Information processing method, program, and information processing device

Abstract

It is easy to confirm whether or not a structure described in a file is sufficiently associated with an effect or the like. In one embodiment of the present invention, file data is received, a text included in the file data is decomposed into a plurality of sentences, sentences of a first classification corresponding to a structure recorded and without an effect recorded and sentences of a second classification corresponding to an effect recorded and without a structure recorded are extracted from each of the plurality of sentences, and one or more groups are created. The group comprises at least one sentence equivalent to the first classification and at least one sentence equivalent to the second classification which correspond to each other, and the sentences which do not belong to any group in the sentences equivalent to the first classification are shown as sentences equivalent to the third classification.

Description

Technical Field

[0001] One embodiment of the present invention relates to an information processing method, a program, and an information processing device. Another embodiment of the present invention relates to an information processing method, a program, and an information processing device related to the specification of the patent application. Background Art

[0002] Examples of patent-related tasks include prior art research, the preparation of patent application documents, patent rights conversion, and invalidation research. Due to the wide variety of patent-related tasks, systems supporting these tasks have been developed in recent years, including patent application document preparation support systems, patent information analysis systems, and patent search systems. Patent Document 1 discloses an application document preparation support system that extracts and displays claims containing input keywords. [Prior technical literature] [Patent Document]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2012-48696 Summary of the Invention Technical problem to be solved by the invention

[0004] One object of one embodiment of the present invention is to easily confirm whether a structure described in a document is sufficiently correlated with relevant information about that structure. Another object of one embodiment of the present invention is to easily confirm whether a structure described in a document is sufficiently correlated with the effects of that structure. Another object of one embodiment of the present invention is to provide a novel information processing device or information processing method related to the specification of a patent application. Another object of one embodiment of the present invention is to support the production of high-quality specifications.

[0005] Note that the inclusion of these objectives does not preclude the existence of other objectives. One embodiment of the present invention does not necessarily achieve all of the above objectives. Objectives other than the above objectives may be extracted from the description, drawings, and claims. Means of solving technical problems

[0006] One embodiment of the present invention is an information processing method, comprising the following steps: receiving file data; decomposing the text included in the file data into multiple sentences; extracting sentences corresponding to the first category that record a structure but not a effect and sentences corresponding to the second category that record an effect but not a structure from the multiple sentences; creating one or more groups, each group including at least one sentence corresponding to the first category and at least one sentence corresponding to the second category; and displaying sentences that do not belong to any group among the sentences corresponding to the first category as sentences corresponding to the third category.

[0007] Preferably, sentences corresponding to the first category are compared with a database to obtain sentences corresponding to the fourth category in which corresponding effects are described, and groups are created by comparing sentences corresponding to the second category with sentences corresponding to the fourth category.

[0008] Preferably, sentences corresponding to the fourth category in which the corresponding effects are described are obtained by comparing each sentence corresponding to the third category with a database, and the sentences corresponding to the fourth category are displayed.

[0009] Preferably, the result of comparing the combination of the sentence corresponding to the first category and the sentence corresponding to the second category in the group with the database is shown.

[0010] Preferably, the first classifier is used when extracting sentences equivalent to the first category and sentences equivalent to the second category, sentences equivalent to the first category and sentences equivalent to the second category are displayed, an evaluation of at least one sentence equivalent to the first category or at least one sentence equivalent to the second category is received, and the combination of the evaluated sentences and the evaluation is used as learning data to perform first classifier learning.

[0011] Preferably, the second classifier is used when creating the groups, the groups are displayed, evaluations of at least one group are received, and a combination of the evaluated groups and the evaluations is used as learning data to perform learning of the second classifier.

[0012] One embodiment of the present invention is a program having a function of causing a processor to execute any of the above-mentioned information processing methods.

[0013] One embodiment of the present invention is an information processing device comprising a receiving unit, a processing unit, and an output unit. The receiving unit has the function of receiving file data. The processing unit has the following functions: a function of decomposing text included in the file data into a plurality of sentences; a function of extracting sentences corresponding to a first category that record a structure but not a effect; a function of extracting sentences corresponding to a second category that record an effect but not a structure; a function of creating a group including at least one sentence corresponding to the first category and at least one sentence corresponding to the second category; and a function of extracting sentences corresponding to the first category that do not belong to any group as sentences corresponding to a third category. The output unit has the function of outputting sentences corresponding to the third category.

[0014] Preferably, one embodiment of the present invention includes a database storing corresponding combinations of sentences describing structures and sentences describing effects.

[0015] Preferably, the processing unit has the following functions: obtaining sentences corresponding to the fourth category in which corresponding effects are recorded by comparing each sentence corresponding to the first category with the database; and creating groups by comparing each sentence corresponding to the second category with sentences corresponding to the fourth category.

[0016] Preferably, the processing unit obtains sentences corresponding to the fourth category in which the corresponding effects are described by comparing each sentence corresponding to the third category with a database, and displays the sentences corresponding to the fourth category.

[0017] Preferably, the processing unit displays a result of comparing a combination of a sentence corresponding to the first category and a sentence corresponding to the second category in the group with a database. Effects of the Invention

[0018] One embodiment of the present invention makes it easy to confirm whether the structure described in a document is sufficiently correlated with the relevant information about that structure. Furthermore, one embodiment of the present invention makes it easy to confirm whether the structure described in a document is sufficiently correlated with the effects of that structure. Furthermore, one embodiment of the present invention provides a novel information processing device or method for a specification related to a patent application. Furthermore, one embodiment of the present invention supports the production of high-quality specifications.

[0019] Note that the description of these effects does not preclude the existence of other effects. One embodiment of the present invention does not necessarily have all of the above effects. Effects other than the above effects can be extracted from the description of the specification, drawings, and claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a diagram showing an example of an information processing system. Figure 2A This is a diagram showing an example of an information processing device. Figure 2B is a diagram showing an example of an information processing system. Figure 3 This is a diagram showing an example of an information processing method. Figure 4 This is a diagram showing an example of an information processing method. Figure 5 This is a diagram showing an example of an information processing method. Figure 6A and Figure 6B This is a diagram showing an example of an information processing method. Figure 7A and Figure 7B This is a diagram showing an example of an information processing method. Figure 8 This is a diagram showing an example of an information processing method. Figure 9A and Figure 9B This is a diagram showing an example of an information processing method. Figures 10A to 10C This is a diagram showing an example of an information processing method. Figure 11 This is a diagram showing an example of an information processing method. Modes for Carrying Out the Invention

[0021] The embodiments are described in detail with reference to the accompanying drawings. Note that the present invention is not limited to the following description. A person skilled in the art will readily appreciate that the embodiments and details can be modified in various ways without departing from the spirit and scope of the present invention. Therefore, the present invention should not be construed as being limited solely to the embodiments described below.

[0022] Note that in the structures of the invention described below, the same reference numerals are used in common across different drawings to represent the same parts or parts having the same function, and their repeated descriptions are omitted. In addition, when parts having the same function are represented, the same hatching is sometimes used without a special reference numeral.

[0023] In addition, for ease of understanding, the positions, sizes, and ranges of various components shown in the drawings may not necessarily represent their actual positions, sizes, and ranges. Therefore, the disclosed invention is not necessarily limited to the positions, sizes, and ranges disclosed in the drawings.

[0024] In this specification, etc., ordinal numbers such as "first" and "second" are used for convenience, but they do not limit the number of components or the order of components (for example, the order of processes). In addition, the ordinal numbers assigned to components in one part of this specification may not be consistent with the ordinal numbers assigned to the same components in other parts of this specification or in the claims.

