design review. Therefore, for example, in a search for documents in a patent search or the like, it is possible to suppress variations in search results and reduce the risk of patent infringement due to omission of searches due to variations.

Further, in the document search system 2 of the present invention, the synonym search unit 8 can search for character information input to the data sheet 4 and documents including those synonyms. As a result, it is possible to search for documents with higher accuracy. Therefore, for example, when a patent search is conducted from a design review, the risk of patent infringement due to omission of the patent search can be reduced.

The conversion unit 10 automatically searches for a document containing character information related to the countermeasure input in the data sheet in order to create a search character string corresponding to the countermeasure content F4 input in the data sheet 4. Can be done as a target. Further, in order to create a search character string corresponding to the failure mode F2 input to the data sheet 4, the search of the document in which the character information related to the failure mode F2 input to the data sheet 4 is described is automatically performed. Can be done As a result, for example, the related documents can be automatically searched from the character information related to the failure mode F2 and the corresponding countermeasure content F4 input in the data sheet 4 such as the design review. Therefore, for example, when searching for a document in a patent search or the like, the failure mode F2 input to the data sheet 4 and the patent document related to the countermeasure content F4 can be automatically searched.

Further, the document search system 2 can search for documents using a search formula. As a result, more complex searches can be performed automatically.

In the document search system 2, the risk evaluation unit 12 can be used to evaluate the risk from the character information related to the countermeasure content F4 input to the data sheet 4. As a result, for example, when searching for a document in a patent search or the like, it is possible to evaluate the risk of patent infringement due to omission of the search for the patent document related to the countermeasure content F4 input in the data sheet 4. In the risk evaluation unit 12, the text information or synonyms arranged in the solution group correspond to the corresponding item of the solution means of the patent document searched by the information search server S (in the present embodiment, the column of the problem solving means of the invention). In addition, the more literature is described, the higher the risk is evaluated. That is, the risk evaluation unit 12 evaluates a document in which a large number of characters matching the character information of the countermeasure content F4 of the data sheet 4 are described by the problem-solving means of the invention of the patent document as high risk. With such an evaluation, it is possible to prevent the product using the countermeasure content F4 described in the data sheet 4 from infringing the patent by being manufactured and sold.

Further, in the risk evaluation unit 12, the character information or synonyms arranged in the task group P and the task group Q correspond to the problem group P or the task group Q of the patent document searched by the information retrieval server S (this implementation). In terms of form, the more documents described in the background technology, the problem to be solved by the invention), the higher the risk is evaluated. That is, the character information of the failure mode F2 of the data sheet 4 relates to the defect of the item A input to the data sheet 4. As a result, when the characters described in the problem to be solved by the invention of the patent document and the character information related to the failure mode F2 of the data sheet 4 match more, the patent document is a document related to the same defect. There is a high probability. Therefore, there is a high risk of infringing the patent by manufacturing and selling a product using the countermeasure content F4 for the failure mode F2 of the item A of the data sheet 4. With such an evaluation, it is possible to determine the possibility that the product using the character information of the data sheet 4 infringes the patent right. Therefore, it is possible to prevent the product using the character information of the data sheet 4 from infringing the patent by being manufactured and sold.

Further, the risk evaluation unit 12 compares the item of the applicant as a result of the search on the information search server S with the company information recorded in the company information database 14, and states that the risk is higher for the company with a larger number of patents. evaluate. That is, when the applicant of the patent document in which the character information of the data sheet 4 matches more has filed a large number of patent applications, it is related to the item A of the data sheet 4, the failure mode F2, and the countermeasure content F4. The probability of holding a large number of patent rights increases. As a result, there is a high probability that the product using the character information described in the data sheet 4 will infringe the patent right by being manufactured and sold. With such an evaluation, it is possible to determine the risk that the product using the character information of the data sheet 4 infringes the patent right. Therefore, it is possible to prevent the product using the character information of the data sheet 4 from infringing the patent by being manufactured and sold.

