Reading support system

Abstract

To provide a reading comprehension support system or a reading comprehension support method which allows a natural language to be input as a query sentence and presents portions strongly related to the input sentence to a reader.SOLUTION: A reading comprehension support system comprises: a document reading part for reading a target document; a document division part for dividing the target document into a plurality of blocks; a first distributed expression acquisition part for acquiring word distributed expression for each of the plurality of blocks; a query sentence reading part for reading a query sentence; a second distributed expression acquisition part for extracting a word included in the query sentence to acquire a word distributed expression; and a similarity acquisition part for comparing word distributed expressions in the query sentence and each of the plurality of blocks to acquire similarity. The similarity acquisition part retrieves from words included in the block a word coincident with the word included in the query sentence, and acquires, for the word coincident with that word, similarity between the word distributed expression in the block and the word distributed expression in the query sentence.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] One aspect of the present invention relates to a document reading support system and a reading support method. [Background technology]

[0002] When reading and comprehending a document, the way in which it is read varies depending on the reader's purpose and the type and nature of the document. In some cases, the reader reads the entire document, while in other cases, the reader's goal is to find the information they need, and it is sufficient to search for the part of the document that contains the information they need and skim only that part. Methods for finding the information they need within a document include using a table of contents or index. In the case of a digitalized document, another method is to search for keywords to find the desired information. A method has also been proposed for analyzing the structure of a document according to set rules (Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2014-219833 [Non-patent literature]

[0004] [Non-Patent Document 1] BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding, Devlin et al. (Submitted on 11 Oct 2018 (v1), last revised 24 May 2019 (this version, v2)), [online], Internet<URL:https: / / arxiv.org / abs / 1810.04805v2> Summary of the Invention [Problem to be solved by the invention]

[0005] When using a table of contents or index, efficiency is low if the word you are directly searching for is not used in the table of contents or index. While a text search using a keyword can search for sentences or paragraphs containing the keyword throughout the entire document, it may not be possible to find the desired information efficiently. Reasons for this include: the keyword yielding too many hits, taking too long to find the desired information; being unable to narrow down the desired information using a single keyword; or being unable to find an appropriate keyword. Furthermore, when analyzing the structure of a document according to rules, the structures to be read are limited, making it difficult to handle documents with a variety of structures. One aspect of the present invention solves at least one of these problems.

[0006] One aspect of the present invention aims to provide a reading support system or a reading support method that enables input of natural language as a query sentence and presents to the reader parts that are highly relevant to the input sentence.

[0007] Note that the description of these problems does not preclude the existence of other problems. One embodiment of the present invention does not necessarily have to solve all of these problems. Problems other than these can be extracted from the description in the specification, drawings, and claims. [Means for solving the problem]

[0008] One aspect of the present invention is a reading comprehension support system including a document reading unit that reads a target document, a document segmentation unit that divides the target document into a plurality of blocks, a first embedded representation acquisition unit that acquires embedded representations of words for each of the plurality of blocks, a query sentence reading unit that reads a query sentence, a second embedded representation acquisition unit that extracts words included in the query sentence and acquires embedded representations of the words, and a similarity acquisition unit that compares the embedded representations of words in the query sentence with each of the plurality of blocks to calculate a similarity. The similarity acquisition unit searches for words that match words included in the query sentence from among the words included in the blocks, and for the matching words, calculates a similarity between the embedded representations of words in the block and the embedded representations of words in the query sentence.

[0009] One aspect of the present invention is a reading comprehension support method including the steps of: reading a target document, dividing the target document into a plurality of blocks, acquiring word embeddings for each of the plurality of blocks, reading a query sentence, extracting words included in the query sentence and acquiring word embeddings, and comparing the word embeddings between the query sentence and each of the plurality of blocks to calculate a similarity. In the similarity calculation step, words included in the blocks that match words included in the query sentence are searched for, and for the matching words, a similarity between the word embeddings in the block and the word embeddings in the query sentence is calculated.

[0010] Each of the multiple blocks may contain one or more paragraphs of the target document.

[0011] Each of the multiple blocks can contain one or more statements.

[0012] The similarity may be acquired only for a predetermined part of speech.