[0025] In this specification, unless otherwise specified, a document refers to a description of a phenomenon using natural language. A document is digitized and machine-readable. In this specification, a text includes one or more sentences.

[0026] (Implementation Method) In this embodiment, referring to Figures 1 to 11 An information processing device and an information processing method according to one embodiment of the present invention will be described.

[0027] There are no particular limitations on the files that can be processed using the information processing device and information processing method according to one embodiment of the present invention. An information processing device according to one embodiment of the present invention can receive a variety of files. Examples of these files include patent application specifications, utility model registration application specifications, books, magazines, newspapers, contracts, papers (including academic papers, dissertations, doctoral theses, short essays, and journal articles), judgments, clauses, product manuals, novels, publications, white papers, technical documents, and working papers. There are no particular limitations on the language used for these files; files written in various languages, including Japanese, English, Chinese, and Korean, can be received.

[0028] The following mainly describes an information processing device and an information processing method according to one embodiment of the present invention using the specification of the patent application as an example.

[0029] In this specification, etc., an invention refers to a technical concept that utilizes the laws of nature. In this specification, etc., examples of the structure of an invention include individual components (also referred to as technical elements) of the invention and combinations of multiple components. Furthermore, the means for solving a technical problem, or a portion of such means, may also be referred to as the structure of an invention.

[0030] The patent application specification preferably describes not only the structure of the invention but also the effects of the invention related to the structure. Furthermore, the specification preferably describes in detail the function of the invention's structure, the effects resulting from that function, and the principles underlying the invention or its function. This makes it easier for readers to understand the invention, grasp its novelty and progressive features, and facilitate claims of novelty and progressiveness by applicants and their agents.

[0031] Note that "A and B are related" can also be replaced with "A and B are related to each other," "A and B correspond to each other," etc. For example, when the structure and effect of an invention are clearly described (for example, in a manner that is understandable to the reader), the structure and effect of the invention can be said to be related. Specifically, when the structure and effect of an invention are described in the same or adjacent paragraphs, the structure and effect of the invention can also be said to be related.

[0032] An information processing device according to one embodiment of the present invention can display a structure of an invention without describing its corresponding effects. By confirming the displayed content, users can easily identify structures with insufficient descriptions of their effects. This allows for efficient identification of drawings that require revision in the specification, thereby improving the quality of the specification.

[0033] Specifically, an information processing device according to one embodiment of the present invention receives file data from a user and decomposes the text included in the file data into a plurality of sentences. Then, it is determined whether each sentence records at least one of a structure and an effect. Based on the determination, sentences corresponding to the first category that record a structure but not an effect and sentences corresponding to the second category that record an effect but not a structure are extracted from the plurality of sentences. Next, one or more groups are created, each group including at least one sentence corresponding to the first category and at least one sentence corresponding to the second category. Furthermore, sentences corresponding to the first category that do not belong to any group are shown as sentences corresponding to the third category.

[0034] The above groups can be considered to be combinations of one or more corresponding sentences describing a structure and one or more corresponding sentences describing an effect. Therefore, sentences corresponding to the third category can be considered to be sentences that describe a structure and for which no corresponding effect can be found in the text. By identifying sentences corresponding to the third category, users can easily identify inadequately described structures. This can reduce omissions in documents, thereby improving document quality.

[0035] The information processing device according to one embodiment of the present invention may also display a list of created groups, thereby allowing the user to check whether the description is excessive or insufficient according to the structure.

[0036] Furthermore, based on the above determination, one or both of sentences that describe both the structure and the effect and sentences that do not describe both the structure and the effect can be extracted. For example, the information processing device of one embodiment of the present invention can also display sentences that describe both the structure and the effect. In this case, the user can also confirm whether the description is excessive or insufficient based on the structure.

[0037] Furthermore, an information processing device according to one embodiment of the present invention can be used not only to confirm the relationship between a structure and an effect, but also to confirm the relationship between a structure and one or more of various related information about the structure. Examples of related information about a structure include principles, functions, and effects. Furthermore, examples of related information about a structure include structures that are subordinate concepts of the structure.

[0038] Furthermore, an information processing device according to one embodiment of the present invention can generate sentences describing structures using a database containing sentences describing structures and sentences describing related information about the structures. For example, for structures that do not have a corresponding principle, function, or effect described, the device can generate sentences describing what should be described. Furthermore, for structures that already have a principle, function, or effect associated with them, the device can determine whether the description is appropriate. Furthermore, for structures already associated with a principle, function, or effect, the device can generate additional descriptions of other principles, functions, or effects.

[0039] <Information Processing System> Figure 1 The diagram shows an information processing system including the information processing device 10 .

[0040] exist Figure 1 In FIG, the information processing device 10 is connected to the terminal 20 a and the terminal 20 b via the network 30 .

[0041] The user of the information processing system can access the information processing apparatus 10 from the terminal 20a, the terminal 20b, etc. Furthermore, the user can receive services using the information processing method according to one embodiment of the present invention by using communication via the network 30.

[0042] The information processing device 10 is a device capable of executing processing using the information processing method of one embodiment of the present invention. Figure 1 In the embodiment of the present invention, a server computer capable of executing the processing of the information processing method according to one embodiment of the present invention is shown as an example.

[0043] Information processing device 10 can perform information processing, such as calculations, using data input from terminal 20a via network 30. Information processing device 10 can transmit the results of this information processing to terminal 20a via network 30. This reduces the computational burden on terminal 20a. While terminal 20a is used as an example for this description, the same applies to terminal 20b.

[0044] Terminal 20a and terminal 20b are information terminal devices such as personal computers (PCs) used by users, and can also be called client PCs. Figure 1 In the figure, a notebook PC terminal 20a and a desktop PC terminal 20b are shown as examples. The number of terminals connected to the information processing device 10 may be one or three or more. Examples of such terminals include desktop information terminals, notebook information terminals, tablet information terminals, and portable information terminals such as smartphones.

[0045] As the network 30, computer networks such as the Internet, which is the basis of the World Wide Web (WWW), an intranet, an extranet, a PAN (Personal Area Network), a LAN (Local Area Network), a CAN (Campus Area Network), a MAN (Metropolitan Area Network), a WAN (Wide Area Network), and a GAN (Global Area Network) can be used. Furthermore, when wireless communication is performed, communication standards such as the fourth-generation mobile communication system (4G), the fifth-generation mobile communication system (5G), and the sixth-generation mobile communication system (6G), or standards standardized by IEEE, such as Wi-Fi (registered trademark) and Bluetooth (registered trademark), can be used as communication protocols or communication technologies.

[0046] For example, the user can use dedicated software or an application installed on the terminal 20a (or the terminal 20b) to use the information processing device 10. Alternatively, the user can use the information processing device 10 through a web browser using the terminal 20a (or the terminal 20b).

[0047] The user uses terminal 20a (or terminal 20b) to input desired document data 21 into information processing device 10. For example, the user can input data for the specification of a patent application. Information processing device 10 uses the information processing method according to one embodiment of the present invention to judge the sentences in the input document data 21 and outputs the judgment result 11 to terminal 20a (or terminal 20b).

[0048] Figure 1 The example in which the information processing device 10 outputs the following determination result is shown: a sentence describing the structure L and a sentence describing the effect M correspond to each other. Figure 1 An example is shown in which the information processing device 10 outputs a determination result that no sentence describing an effect corresponding to the structure N is found in the document.

[0049] Figure 2A 2 is a block diagram of the information processing device 10. The information processing device 10 includes a receiving unit 110, a storage unit 120, a processing unit 130, an output unit 140, and a transmission channel 150.