Further, the risk evaluation unit 12 refers to the numerical values of the influence degree F3A, the occurrence frequency F3B, and the detection difficulty level F3C read by the reading unit 6. The risk evaluation unit 12 determines that the higher the value of the degree of influence F3A, the higher the risk. That is, when the degree of influence F3A of the failure mode F2 is a high value, there is a high probability that the countermeasure content F4 for the failure mode F2 is an important element that makes it difficult for the product using the character information of the data sheet 4 to change the design. .. Therefore, if the characters described in the patent document and the character information of the data sheet 4 match more, the patent right of another person becomes an obstacle in ensuring the quality of the product using the character information of the data sheet 4. The risk is high. The risk evaluation unit 12 determines that the higher the value of the degree of impact F3A, the higher the risk. Therefore, the risk evaluation unit 12 automatically evaluates the risk before the product such as design review is manufactured and sold, and the data sheet 4 It is possible to prevent infringement of patents by manufacturing and selling products using the character information of.

<Other embodiments>
(A) In the above embodiment, the data sheet 4 has been described by taking as an example a data sheet in which character information is input, but the present invention is not limited to this, and various changes can be made without departing from the gist of the invention. It is possible. For example, the data sheet 4 is a handwritten character, and the reading unit 6 may read the handwritten character by the character recognition function.

(B) In the above embodiment, the character information and synonyms of the item A are arranged in the item group, the character information and the synonyms of the function F1 and the failure mode F2 are arranged in the task group P, and the failure influence F3 and the failure cause F5 are arranged. It was arranged in the problem group Q, and the countermeasure content F4 was arranged in the solution group, but it is not limited to this, the item A is arranged in the item column, the failure mode F2 is arranged in the problem group, and the countermeasure content is arranged. F4 may be aligned with the solution group. The other failure influence F3, failure cause F5, and function F1 may be arranged in any of the above groups.

(C) In the above embodiment, the patent number, the filing date, and the list of applicants are given as an example, but the patent number, the filing date, the applicant, the abstract, and the contents described in the specification may be added to the list. good.

(D) In the above embodiment, the search character string or the search formula is created from the character information of the item A, but the patent classification searched by the patent classification search unit 16 may be used. In this case, the patent classification is used in the search string or search formula. Further, the conversion unit 10 creates a search character string or a search formula that can be searched so as to apply to the patent classification stored in the patent classification search unit 16.

(E) In the above embodiment, the data sheet 4 has been described using the format for FMEA, but the present invention is not limited to this, and for example, the data sheet 104 having the format shown in FIG. 8 may be used. In this case, the column of "parts" corresponds to item A, the column of "contents of concern related to change" corresponds to the failure mode F2, and the column of "when the concern occurs" corresponds to the cause of failure F5. In addition, the column of "impact on customers" corresponds to the failure impact F3. The column of "What kind of design was made to eliminate anxiety" corresponds to the countermeasure content F4.

2: Document search system 4,104: Data sheet 6: Reading unit 8: Synonym search unit 10: Conversion unit 12: Risk evaluation unit A: Item F1: Function F2: Failure mode F4: Countermeasures

Claims (5)

A document search system that searches for patent documents that include at least digitized textual information.
A data sheet in which at least an item, a function of the item, a failure mode based on the function, and a countermeasure corresponding to the failure mode are input in character information.
A reading unit that reads the character information of the data sheet,
A conversion unit that converts the document into a searchable character string based on the character information read by the reading unit, and a conversion unit.
Literature search system with. Further provided with a synonym search unit for searching for synonyms of character information read by the reading unit.
The document search system according to claim 1, wherein the conversion unit converts the character information read by the reading unit and the synonyms searched by the synonym search unit into the search character string. The patent document describes at least problems and solutions.
The conversion unit
The countermeasure is converted into the search character string corresponding to the solution, and the countermeasure is converted into the search character string.
The document search system according to claim 1 or 2, wherein the failure mode is converted into the search character string corresponding to the problem. The document search system according to any one of claims 1 to 3, wherein the search character string is a search formula used for searching the patent document.


The document search system according to any one of claims 1 to 4, further comprising a risk evaluation unit that performs risk evaluation based on the result of searching with the search character string.

Images