[0013] The similarity may be obtained by calculating the cosine similarity.

[0014] If there are multiple words that match between the query sentence and the block, the sum of the similarities of the distributed representations for each word may be used as the score for the block. [Effects of the Invention]

[0015] According to one aspect of the present invention, it is possible to provide a reading support system or a reading support method that enables input of a natural language as a query sentence and presents to the reader parts that are highly relevant to the input sentence.

[0016] 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 these effects. Effects other than these can be extracted from the description in the specification, drawings, and claims. [Brief explanation of the drawings]

[0017] [Figure 1] FIG. 1 is a diagram illustrating an example of a reading support system. [Figure 2] FIG. 2 is a flowchart showing an example of a reading support method. [Figure 3] FIG. 3 is a flowchart showing an example of a reading support method. [Figure 4] FIG. 4 is a diagram illustrating embedded representations of words. [Figure 5] FIG. 5 is a diagram illustrating an example of a method for calculating the similarity. [Figure 6] FIG. 6 is a diagram illustrating an example of hardware of a reading support system. [Figure 7] FIG. 7 is a diagram illustrating an example of hardware of a reading support system. DETAILED DESCRIPTION OF THE INVENTION

[0018] The embodiments will be described in detail with reference to the drawings. However, the present invention is not limited to the following description, and it will be readily understood by those skilled in the art that various changes can be made in form and detail without departing from the spirit and scope of the present invention. Therefore, the present invention should not be interpreted as being limited to the description of the embodiments shown below.

[0019] In the configuration of the invention described below, the same parts or parts having similar functions are denoted by the same reference numerals in different drawings, and repeated explanations thereof will be omitted. Furthermore, when referring to similar functions, the same hatch pattern may be used and no particular reference numeral may be assigned.

[0020] Furthermore, for ease of understanding, the position, size, range, etc. of each component shown in the drawings may not represent the actual position, size, range, etc. Therefore, the disclosed invention is not necessarily limited to the position, size, range, etc. disclosed in the drawings.

[0021] (Embodiment 1) In this embodiment, a reading assistance system and a reading assistance method according to one embodiment of the present invention will be described with reference to FIGS.

[0022] In the reading comprehension assistance method of this embodiment, first, a document that a user wants to read (target document) and a sentence related to the information the user needs (query sentence) are acquired. The target document is divided into multiple blocks (e.g., paragraphs), and word embeddings are acquired for each block. Also, embeddings for words included in the query sentence are acquired. Next, words included in the blocks are searched for that match a word included in the query sentence. Then, for each matching word, the similarity (e.g., cosine similarity) between the embeddings for the word in the block and the embeddings for the word in the query sentence is calculated. If there are multiple matching words, the sum of the similarities of the embeddings for each word is used as the score for the block. A block with a relatively high score is considered to have a high relevance to the query sentence. This makes it possible to present parts of the target document that are highly related or similar to the information in question. For example, the blocks of the target document can be sorted in descending order of score, and the blocks can be presented in descending order of relevance.

[0023] In the reading comprehension support method of this embodiment, when a question sentence in natural language is input, parts of the target document that are related to the question sentence can be presented. Since different embedded representations are used for the same word depending on the sentence, it is possible to present blocks that are more related or similar to the question sentence.

[0024] The query can contain one or more sentences. Because there is no need to select keywords to use in the search, users can easily find the desired information from documents.

[0025] Unless otherwise specified herein, a document is a description of an event in natural language that is computerized and machine-readable. Examples of documents include, but are not limited to, patent applications, legal precedents, contracts, terms and conditions, product manuals, novels, publications, white papers, and technical documents. In addition, in this specification, a sentence includes one or more sentences.

[0026] In this specification, a word is the smallest linguistic unit that has linguistic sounds, meaning, and grammatical functions. However, embeddings may be obtained for subwords obtained by further dividing a word. For example, the English word "transformer" can be divided into the subwords "transform" and "er," and an embedding representation can be assigned to each of them. Alternatively, an embedding representation can be assigned to a phrase consisting of two or more words. In this specification, the subwords obtained by dividing a word are also referred to as words. In this specification, a phrase, word, or subword to which an embedding representation is assigned may also be referred to as a token.