[0050] In the drawings of this specification, components are categorized by function and shown as separate blocks. However, in practice, components are difficult to fully separate by function, and a single component may involve multiple functions. For example, a portion of processing unit 130 may also function as receiving unit 110. Furthermore, a single function may involve multiple components. For example, processing performed in processing unit 130 may sometimes be performed on different servers depending on the process.

[0051] [Receiving Unit 110] The receiving unit 110 receives document data from a user. The document data preferably includes text data. Alternatively, the processing unit 130 may create text data using the document data received by the receiving unit 110.

[0052] Furthermore, the receiving unit 110 may also receive a user's evaluation of the determination result of the information processing device 10 .

[0053] The information supplied to the receiving unit 110 is supplied to one or both of the storage unit 120 and the processing unit 130 through the transmission path 150 .

[0054] [Storage unit 120] The storage unit 120 has a function of storing programs executed by the processing unit 130. The storage unit 120 may also have a function of storing data generated by the processing unit 130 (eg, calculation results, analysis results, inference results) and data input to the receiving unit 110.

[0055] like Figure 2A As shown, the storage unit 120 may also include a database 121. Furthermore, the information processing device 10 may also include a database 121 in addition to the storage unit 120 (see the database 121 of the information processing device 10a described later). The information processing device 10 may also have the function of extracting data from a database located outside the storage unit 120, outside the information processing device 10, or outside the information processing system. Furthermore, the information processing device 10 may have the function of extracting data from both the database 121 included in the information processing device 10 and an external database.

[0056] In addition, a file server may be used instead of a database. For example, when using files included in a file server, the database preferably has the path to the files stored in the file server.

[0057] In the database 121, sentences describing structures and sentences describing related information of the structures are stored in groups. For example, sentences describing structures and sentences describing at least one of the related principles, functions, effects, and subordinate concepts of the structures are stored in groups.

[0058] The storage unit 120 includes at least one of a volatile memory and a non-volatile memory. Examples of volatile memory include DRAM (Dynamic Random Access Memory) and SRAM (Static Random Access Memory). Examples of non-volatile memory include ReRAM (Resistive Random Access Memory, also known as resistive random access memory), PRAM (Phase change Random Access Memory), FeRAM (Ferroelectric Random Access Memory), MRAM (Magnetoresistive Random Access Memory, also known as magnetoresistive random access memory), and flash memory. In addition, the storage unit 120 may also include a recording medium drive. Examples of recording medium drives include a hard disk drive (HDD) and a solid state drive (SSD).

[0059] [Processing unit 130] The processing unit 130 has the function of performing processing such as calculation, analysis, and inference using data supplied from one or both of the receiving unit 110 and the storage unit 120. The processing unit 130 may supply the generated data (e.g., calculation results, analysis results, and inference results) to one or both of the storage unit 120 and the output unit 140.

[0060] The processing unit 130 has a function of acquiring data from one or both of the storage unit 120 and the database 121 .

[0061] Processing unit 130 has the function of breaking down the text in the document data received by receiving unit 110 into multiple sentences or clauses. In this embodiment, the example of processing unit 130 breaking down the text into multiple sentences is used as an example. However, sentences can also be referred to as clauses. Furthermore, depending on the situation, the text may be broken down into phrases. In other words, sentences can sometimes be referred to as phrases.

[0062] Processing unit 130 has the function of determining whether a sentence describes at least one of a constituent element and related information about the constituent element (one or more of a principle, function, effect, and the composition of a subordinate concept). For example, processing unit 130 has the function of determining whether a sentence describes at least one of a structure and an effect. Alternatively, processing unit 130 has the function of determining whether a sentence describes at least one of a structure, function, and effect.

[0063] Processing unit 130 has the function of creating a group including at least one sentence corresponding to a first category that describes a structure but does not describe information related to the structure, and at least one sentence corresponding to a second category that describes information related to a structure but does not describe the structure itself. For example, processing unit 130 has the function of creating a group including at least one sentence corresponding to a first category that describes a structure but does not describe an effect, and at least one sentence corresponding to a second category that describes an effect but does not describe a structure.

[0064] The processing unit 130 may also have a function of comparing sentences in the document data received by the receiving unit 110 with sentences stored in the database 121 .

[0065] The processing unit 130 may specify and output sentences that do not belong to a group among the sentences corresponding to the first category. Alternatively, the created sentence group may be output.

[0066] The processing unit 130 may include, for example, a computing circuit, a central processing unit (CPU), or a graphics processing unit (GPU).

[0067] The processing unit 130 may also include a microprocessor such as a DSP (Digital Signal Processor). The microprocessor may also be implemented as a PLD (Programmable Logic Device) such as an FPGA (Field Programmable Gate Array) or an FPAA (Field Programmable Analog Array). Furthermore, the processing unit 130 may also include a quantum processor. The processing unit 130 can perform various information processing and program control by interpreting and executing instructions from various programs through the processor. Programs executable by the processor are stored in at least one of a memory area included in the processor and the storage unit 120.

[0068] The processing unit 130 may include a main memory. The main memory includes at least one of a volatile memory such as a RAM (Random Access Memory) and a non-volatile memory such as a ROM (Read Only Memory).

[0069] RAM, for example, DRAM or SRAM, is used. Virtual memory space is allocated within the RAM as a workspace for processing unit 130 and is used by processing unit 130. The operating system, application programs, program modules, program data, and lookup tables stored in storage unit 120 are loaded into the RAM during execution. Processing unit 130 directly accesses and operates these data, programs, and program modules loaded into the RAM.

[0070] ROM can store non-rewritable BIOS (Basic Input / Output System) and firmware. Examples of ROM include mask ROM, OTPROM (One Time Programmable Read Only Memory), and EPROM (Erasable Programmable Read Only Memory). Examples of EPROM include UV-EPROM (Ultra-Violet Erasable Programmable Read Only Memory), which can erase stored data by ultraviolet light, EEPROM (Electrically Erasable Programmable Read Only Memory), and flash memory.

[0071] It is preferable that artificial intelligence (AI) be used for at least a part of the processing of the information processing device.

[0072] The information processing device preferably uses an artificial neural network (ANN: Artificial Neural Network, hereinafter also referred to as a neural network). A neural network can be implemented by a circuit (hardware) or a program (software).

[0073] In this specification, neural networks refer to any model that mimics biological neural circuit networks and determines the strength of connections between neurons through learning, thereby acquiring problem-solving capabilities. A neural network consists of an input layer, intermediate layers (hidden layers), and an output layer.

[0074] In this specification and other descriptions of a neural network, determining the connection strength (also called weight coefficient) between neurons based on existing information may be referred to as "learning."

[0075] In this specification and other documents, constructing a neural network using the combination strengths obtained through learning and deriving new conclusions from this structure may be referred to as "inference."

[0076] [Output unit 140] The output unit 140 outputs information based on the processing results of the processing unit 130. The output unit 140 can supply at least one of the calculation results, analysis results, and inference results of the processing unit 130 to the outside of the information processing device 10. The output unit 140 can output information to a terminal or display used by the user.

[0077] As a method of presenting information, for example, one or both of the following methods may be performed: displaying the information on a display screen of a terminal used by the user; and outputting the information to a file in a CSV format or the like.

[0078] [Transmission channel 150] The transmission channel 150 has the function of transmitting data. Data transmission and reception between the receiving unit 110 , the storage unit 120 , the processing unit 130 , and the output unit 140 can be performed through the transmission channel 150 .

[0079] Figure 2B 2 is a block diagram of an information processing system 100. The information processing system 100 includes an information processing apparatus 10a and a terminal 20.

[0080] Figure 2B The information processing system 100 shown can be said to be Figure 1 The information processing device 10a is a specific example of an information processing system. Figure 2A The terminal 20 is equivalent to the modified example of the information processing device 10 shown in FIG. Figure 1 Terminal 20a or terminal 20b is shown.

[0081] The information processing device 10a includes a communication unit 171a, a transmission channel 150, a storage unit 120, a database 121, and a processing unit 130. The transmission channel 150, the storage unit 120, the database 121, and the processing unit 130 can refer to the above description.

[0082] The communication unit 171a has the functions of both the reception unit 110 and the output unit 140. In addition to the communication unit 171a, one or both of the reception unit 110 and the output unit 140 may be provided.

[0083] The terminal 20 includes a communication unit 171b, a transmission channel 174, an input unit 115, a storage unit 125, a processing unit 135, and a display unit 145. Examples of the terminal 20 include various PCs, such as tablet computers, notebook computers, and desktop computers, as well as various portable information terminals. Furthermore, the terminal 20 may be a desktop PC that does not include a display unit 145, or the terminal 20 may be connected to a monitor or the like that serves as the display unit 145.

[0084] The user of the information processing system 100 can input document data into the information processing apparatus 10a through the input unit 115 of the terminal 20. The input contents are transmitted from the communication unit 171b to the communication unit 171a.

[0085] The information received by the communication unit 171a is stored in the memory or storage unit 120 included in the processing unit 130 through the transmission channel 150. In addition, the information can also be received from the communication unit 171a through the receiving unit (see Figure 2A The receiving unit 110 shown in FIG. 1 supplies data to the processing unit 130 .

[0086] The processing result of the processing unit 130 is stored in the memory or storage unit 120 included in the processing unit 130 through the transmission channel 150. Then, the processing result is output from the information processing device 10a to the display unit 145 of the terminal 20. The processing result is sent from the communication unit 171a to the communication unit 171b. In addition, various data included in the database 121 can also be sent from the communication unit 171a to the communication unit 171b based on the processing result of the processing unit 130. In addition, the processing result can also be output from the processing unit 130 to the output unit ( Figure 2A The output portion 140 shown is supplied to the communication portion 171a.

[0087] [Communication Unit 171a and Communication Unit 171b] The communication units 171a and 171b can be used to transmit and receive data between the information processing device 10a and the terminal 20. A hub, a router, a modem, or the like can be used as the communication units 171a and 171b. Data transmission and reception can be performed by wired or wireless means (e.g., radio waves, infrared rays, etc.).

[0088] As a communication method between the communication unit 171a and the communication unit 171b, a configuration that can be used in the above-mentioned network 30 can be adopted.

[0089] [Transmission Channel 174] The transmission channel 174 has a function of transmitting data. Data can be sent and received between the communication unit 171 b , the input unit 115 , the storage unit 125 , the processing unit 135 , and the display unit 145 via the transmission channel 174 .

[0090] [Input unit 115] The user can use the input unit 115 when sending document data. Furthermore, the user can use the input unit 115 when sending comments on the determination results of the information processing device 10. For example, the input unit 115 can have the functions of operating the terminal 20, and specifically, examples thereof include a mouse, keyboard, touch panel, microphone, scanner, and camera.

[0091] The information processing system 100 may also have a function of converting audio data into text data. For example, at least one of the processing unit 130 and the processing unit 135 may have this function.

[0092] The information processing system 100 may also include an optical character recognition (OCR) function, thereby recognizing characters included in image data and generating text data. For example, at least one of the processing unit 130 and the processing unit 135 may include this function.

[0093] [Storage unit 125] The storage unit 125 may store one or both of the document data and the data supplied from the information processing device 10 a . Furthermore, the storage unit 125 may include at least a portion of the data that the storage unit 120 may include.

[0094] [Processing unit 135] The processing unit 135 has a function of performing calculations using data supplied from the communication unit 171 b , the storage unit 125 , the input unit 115 , etc. The processing unit 135 may also have a function of executing at least part of the processing that can be performed by the processing unit 130 .

[0095] Each of the processing unit 130 and the processing unit 135 may include one or both of a transistor including a metal oxide in a channel formation region (OS transistor) and a transistor including silicon in a channel formation region (Si transistor).

[0096] In this specification, etc., a transistor using an oxide semiconductor or a metal oxide in a channel formation region is referred to as an oxide semiconductor transistor or an OS transistor. The channel formation region of an OS transistor preferably includes a metal oxide.

[0097] In this specification, metal oxide refers to a metal oxide in a broad sense. Metal oxides are classified into oxide insulators, oxide conductors (including transparent oxide conductors), and oxide semiconductors (also referred to as OS). For example, when a metal oxide is used in a semiconductor layer of a transistor, the metal oxide is sometimes referred to as an oxide semiconductor.

[0098] The metal oxide contained in the channel formation region preferably contains indium (In). When the metal oxide contained in the channel formation region contains indium, the carrier mobility (electron mobility) of the OS transistor is improved. In addition, the metal oxide contained in the channel formation region is preferably an oxide semiconductor containing element M. Element M is preferably at least one of aluminum (Al), gallium (Ga) and tin (Sn). Other elements that can be used as element M include boron (B), silicon (Si), titanium (Ti), iron (Fe), nickel (Ni), germanium (Ge), yttrium (Y), zirconium (Zr), molybdenum (Mo), lanthanum (La), cerium (Ce), neodymium (Nd), hafnium (Hf), tantalum (Ta) and tungsten (W). Note that as element M, multiple of the above elements can sometimes be combined. Element M is, for example, an element with a high bond energy with oxygen. Element M is, for example, an element with a higher bond energy with oxygen than indium. In addition, the metal oxide contained in the channel formation region preferably contains zinc (Zn). Metal oxides containing zinc are sometimes easy to crystallize.

[0099] The metal oxide contained in the channel formation region is not limited to a metal oxide containing indium. The semiconductor layer may be a metal oxide containing zinc, gallium, or tin, such as zinc tin oxide or gallium tin oxide, which does not contain indium.

[0100] Processing unit 130 preferably includes an OS transistor. Because the off-state current of an OS transistor is extremely low, using the OS transistor as a switch to retain charge (data) flowing into a capacitor serving as a storage element ensures a long data retention period. By applying this characteristic to at least one of the registers and cache memory included in processing unit 130, processing unit 130 can be operated only when necessary. Otherwise, previously processed information is stored in the storage element, allowing processing unit 130 to be shut down. This enables normally off computing, reducing power consumption in the information processing system.

[0101] [Display unit 145] The display unit 145 has a function of displaying the output result. As the display unit 145, a liquid crystal display device, a light-emitting display device, etc. can be cited. As light-emitting elements that can be used for the light-emitting display device, LED (Light Emitting Diode), OLED (Organic LED), QLED (Quantum-dot LED), and semiconductor lasers can be cited. In addition, the following display devices can be used in the display unit 145: a display device using MEMS (Micro Electro Mechanical Systems) elements using a shutter method or an optical interference method; a display device using display elements using a microcapsule method, an electrophoresis method, an electrowetting method, or an electronic powder fluid (registered trademark) method; etc.

[0102] Reference Figures 3 to 11 The information processing method and output method in an information processing device according to one embodiment of the present invention will be described. Note that a display method is provided below as an example of an output method. Specifically, the method for displaying the results of the information processing method according to one embodiment of the present invention will be described below.