[0027] In this embodiment, the embedded representations of words are obtained using a language model that can obtain different embedded representations for the same word depending on the distribution of surrounding words or context. Alternatively, the embedded representations can be obtained using a language model that can obtain different embedded representations for the same word depending on the context. Furthermore, a language model that can obtain embedded representations in which information about the position of a word in a sentence, segments (information about the connection of sentences), and tokens is embedded may be used as the embedded representations of words. Furthermore, a language model that has a self-attention function and obtains embedded representations by learning from both directions of a sentence may be used. BERT (Bidirectional Encoder Representations from Transformers) (see Non-Patent Document 1) is an example of a language model that can obtain different embedded representations for the same word depending on the distribution of surrounding words or context.

[0028] Figure 4 plots the embedded representations obtained by BERT for "carbon" in six English sentences containing "carbon" on the XY coordinate. The three plots (squares) on the left half are sentences containing "carbon" as an impurity in materials, while the three plots (diamonds) on the right half are sentences about "carbon" as an anode material. Figure 4 shows that even for the same "carbon," different embedded representations can be obtained depending on the context and sentence.

[0029] By using a language model that can obtain different distributed representations of the same word depending on the sentence it is included in, it is possible to accurately find blocks that are highly relevant to the information the user needs. For example, if the query sentence contains "carbon" as an anode material, the score of blocks that contain "carbon" as an anode material will be relatively high, and the score of blocks that contain "carbon" as an impurity will be relatively low.

[0030] [Reading support system] FIG. 1 is a block diagram showing the configuration of a reading support system 100. As shown in FIG.

[0031] The reading support system 100 may be provided in an information processing device such as a personal computer used by a user, or may be configured such that a processing unit of the reading support system 100 is provided in a server and the system is accessed and used from a client PC via a network.

[0032] The reading support system 100 includes a document reading unit 101, a question input unit 102, a block division unit 103, a distributed representation acquisition unit 104a, a distributed representation acquisition unit 104b, a word selection unit 105, a similarity calculation unit 106, a score display unit 107, and a sentence display unit 108.

[0033] The document reading unit 101 reads a document to be read and understood.

[0034] The document read by the document reading unit 101 may be a document stored in a personal computer used by a user, or may be a document stored in a storage device connected via a network.

[0035] The question input section 102 is a section where the user inputs a sentence to be specified for search.

[0036] The question sentence (also called a query sentence) can be input by directly entering any sentence, or by pasting text copied from a document file. Alternatively, a part of a document read by the document reading unit 101 can be arbitrarily specified by the user and read into the question sentence input unit 102.

[0037] The block dividing unit 103 divides the read document into blocks. The block dividing unit 103 can be called a document dividing unit.

[0038] A paragraph may be divided into a block, a sentence separated by a period or a comma may be divided into a block, or a predetermined number of paragraphs or sentences may be divided into a block. Some documents have paragraph numbers included in the document from the beginning, and the document may be divided into blocks according to the paragraph numbers.

[0039] The embedded representation acquisition unit 104a processes the document read by the document reading unit 101 for each block, and acquires embedded representations of the words included in the block.

[0040] The distributed expression acquisition unit 104b acquires distributed expressions of words included in the sentence input to the question input unit 102.

[0041] It is preferable that the distributed expression acquisition unit 104a and the distributed expression acquisition unit 104b basically use the same language model.

[0042] The word selection unit 105 is a part that selects words to be used in similarity calculation from among the words contained in the input question sentence.

[0043] You can select all words, select a specific part of speech such as a noun, or allow the user to freely select words. At least one word must be selected, and even if only one word is selected, different embedded representations can be obtained depending on the sentence and context, making scoring possible.

[0044] The similarity calculation unit 106 calculates the similarity of each block to the question sentence using the distributed representations of words obtained by the distributed representation acquisition unit 104a and the distributed representation acquisition unit 104b. The similarity calculation unit 106 can be called a similarity acquisition unit.

[0045] The score display unit 107 can display the score calculated by the similarity calculation unit 106 .

[0046] The text display unit 108 can display the document read by the document reading unit 101. The text display unit 108 may further display the text input to the question input unit 102.