[0103] <Information Processing Method 1> The information processing method 1 of this embodiment includes Figure 3 The processing of steps S1 to S5 is shown. Figure 4 This is a diagram for explaining information processing method 1. Figure 4 It can also be said that this is an example of a graphical user interface (GUI) of the information processing system according to this embodiment. Figure 4 The formats, icons, and tables shown in the drawings of the GUI according to this embodiment are merely examples and are not particularly limited. The GUI may be configured as a web page that a user accesses via a network. Alternatively, the GUI may be configured as a screen of a program application executed on a terminal utilized by the user.

[0104] [Step S1] In step S1 , the receiving unit 110 receives file data from a user via the terminal 20 a or the terminal 20 b .

[0105] Figure 4 The area 51 shown is where the user can select a file. For example, when the user clicks or touches the area 51, a dialog box for selecting a file is displayed, and the user can select the desired file data. Alternatively, the user can also receive file data in the area 51 by performing operations such as dragging and dropping.

[0106] The document data preferably includes text data. The document data may also include data other than text data (e.g., image data). Note that the processing after step S2 is primarily performed using text data. Therefore, after step S1, the processing unit 130 may also perform processing to convert the document data into text data or extract text data from the document data, as needed.

[0107] [Step S2] In step S2 , the processing unit 130 decomposes the text in the document data received in step S1 into a plurality of sentences.

[0108] Note that, in this embodiment, an example is shown in which a text is decomposed into sentences, but the present invention is not limited to this, and a text may be decomposed into a plurality of clauses.

[0109] For example, the text can be divided into sentences, clauses or phrases according to the demarcation symbols (also called demarcation marks) in the text. For example, the text can be divided into sentences, clauses or phrases according to one or more of a period, a comma, a bracket, an exclamation mark, a question mark, a blank, a line break, a colon, a semicolon, etc.

[0110] Alternatively, the file data received in step S1 can be analyzed for its structure (also known as a hierarchical structure) to extract the file's title, paragraphs, and the like. Displaying the judgment results along with the title, paragraph, and the like is preferable to displaying the judgment results for each sentence alone, as it is easier for the user to see and understand the results. Furthermore, by combining the judgment results with the title information and performing subsequent processing, it is possible to prevent sentences included in different titles from being associated with each other. Furthermore, by combining the judgment results with the paragraph information and performing subsequent processing, it is possible to preferentially associate sentences included in the same paragraph, and the like. This improves judgment accuracy.

[0111] [Step S3] In step S3, the processing unit 130 determines whether each of the plurality of sentences obtained in step S2 describes at least one of a structure and an effect. For example, the plurality of sentences are classified into the following four categories: (1) sentences that describe a structure but not an effect (hereinafter simply referred to as sentences that describe a structure); (2) sentences that describe an effect but not a structure (hereinafter simply referred to as sentences that describe an effect); (3) sentences that describe both a structure and an effect; and (4) sentences that describe neither a structure nor an effect.

[0112] Note that in this embodiment, an example of determining whether an effect is recorded is shown, but it is not limited to this. It can be determined whether one or more information in the relevant structure (for example, one or more of the functions, effects, principles, and structures of lower-level concepts) is recorded.

[0113] There is no particular limitation on the sentence classification method. For example, one or both of a rule-based method and a method utilizing machine learning may be used.

[0114] As a rule-based method, for example, a method using regular expressions is preferred. A regular expression is an example of a method that uses a string to represent a set of strings. By using a regular expression, a specified string can be easily and accurately retrieved or extracted. When a sentence containing a specified string belongs to one of the four categories described above, it is preferred to use a regular expression to search for that string to classify the sentence.

[0115] As a method using machine learning, for example, a method using natural language processing and a recognizer (also called a classifier) ​​is preferred. For example, a sentence is vectorized and the vectorized sentence is input to a classifier. The classifier can output a determination of which of the four categories the input sentence belongs to.

[0116] Methods for vectorizing sentences include Bag-of-Words (Bag of Words model), TF-IDF (Term Frequency-Inverse Document Frequency), Doc2Vec, fastText, and BERT (Bidirectional Encoder Representations from Transformer).

[0117] The classifier can use naive Bayes, support vector machine, random forest, neural network (including multilayer perceptron), etc. In particular, it is preferable to use a natural language processing model using Transformer (one of the architectures of neural networks), and more specifically, a general natural language processing model such as BERT is preferably used.

[0118] In particular, for vectorization, it is preferable to use a language model that can represent the same word differently depending on the distribution of surrounding words or the context. Furthermore, using a model that integrates sentence vectorization and classification can improve classification accuracy. For example, it is preferable to use the aforementioned BERT for both sentence vectorization and classification.

[0119] When machine learning is used, humans do not need to cite all the classification rules, and can classify sentences based on regularities that humans have not yet mastered. Therefore, sentences can be classified with high accuracy, which is preferable.

[0120] Alternatively, a database of sentences including classification information may be prepared and each sentence may be classified by comparing it with the database. For example, the above structure may be compared with the database 121 that stores information related to the structure in groups.

[0121] [Step S4] In step S4 , the processing unit 130 creates a pair of a sentence describing a structure and a sentence describing an effect.

[0122] The sentences with structures and the sentences with effects that are recorded in a group can be included in the same paragraph or in different paragraphs. In addition, the paragraph including the sentences with structures and the paragraph including the sentences with effects that are recorded can be adjacent to or far away from each other. The sentences with structures and the sentences with effects that are recorded in a group can also be set to the following conditions: the number of paragraphs between the paragraph including the sentences with structures and the paragraph including the sentences with effects that are recorded, or the number of sentences, words or characters between the two sentences are within the specified range. In addition, the sentences with structures and the sentences with effects that are recorded in a group are preferably included in the same title. Thus, the association error between the sentences with structures and the sentences with effects that are recorded can be prevented.

[0123] There is no particular limitation on the method for creating a group, and for example, one or both of a rule-based method and a method utilizing machine learning may be used.

[0124] When creating groups based on rules, the following conditions can be given as examples: (a1) adjacent sentences with a structure and a sentence with an effect are grouped together; (a2) a sentence with a structure and the immediately following sentence with an effect are grouped together; (a3) ​​sentences with a structure and a sentence with an effect that are included in the same paragraph are grouped together; (b1) multiple sentences with a structure are grouped together when they are consecutive and included in the same paragraph; (b2) two sentences with a structure are grouped together when the number of paragraphs, sentences, words, or characters between them falls within a specified range; (c1) multiple sentences with an effect are grouped together when they are consecutive and included in the same paragraph; (c2) two sentences with an effect are grouped together when the number of paragraphs, sentences, words, or characters between them falls within a specified range; etc. Multiple of these conditions can also be used in combination.

[0125] As a method using machine learning, for example, a method using natural language processing and a recognizer (also called a classifier) ​​is preferred. For example, two sentences are vectorized and the two vectorized sentences are input to a classifier. The classifier can output a determination result of the relationship between the two input sentences.

[0126] The sentence vectorization method and classifier examples are described above.

[0127] In particular, for vectorization, it is preferable to use a language model that can represent the same word differently depending on the distribution of surrounding words or the context. Furthermore, using a model that integrates sentence vectorization and classifiers can improve judgment accuracy. For example, it is preferable to use the aforementioned BERT for both sentence vectorization and for determining the relationship between two sentences.

[0128] When machine learning is used, humans do not need to cite all the judgment rules, and can classify the relationship between two sentences, including regularities that humans do not understand. Therefore, it is possible to create sentences with high accuracy, which is preferable.

[0129] For example, it is preferred to output the following judgment results for two sentences and create the above-mentioned groups based on these judgment results: (x) both sentences are sentences that record structures and are highly related to each other; (y) both sentences are sentences that record structures and sentences that record the effects of the structures; (z) both sentences are sentences that have a low relationship with each other; etc.

[0130] [Step S5] In step S5, sentences that do not belong to the group among the sentences having the structure are displayed. The determination result of the processing unit 130 is outputted to the terminal or display used by the user via the output unit 140.

[0131] It can be seen that in step S3, (1) sentences that describe the structure but not the effects, which do not belong to any of the groups created in step S4, describe the structure of the invention but not the corresponding effects. Therefore, by indicating the existence of such sentences to the user, the user can easily understand the structure of the invention whose effects are not adequately described. In addition, by adding the effects, the structure of the invention and the effects can be accurately linked, thereby improving the quality of the specification.

[0132] Figure 4 The area 50 shown shows an example in which the determination result 54 of the processing unit 130 is output as each sentence 55. Specifically, Figure 4 The illustrated area 50 shows a table including a title 52, a paragraph 53, a determination result 54, and a sentence 55. Note that the contents of the table are not limited to this.

[0133] By analyzing the file structure of the document data received in step S1, it is possible to extract the title 52 and the paragraph 53. Although the title 52 and the paragraph 53 may not be displayed, by displaying them, the user's convenience in browsing the results can be improved.

[0134] For example, area 56 shows icons of each title ("Mode 1", "Mode 2", "Mode 3"). Area 56 has a function of moving the display to the position of a desired title by the user selecting the desired title.

[0135] The determination result 54 shows into which of the above-mentioned (1) to (4) the sentence 55 is classified and whether it is a sentence included in a group.

[0136] Figure 4 Specific examples of sentences 73a to 73f are shown. Sentence 73a" Figure 1 2 is a cross-sectional view showing a transistor” and sentence 73f “FIG. 2 is a three-dimensional view showing a transistor” were judged to be sentences 74 that did not describe both the structure and the effect. Sentence 73b “Layer A preferably uses element B”, sentence 73c “Alternatively, layer A may also use element C” and sentence 73e “Layer D preferably uses element E” were judged to be sentences 71 that described the structure but did not describe the effect. Sentence 73d “Thus, the barrier property of layer A is improved, thereby improving the reliability of the device” was judged to be a sentence 72 that described the effect but did not describe the structure. Note that although in Figure 4 Although not shown in the figure, an example of a sentence that can be judged to record both the structure and the effect is "When element B or element C is used for layer A, the barrier property of layer A is improved, thereby improving the reliability of the device, so it is preferred."

[0137] Sentence 73b, sentence 73c, and sentence 73d are determined to be one group 75. By showing group 75, it can be confirmed that the structure and the effect are described in a correlated manner. Figure 4 An example is shown in which the group 75 corresponds to a part of a plurality of sentences included in the paragraph

[0033] .

[0138] On the other hand, sentence 73e is shown as a sentence 76 that has a structure and does not belong to a group.

[0139] The user preferably sequentially checks at least the sentences 76 included in the document. This allows efficient confirmation of whether or not the description of the structure is insufficient.

[0140] For example, icons ("Back" and "Forward") are displayed in area 57. By selecting an icon in area 57, the user can move the display to the position of the previous or next sentence 76 of the currently displayed or selected portion. This is preferred because it allows efficient confirmation of all sentences 76. Note that this function is not limited to sentence 76; the same function can also be applied to sentence 71, sentence 72, group 75, and so on.

[0141] Furthermore, the total number of sentences 76 that have a structure and do not belong to a group may be displayed. This allows the user to clearly understand at a glance the number of sentences 76 that should be confirmed or whether there are any sentences 76 that should be confirmed. Similarly, one or more of the total number of groups 75, the total number of sentences 71, and the total number of sentences 72 may be displayed.

[0142] As described above, using an information processing device according to one embodiment of the present invention, it is possible to confirm the relationship between structure and effect in a specification. Furthermore, it is possible to easily discover structures that are not related to effects. Thus, an information processing device according to one embodiment of the present invention can support the creation of specifications so that the relationship between structure and effect is fully established.

[0143] <Information Processing Method 2> The information processing method 2 of this embodiment includes Figure 5 The processing of steps S11 to S15 shown in FIG. 2 is preferably performed by the information processing method 2. Figure 6A The processing of step S16 and step S17 shown and Figure 6B One or both of the processes of step S26 and step S27 shown. Figure 7A and Figure 7B This is a diagram for explaining information processing method 2. Figure 7A and Figure 7B This can also be said to be an example of the GUI of the information processing system according to this embodiment.

[0144] Note that, in each information processing method described below, description of the same parts as in information processing method 1 may be omitted.

[0145] [Step S11 and Step S12] Step S11 is the same process as step S1 , and step S12 is the same process as step S2 , so the above description can be referred to.

[0146] [Step S13] In step S13 , the processing unit 130 uses the first classifier to determine whether each of the plurality of sentences obtained in step S12 describes at least one of a structure and an effect.

[0147] The first classifier preferably uses the method using machine learning described in Information Processing Method 1, and more preferably uses BERT.

[0148] [Step S14] In step S14 , the processing unit 130 creates a group of sentences describing structures and sentences describing effects using the second classifier.

[0149] The second classifier preferably uses the method using machine learning described in Information Processing Method 1, and more preferably uses BERT.

[0150] [Step S15] Step S15 is the same process as step S5, and therefore the above description can be referred to.

[0151] The information processing device of one embodiment of the present invention preferably receives user evaluations of determination results at the receiving unit 110. For example, when a classifier is used for determination, the user's evaluations are preferably used to train (or retrain) the classifier. This can improve the accuracy of the classifier. Furthermore, when determination is performed using methods other than classifiers, the accuracy of the determination can be improved by changing the classification conditions based on the evaluations.

[0152] [Step S16] In step S16 , the receiving unit 110 receives an evaluation of whether the judgment of the sentence 71 in which the structure is described and the sentence 72 in which the effect is described is correct.

[0153] Figure 7A The area 58 shown is where the user can enter their evaluation of the judgment. Area 58 is preferably configured with, for example, radio buttons, check boxes, text boxes, etc. The user can, for example, indicate that the judgment was incorrect. Alternatively, the user can enter the correct judgment content (any of (1) to (4) above). After the user enters their evaluation of at least one judgment, they select the send button 59, whereby the receiving unit 110 can receive the user's evaluation.

[0154] Figure 7A Specific examples of sentences 73g and 73h are shown. Sentence 73g, "Layer S preferably uses element T," is judged as sentence 71, which describes a structure but not an effect. Sentence 73h, "Layer S can be 1 nm or more and 10 nm or less," is judged as sentence 72, which describes an effect but not a structure.

[0155] Sentences 73g and 73h are determined to be one group 75. However, sentence 73h is not a sentence describing an effect, so it can be said that the determination is wrong.

[0156] Figure 7A The example in which the evaluation 60 is input for sentence 73h, which is determined to have an effect, is shown. On the other hand, no evaluation is input for sentence 73g, which is determined to have a structure, 71. This is preferable because it simplifies the user's work by not requiring the user to input an evaluation for sentences that have been correctly determined.

[0157] in addition, Figure 7A 73g and 73h included in group 75 are included in an example of adjacent paragraphs. Figure 7B As shown, the two sentences included in group 75 can also be included in separate paragraphs. Figure 7B In the example, sentence 73i is determined to be a sentence 71 having a structure, and sentence 73j is determined to be a sentence 72 having an effect. Furthermore, these two sentences are determined to be a group 75. There is another paragraph between the paragraph including sentence 73i and the paragraph including sentence 73j, and they are separated from each other.

[0158] [Step S17] In step S17, the processing unit 130 uses each sentence and the labels based on the received evaluations as learning data to train the first classifier. This improves the accuracy of the first classifier in determining the sentences 71 with structures and the sentences 72 with effects.