[0047] It is preferable that the score display unit 107 and the sentence display unit 108 are synchronized. For example, the display method of the target document may be changed based on the score value, such as sorting sentence blocks in descending order of score, or displaying only blocks with scores equal to or greater than a predetermined value.

[0048] [Reading support methods] 2 and 3 are flowcharts illustrating the flow of processing executed by the reading support system 100. In other words, each of Fig. 2 and Fig. 3 can be said to be a flowchart illustrating an example of a reading support method according to one aspect of the present invention.

[0049] [Step S1: Obtain the target document] First, the document to be read is read by the document reading unit 101 of the reading support system 100 .

[0050] [Step S2: Divide the target document into multiple blocks] Next, the block dividing unit 103 divides the target document into a plurality of blocks.

[0051] [Step S3: Obtain distributed representations of words for each block] Next, the sentence is input to the distributed representation acquisition unit 104a for each block, and distributed representations of words are acquired. Specifically, the target document is input to a language model such as BERT for each block, and distributed representations of words are acquired.

[0052] [Step S4: Obtain the query sentence] Furthermore, a query sentence is acquired by the question sentence input unit 102 of the reading comprehension support system 100. The query sentence may be a sentence arbitrarily input by the user, or may be a sentence from a part of the target document that is of great interest to the user. FIG. 2 shows an example in which steps S4 and S5 are performed after step S3, but as shown in FIG. 3, steps S1 to S3 and steps S4 and S5 can be performed independently, and the order does not matter.

[0053] [Step S5: Obtain distributed representations of words contained in the query sentence] Next, the query sentence is input to the distributed representation acquisition unit 104b to acquire distributed representations of words. Specifically, the query sentence is input to a language model such as BERT to acquire distributed representations of words.

[0054] [Step S6: Calculate the block score] Next, the similarity calculation unit 106 searches for matching words between the words contained in each block and the words contained in the query sentence, and only if the words match, calculates the cosine similarity between the embedded representations of the matching words, and obtains the score of the block by calculating the sum of the cosine similarities within the block.

[0055] The word selection unit 105 may select words to be used in the similarity calculation from among the words included in the query sentence, and calculate the similarity only for the selected words.

[0056] Although the present embodiment shows an example in which the similarity is calculated using cosine similarity, other similarity calculation methods may be used.

[0057] The method for calculating the score for each block will be explained using Figure 5. Figure 5 shows an example of comparing blocks 1, 2, 3, and 4 of a target document with a query sentence. First, each block of the target document is searched for words that match words in the query sentence, and the cosine similarity of the embedded representations of those words is calculated only for those matching words. If there are multiple matching words in one block, the cosine similarity for each word is added to calculate the score for that block. For example, in block 1 shown in Figure 5, two words, word W1 and word W2, match in the query sentence. In this case, the score for block 1 is the sum of the cosine similarity for word W1 and the cosine similarity for word W2.

[0058] [Step S7: Output the calculated score] Then, blocks with high calculated scores can be presented to the user as blocks that are likely to contain the desired information.

[0059] As described above, the reading support system and method of this embodiment, when a user provides a document they wish to read and a sentence related to the information they need, can present blocks in the document that are highly relevant to the information the user needs. This eliminates the need for the user to select keywords, making it easier for the user to find the desired information in the document.

[0060] The reading support system and method of this embodiment use a language model that can obtain different distributed representations of the same word depending on the sentence it is in. This makes it possible to find with high accuracy blocks that are highly relevant to the information the user needs.

[0061] 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.

[0062] (Embodiment 2) In this embodiment, a reading assistance system according to one embodiment of the present invention will be described with reference to FIGS.

[0063] The reading support system of this embodiment can easily search for and acquire desired information from a document using the reading support method shown in the first embodiment.

[0064] <Reading support system configuration example 1> Figure 6 shows a block diagram of reading comprehension support system 200. Note that in the drawings attached to this specification, the components are classified by function and shown as independent blocks in the block diagram, but in reality, it is difficult to completely separate the components by function, and one component may be involved in multiple functions. Also, one function may be involved in multiple components; for example, the processing performed by processing unit 120 may be executed by different servers depending on the processing.

[0065] The reading support system 200 includes at least a processing unit 120. The reading support system 200 shown in FIG.

[0066] [Input section 110] A question sentence (query sentence) is supplied to the input unit 110 from outside the reading comprehension assistance system 200. A target document may also be supplied to the input unit 110 from outside the reading comprehension assistance system 200. The target document and query sentence supplied to the input unit 110 are each supplied to the processing unit 120, the storage unit 130, or the database 140 via a transmission path 160.

[0067] The target document and the query sentence are input as, for example, text data, audio data, or image data, and the target document is preferably input as text data.

[0068] Methods for inputting a query sentence include, for example, key input using a keyboard or touch panel, voice input using a microphone, reading from a recording medium, image input using a scanner or camera, and acquisition via communication.

[0069] The reading assistance system 200 may have a function of converting voice data into text data. For example, the processing unit 120 may have this function. Alternatively, the reading assistance system 200 may further have a voice conversion unit having this function.

[0070] The reading support system 200 may have an optical character recognition (OCR) function, which allows it to recognize characters included in image data and create text data. For example, the processing unit 120 may have this function. Alternatively, the reading support system 200 may further have a character recognition unit having this function.

[0071] [Processing section 120] The processing unit 120 has a function of performing calculations using data supplied from the input unit 110, the storage unit 130, the database 140, etc. The processing unit 120 can supply the calculation results to the storage unit 130, the database 140, the display unit 150, etc.

[0072] The processing unit 120 has a function of dividing a document into a plurality of blocks. For example, the processing unit 120 may have a function of dividing a document into a plurality of blocks, such as by chapter, by paragraph, or by a predetermined number of sentences.

[0073] The processing unit 120 has a function of acquiring embedded representations of words, such as words included in a block of a target document or words included in a query sentence.

[0074] The processing unit 120 has a function of extracting words from a query sentence, thereby making it possible to select words to be used in similarity calculation from among the words contained in the query sentence.

[0075] The processing unit 120 has a function of calculating the similarity between embedded representations of words.

[0076] The processing unit 120 may be a transistor having a metal oxide in its channel formation region. Because the off-state current of the transistor is extremely low, using the transistor as a switch for retaining charge (data) flowing into a capacitor functioning as a memory element can ensure a long data retention period. By utilizing this characteristic in at least one of the register and cache memory of the processing unit 120, the processing unit 120 can be operated only when necessary, and can be turned off at other times by saving the information from the previous processing in the memory element. In other words, normally-off computing becomes possible, enabling a reading comprehension assistance system to consume less power.

[0077] Note that in this specification and the like, a transistor whose channel formation region is formed using an oxide semiconductor is referred to as an oxide semiconductor transistor or an OS transistor. The channel formation region of an OS transistor preferably contains a metal oxide.

[0078] 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 increased. Furthermore, the metal oxide contained in the channel formation region is preferably an oxide semiconductor containing element M. The element M is preferably aluminum (Al), gallium (Ga), or tin (Sn). Other elements applicable to 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). However, a combination of the above elements may be used as element M. The element M is, for example, an element having a high binding energy with oxygen. For example, it is an element whose bond energy with oxygen is higher than that of indium. Furthermore, the metal oxide contained in the channel formation region preferably contains zinc (Zn). Metal oxides containing zinc may be more likely to crystallize.

[0079] The metal oxide contained in the channel formation region is not limited to a metal oxide containing indium. The semiconductor layer may be, for example, a metal oxide containing zinc but not indium, such as zinc tin oxide or gallium tin oxide, a metal oxide containing gallium, or a metal oxide containing tin.

[0080] The processing section 120 may also use a transistor containing silicon in the channel formation region.

[0081] The processing section 120 may include a combination of a transistor including an oxide semiconductor in a channel formation region and a transistor including silicon in a channel formation region.

[0082] The processing unit 120 includes, for example, an arithmetic circuit or a central processing unit (CPU).