[0159] [Step S26] In step S26 , the receiving unit 110 receives an evaluation of whether the determination of the group 75 is correct.

[0160] [Step S27] In step S27, the processing unit 130 uses each group and the labels based on the received evaluations as learning data to perform training on the second classifier. This improves the accuracy of creating the groups 75 in the second classifier.

[0161] exist Figure 7A The area 58 shown may also use the evaluation of the judgment of the sentence 71 or the sentence 72 as the evaluation of the group 75 including the evaluated sentence 71 or the sentence 72. Alternatively, the evaluation of the group 75 may be received separately.

[0162] <Information Processing Method 3> The information processing method 3 of this embodiment includes Figure 8 Alternatively, the information processing method 3 preferably includes the processing of steps S31 to S34 and Figure 9A The processing of step S45 shown in FIG. Figure 10A and Figure 10B A diagram illustrating information processing method 3 is shown.

[0163] [Step S31 to Step S33] Step S31 is the same process as step S1 , step S32 is the same process as step S2 , and step S33 is the same process as step S3 or step S13 , and therefore, the above descriptions can be referred to.

[0164] [Step S34] In step S34 , the processing unit 130 compares the sentence 71 having the structure described with the database 121 , thereby creating a pair 75 of the sentence 71 having the structure described and the sentence 72 having the effect described.

[0165] Figure 10A An example is shown in which group 75a is created based on group 85a included in database 121. Figure 10B A diagram showing comparison between the explanatory sentence 73e and the database 121 is shown.

[0166] Database 121 stores sentences with structures and sentences with effects in groups. For example, group 85a includes sentence 81a with a structure and sentence 82a with an effect. Similarly, group 85b includes sentence 81a with a structure and sentence 82b with an effect, and group 85c includes sentence 81c with a structure and sentence 82c with an effect.

[0167] First, in step S33, sentence 73b is determined to be the sentence 71 having the structure described therein. Then, in step S34, sentence 73b is compared with database 121 to extract sentence 81a as a sentence that is identical or similar to sentence 73b.

[0168] Similarly, in step S33, sentence 73e is determined to be the sentence having the structure 71. Then, in step S34, sentence 73e is compared with the database 121, and sentence 81c is extracted as a sentence that is consistent with or similar to sentence 73e.

[0169] In order to specify a sentence that is consistent or similar to a certain sentence, it is preferable to calculate the similarity between the two sentences. For example, the similarity between the two sentences can be calculated based on the literal consistency of the two sentences (or the consistency of the character strings). Alternatively, each sentence can be vectorized (digitized) to calculate the similarity or distance between the vectors of the two sentences. The method of vectorizing sentences is as described above.

[0170] Examples of methods for determining the similarity between two vectors include cosine similarity, covariance, unbiased covariance, and Pearson correlation coefficient, among which cosine similarity is particularly preferably used.

[0171] Examples of methods for determining the distance between two vectors include Euclidean distance, standard (normalized, average) Euclidean distance, Mahalanobis distance, Manhattan distance, Chebyshev distance, and Minkowski distance.

[0172] Note that there is no particular limitation on the method of calculating the similarity or distance between two sentences, and various models capable of calculating the similarity or distance can be used.

[0173] Group 85a includes sentence 82a having an effect corresponding to sentence 81a. Group 85b includes sentence 82b having an effect corresponding to sentence 81a. Group 85c includes sentence 82c having an effect corresponding to sentence 81c.

[0174] For example, by comparing each of the sentences 72 determined to have an effect in step S33 with sentence 82a, sentence 82b, or sentence 82c, sentences that are identical or similar to sentence 82a, sentence 82b, or sentence 82c can be extracted. Alternatively, each of the multiple sentences obtained in step S32 can be compared with sentence 82a, sentence 82b, or sentence 82c.

[0175] Figure 10A The example in which sentence 73d is determined to be identical or similar to sentence 82a is shown. By the above method, group 75a can be created.

[0176] Figure 10B This shows an example in which no sentence identical or similar to sentence 82c is found in the document.

[0177] Note that when extracting a plurality of sentences as sentences that are identical or similar to sentence 82 a , it is preferable to extract, for example, a sentence that is closest to sentence 73 b in the text among the plurality of sentences.

[0178] [Step S35] In step S35 , sentences that do not belong to a group among the sentences 71 in which the structure is described are displayed.

[0179] Figure 10B The sentence 73e in FIG is the sentence 76a for which no group is created in step S34. Therefore, in step S35, the sentence 73e is displayed.

[0180] [Step S45] In step S45 , sentences having effects recorded thereon corresponding to sentences not belonging to a group among the sentences 71 having structures recorded thereon are extracted and displayed from the database 121 .

[0181] When no sentence identical or similar to sentence 82c is found in the file, sentence 82c may be output as a sentence having an effect corresponding to sentence 81c recorded therein.

[0182] The user can confirm the content of sentence 82c and determine whether to add sentence 82c to the file as an effect corresponding to the structure described in sentence 73e.

[0183] Furthermore, when group 85c includes sentences describing functions, principles, or structures of subordinate concepts, these sentences are preferably displayed in addition to sentence 82c. Thus, the more abundant the amount and variety of data in database 121, the more the information processing apparatus of this embodiment can support the creation of manuals and improve their quality.

[0184] Note that sentences having other effects recorded thereon corresponding to the sentences included in the group in the sentence 71 having the structure recorded thereon (eg, the sentence 73 b ) may be extracted from the database 121 and displayed.

[0185] For example, when no sentence identical or similar to sentence 82b is found in the document, sentence 82b may be output as a sentence describing an effect corresponding to sentence 81a.

[0186] The user can confirm the content of sentence 82b and determine whether to add sentence 82b to the file as an effect corresponding to the structure described in sentence 73b.

[0187] In addition, when the database includes functions, principles, structures of subordinate concepts, etc. in addition to the effects corresponding to the structures, they can also be displayed together.

[0188] In this way, an information processing device according to one embodiment of the present invention can provide information about the structure described in a document (e.g., one or more of the following: function, effect, principle, and subordinate concept). This allows users to improve the quality of the documents they create with less burden.

[0189] <Information Processing Method 4> The information processing method 4 of this embodiment includes Figure 9B The processing of step S51 and step S52 shown in FIG. Information processing method 4 is preferably performed in combination with information processing method 1 or information processing method 2. In addition, Figure 10C A diagram illustrating the information processing method 4 is shown.

[0190] [Step S51] After creating the groups in step S4 or step S14, step S51 is performed. In step S51, the processing unit 130 compares each group with the database 121 to determine whether there are any errors or deficiencies in the association between the structure and the effect.

[0191] First, in step S4 or step S14, group 75b is created. Here, when group 75b is created, no comparison with database 121 is performed. Figure 10C As shown, group 75b includes sentence 73g in which the structure is described and sentence 73h in which the effect is described.

[0192] Then, in step S51, by comparing sentence 73g with database 121, sentence 81d is extracted as a sentence that is consistent with or similar to sentence 73g.

[0193] Group 85d includes sentence 82d in addition to sentence 81d, which describes an effect corresponding to sentence 81d. Next, by comparing sentence 73h with sentence 82d, it is determined whether sentence 73h is consistent with or similar to sentence 82d.

[0194] If it is determined that sentence 73h is identical or similar to sentence 82d, it can be determined that sentence 73g and sentence 73h are correctly combined. On the other hand, if it is determined that sentence 73h is not identical or similar to sentence 82d, it is possible that sentence 73g and sentence 73h are not correctly combined.