[0083] The processing unit 120 may have a microprocessor such as a DSP (Digital Signal Processor) or a GPU (Graphics Processing Unit). The microprocessor may be implemented by a PLD (Programmable Logic Device) such as an FPGA (Field Programmable Gate Array) or an FPAA (Field Programmable Analog Array). The processing unit 120 can perform various data processing and program control by interpreting and executing instructions from various programs using the processor. Programs that can be executed by the processor are stored in at least one of a memory area of ​​the processor and the storage unit 130.

[0084] The processing unit 120 may include a main memory, which includes at least one of a volatile memory such as a RAM and a non-volatile memory such as a ROM.

[0085] The RAM may be, for example, a dynamic random access memory (DRAM) or a static random access memory (SRAM), and is used as a virtual memory space allocated to the processing unit 120 as a working space. The operating system, application programs, program modules, program data, lookup tables, and the like stored in the storage unit 130 are loaded into the RAM for execution. The data, programs, and program modules loaded into the RAM are each directly accessed and operated by the processing unit 120.

[0086] ROM can store BIOS (Basic Input / Output System) and firmware, which do not require rewriting. 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 allows stored data to be erased by exposure to ultraviolet light, EEPROM (Electrically Erasable Programmable Read Only Memory), and flash memory.

[0087] [Storage section 130] The storage unit 130 has a function of storing a program executed by the processing unit 120. The storage unit 130 may also have a function of storing, for example, calculation results generated by the processing unit 120 and data input to the input unit 110.

[0088] The storage unit 130 includes at least one of a volatile memory and a nonvolatile memory. The storage unit 130 may include a volatile memory such as a DRAM or an SRAM. The storage unit 130 may include a nonvolatile memory such as a ReRAM (Resistive Random Access Memory, also called a resistance change memory), a PRAM (Phase change Random Access Memory), an FeRAM (Ferroelectric Random Access Memory), an MRAM (Magnetoresistive Random Access Memory, also called a magnetoresistive memory), or a flash memory. The storage unit 130 may also include a recording media drive such as a hard disk drive (HDD) or a solid state drive (SSD).

[0089] [Database 140] The reading support system may include a database 140. For example, the database 140 has a function of storing a plurality of documents. For example, one of the documents stored in the database 140 can be set as a target document, and the reading support method of one aspect of the present invention can be used to perform reading of the target document. Note that the storage unit 130 and the database 140 do not need to be separated from each other. For example, the reading support system may include a storage unit having the functions of both the storage unit 130 and the database 140.

[0090] The memories included in the processing unit 120, the storage unit 130, and the database 140 can be considered to be examples of non-transitory computer-readable storage media.

[0091] [Display section 150] The display unit 150 has a function of displaying the calculation results of the processing unit 120. The display unit 150 also has a function of displaying the target document. The display unit 150 may also have a function of displaying the query sentence.

[0092] The reading support system 200 may also have an output unit. The output unit has a function of supplying data to the outside.

[0093] [Transmission Line 160] The transmission path 160 has a function of transmitting various types of data. Data can be transmitted and received between the input unit 110, the processing unit 120, the storage unit 130, the database 140, and the display unit 150 via the transmission path 160. For example, data such as a target document is transmitted and received via the transmission path 160.

[0094] <Reading support system configuration example 2> 7 shows a block diagram of the reading support system 210. The reading support system 210 includes a server 220 and a terminal 230 (such as a personal computer).

[0095] The server 220 includes a communication unit 161a, a transmission path 162, a processing unit 120, and a storage unit 170. Although not shown in Fig. 7, the server 220 may further include an input / output unit and the like.

[0096] The terminal 230 includes a communication unit 161b, a transmission path 164, a processing unit 180, a storage unit 130, and a display unit 150. Although not shown in FIG. 7, the terminal 230 may further include a database or the like.

[0097] A user of the reading support system 210 inputs a question (query sentence) into the input unit 110 of the terminal 230. The question sentence is transmitted from the communication unit 161b of the terminal 230 to the communication unit 161a of the server 220.

[0098] The question received by the communication unit 161a is stored in the storage unit 170 via the transmission path 162. Alternatively, the question may be supplied to the processing unit 120 directly from the communication unit 161a.

[0099] The block division, distributed representation acquisition, and similarity calculation described in the first embodiment each require high processing power. The processing unit 120 of the server 220 has higher processing power than the processing unit 180 of the terminal 230. Therefore, it is preferable that these processes are each performed by the processing unit 120.

[0100] Then, the processing unit 120 calculates a score for the block. The score is stored in the storage unit 170 via the transmission path 162. Alternatively, the score may be supplied directly from the processing unit 120 to the communication unit 161a. The score is transmitted from the communication unit 161a of the server 220 to the communication unit 161b of the terminal 230. The score is displayed on the display unit 150 of the terminal 230.

[0101] [Transmission path 162 and transmission path 164] The transmission paths 162 and 164 have a function of transmitting data. Data can be transmitted and received between the communication unit 161a, the processing unit 120, and the storage unit 170 via the transmission path 162. Data can be transmitted and received between the input unit 110, the communication unit 161b, the processing unit 180, the storage unit 130, and the display unit 150 via the transmission path 164.

[0102] [Processing Unit 120 and Processing Unit 180] The processing unit 120 has a function of performing calculations using data supplied from the communication unit 161a, the storage unit 170, etc. The processing unit 180 has a function of performing calculations using data supplied from the communication unit 161b, the storage unit 130, the display unit 150, etc. For the processing unit 120 and the processing unit 180, refer to the description of the processing unit 120. It is preferable that the processing unit 120 has higher processing power than the processing unit 180.

[0103] [Storage section 130] The storage unit 130 has a function of storing a program executed by the processing unit 180. The storage unit 130 also has a function of storing the calculation results generated by the processing unit 180, the data input to the communication unit 161b, the data input to the input unit 110, and the like.

[0104] [Storage section 170] The storage unit 170 has a function of storing a plurality of documents, calculation results generated by the processing unit 120, data input to the communication unit 161a, and the like.

[0105] [Communication Unit 161a and Communication Unit 161b] Using the communication units 161a and 161b, data can be transmitted and received between the server 220 and the terminal 230. A hub, a router, a modem, or the like can be used as the communication units 161a and 161b. Data can be transmitted and received using either a wired connection or wirelessly (for example, radio waves, infrared rays, etc.).

[0106] This embodiment mode can be combined with other embodiment modes as appropriate. [Explanation of symbols]

[0107] W1: word, W2: word, 1: block, 2: block, 3: block, 4: block, 100: reading support system, 101: document reading unit, 102: question input unit, 103: block division unit, 104a: distributed representation acquisition unit, 104b: distributed representation acquisition unit, 105: word selection unit, 106: similarity calculation unit, 107: score display unit, 108: sentence display unit, 110: input unit, 120: processing unit, 130: memory unit, 140: database, 150: display unit, 160: transmission path, 161a: communication unit, 161b: communication unit, 162: transmission path, 164: transmission path, 170: memory unit, 180: processing unit, 200: reading support system, 210: reading support system, 220: server, 230: terminal

Claims

1. a first part for reading a target document; a second part for dividing the scanned target document into a plurality of blocks; a third part for obtaining a distributed representation of words for each of the plurality of blocks; a fourth part for reading a query sentence; a fifth part for extracting words included in the query sentence and obtaining embedded representations of the words; a sixth part for comparing embedded representations of words between the query sentence and the words in each of the blocks, calculating similarities between the words, and calculating scores for the blocks; Including, In the sixth part, for each of the plurality of blocks, words included in the block are searched for that match a word included in the query sentence, and for the matched words, a similarity between a distributed representation of the word in the block and a distributed representation of the word in the query sentence is calculated, and a score of the block is calculated; In the sixth part, if there are multiple words that match between the query sentence and the block, the sum of the similarities of the distributed representations for each word is set as the score of the block; The reading assistance system, wherein the third portion and the fifth portion have transistors each including a metal oxide in a channel formation region.

2. In claim 1, A reading support system, wherein each of the plurality of blocks includes one or more paragraphs of the target document.

3. In claim 1, A reading support system, wherein each of the plurality of blocks includes one or more sentences.

4. In any one of claims 1 to 3, a reading comprehension assistance system that, in the sixth part, calculates the similarity only for selected words among words included in the query sentence and calculates a score for the block.

5. In any one of claims 1 to 4, The metal oxide includes indium.

Images