[0195] [Step S52] In step S52, as a result of the determination, group 75b determined to be a possible incorrect combination in step S51 is shown. In addition, sentence 82d is shown as a sentence describing an effect corresponding to sentence 73g.

[0196] The user can check the output judgment result to confirm whether the sentence 73g and the sentence 73h are a correct combination. In addition, it is also possible that the sentence 73g and the sentence 73h are a new combination of structure and effect that is not stored in the database 121.

[0197] In addition, the user can determine whether to add sentence 82d to the file as an effect corresponding to the structure described in sentence 73g.

[0198] <Information Processing Method 5> The information processing method 5 of this embodiment includes Figure 11 The processing of steps S61 to S67 is shown.

[0199] In information processing method 5, an example of determining by paragraph rather than by sentence is shown. Depending on the document, by determining by paragraph, the determination result can sometimes be simplified and easily confirmed. In addition, when a structure and related information of the structure are recorded in the same paragraph or adjacent paragraphs, it is often easier to understand the content than when they are recorded in distant paragraphs. Therefore, by determining by paragraph, it is possible to evaluate whether the structure and related information of the structure are associated in an easily understandable manner. On the other hand, depending on the document, the determination accuracy is sometimes higher when determining by sentence. In addition, both determination by sentence and determination by paragraph can be performed.

[0200] [Step S61] Step S61 is the same process as step S1, and therefore the above description can be referred to.

[0201] [Step S62] In step S62, the processing unit 130 analyzes the structure of the document, for example, extracting titles and paragraphs.

[0202] [Step S63] Step S63 is the same processing as step S2, so the above description can be referred to.

[0203] [Step S64] Step S64 is the same processing as step S3 or step S13, and therefore the above description can be referred to.

[0204] [Step S65] In step S65, the processing unit 130 determines whether each of the plurality of paragraphs describes at least one of a structure and an effect based on the determination result obtained in step S64. For example, the plurality of paragraphs are classified into the following four categories: (5) paragraphs that describe a structure but not an effect (hereinafter simply referred to as paragraphs that describe a structure); (6) paragraphs that describe an effect but not a structure (hereinafter simply referred to as paragraphs that describe an effect); (7) paragraphs that describe both a structure and an effect; and (8) paragraphs that describe neither a structure nor an effect.

[0205] [Step S66] In step S66 , the processing unit 130 creates a group of a paragraph describing a structure and a paragraph describing an effect.

[0206] [Step S67] In step S67, paragraphs that do not belong to the group among the paragraphs in which the structure is described are displayed. The determination result of the processing unit 130 is outputted to the terminal or display used by the user via the output unit 140.

[0207] As described above, using an information processing device according to one embodiment of the present invention, it is possible to confirm the relationship between a structure in a document and information related to that structure. Furthermore, it is possible to easily discover structures that are not related to that information. Thus, an information processing device according to one embodiment of the present invention can support the creation of high-quality documents in which the relationship between structure and information related to that structure is sufficiently clear.

[0208] This embodiment mode can be combined with other embodiment modes as appropriate. In addition, in this specification, when a plurality of configuration examples are shown in one embodiment mode, the configuration examples can be combined as appropriate. [Explanation of symbols]

[0209] 10a: Information processing device, 10: Information processing device, 11: Judgment result, 20a: Terminal, 20b: Terminal, 20: Terminal, 21: File data, 30: Network, 50: Area, 51: Area, 52: Title, 53: Paragraph, 54: Judgment result, 55: Sentence, 56: Area, 57: Area, 58: Area, 59: Send button, 60: Evaluation, 71: Sentence, 72: Sentence, 73a: Sentence, 73b: Sentence, 73c: Sentence, 73d: Sentence, 73e: Sentence, 73f: Sentence, 73g: Sentence, 73h: Sentence, 73i: Sentence, 73j: Sentence, 74: Sentence, 75a: group, 75b: group, 75: group, 76a: sentence, 76: sentence, 81a: sentence, 81c: sentence, 81d: sentence, 82a: sentence, 82b: sentence, 82c: sentence, 82d: sentence, 85a: group, 85b: group, 85c: group, 85d: group, 100: information processing system, 110: receiving unit, 115: input unit, 120: storage unit, 121: database, 125: storage unit, 130: processing unit, 135: processing unit, 140: output unit, 145: display unit, 150: transmission channel, 171a: communication unit, 171b: communication unit, 174: transmission channel.

Claims

1. An information processing method comprising the following steps: Receive file data; Decomposing the text included in the document data into a plurality of sentences; extracting sentences corresponding to a first category in which a structure is described but an effect is not described, and sentences corresponding to a second category in which an effect is described but a structure is not described, from the plurality of sentences; creating one or more groups, the groups including at least one sentence corresponding to the first category and at least one sentence corresponding to the second category; and Among the sentences corresponding to the first category, sentences that do not belong to any of the groups are indicated as sentences corresponding to the third category.

2. The information processing method according to claim 1, The sentences corresponding to the first category are compared with the database to obtain the sentences corresponding to the fourth category that record the corresponding effects, and the groups are created by comparing the sentences corresponding to the second category with the sentences corresponding to the fourth category.

3. The information processing method according to claim 1, Here, sentences corresponding to the fourth category in which corresponding effects are described are obtained by comparing each sentence corresponding to the third category with a database, and the sentences corresponding to the fourth category are displayed.

4. The information processing method according to claim 1, The result of comparing the combination of the sentence corresponding to the first category and the sentence corresponding to the second category in the group with the database is shown.

5. The information processing method according to claim 1, The first classifier is used when extracting sentences corresponding to the first classification and sentences corresponding to the second classification, Show sentences corresponding to the first category and sentences corresponding to the second category, receiving an evaluation of at least one sentence corresponding to the first category or at least one sentence corresponding to the second category, Furthermore, a combination of the evaluated sentence and the evaluation is used as learning data to perform training on the first classifier.

6. The information processing method according to claim 1, wherein a second classifier is used in creating said groups, Showing the group, receiving an evaluation of at least one of the groups, Furthermore, a combination of the evaluated group and the evaluation is used as learning data to perform learning on the second classifier.

7. A program for causing a processor to execute the information processing method according to claims 1 to 6.

8. An information processing device comprising: Receiving Department; Processing Department; as well as Output section, The receiving unit has the function of receiving file data. The processing unit has: A function of decomposing the text included in the document data into a plurality of sentences; A function to extract sentences corresponding to the first category that record a structure but do not record an effect; A function to extract sentences corresponding to the second category that record effects but do not record structures; a function of creating a group including at least one sentence corresponding to the first category and at least one sentence corresponding to the second category; and A function of extracting sentences that do not belong to any of the groups from among the sentences corresponding to the first category as sentences corresponding to the third category, Furthermore, the output unit has a function of outputting sentences corresponding to the third category.

9. The information processing apparatus according to claim 8, further comprising: database, The database contains corresponding combinations of sentences with structures and sentences with effects.

10. The information processing device according to claim 9, The processing unit comprises: A function of obtaining sentences corresponding to the fourth category having corresponding effects recorded therein by comparing each sentence corresponding to the first category with the database; and The function of creating the group is to collate the sentences corresponding to the second category with the sentences corresponding to the fourth category.

11. The information processing device according to claim 9, The processing unit compares each sentence corresponding to the third category with the database to obtain sentences corresponding to the fourth category in which the corresponding effect is described, and displays the sentences corresponding to the fourth category.

12. The information processing device according to claim 9, The processing unit displays a result of comparing a combination of a sentence corresponding to the first category and a sentence corresponding to the second category in the group with the database.

Images