Document search apparatus

Abstract

To realize highly accurate document search, particularly search for a document relating to intellectual property through a simple input method.SOLUTION: A document search system includes a processing unit. The processing unit has a function of extracting a keyword included in text data, a function of extracting a related term of the keyword from words included in multiple pieces of first reference text analysis data, a function of adding a weight to each of the keyword and the related term, a function of scoring each of multiple pieces of second reference text analysis data on the basis of the weight, a function of ranking the pieces of second reference text analysis data on the basis of the scores to generate ranking data, and a function of outputting the ranking data.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] One aspect of the present invention is a document search system, a document search method, a program, and a non-transitory computer. One aspect of the present invention relates to a document search system for intellectual property and It relates to document retrieval methods.

[0002] Note that one embodiment of the present invention is not limited to the above technical field. Examples of the semiconductor device include a semiconductor device, a display device, a light-emitting device, a power storage device, a memory device, an electronic device, a lighting device, Input devices (e.g., touch sensors, etc.), input / output devices (e.g., touch panels, etc.), These driving methods and manufacturing methods can be cited as examples. [Background technology]

[0003] Conducting a prior art search on pre-filing inventions to determine whether relevant intellectual property rights exist The patent documents and papers obtained from the prior art search can be used to These are used to confirm the novelty and inventive step of an invention and to decide whether or not to apply for a patent. In addition, by conducting invalidation document searches of patent documents, you can check whether your own patent rights are valid. Investigate whether your patent is likely to be invalid or whether you can invalidate patents owned by others. This can be done.

[0004] For example, a user can enter a keyword into a system for searching patent documents. It is possible to output patent documents containing the keyword.

[0005] In order to perform a high-precision prior art search using such a system, appropriate keywords are required. Searching for patents and extracting the necessary patents from the many patents output It requires high skill from the user.

[0006] In addition, the use of artificial intelligence (AI) is being considered for various purposes. By utilizing networks, it is possible to develop computers with higher performance than conventional von Neumann type computers. In recent years, artificial neural networks have been developed on electronic circuits. Various research projects are being conducted to build such a system.

[0007] For example, Patent Document 1 discloses a transistor having an oxide semiconductor in a channel formation region. The memory device stores the weight data required for calculations using an artificial neural network. The invention has been disclosed. [Prior art documents] [Patent documents]

[0008] [Patent Document 1] US Patent Publication No. 2016 / 0343452 Summary of the Invention [Problem to be solved by the invention]

[0009] Therefore, one aspect of the present invention is to provide a document retrieval system capable of retrieving documents with high accuracy. Another object of the present invention is to provide a document search system capable of searching documents with high accuracy. Another object of the present invention is to provide a simple and easy input method. One of the challenges is to realize highly accurate document retrieval, especially for documents related to intellectual property. do.

[0010] Note that the description of these problems does not preclude the existence of other problems. However, it is not necessary to solve all of these problems. It is possible to extract other issues from the description in the section. [Means for solving the problem]

[0011] One aspect of the present invention is a document search system having a processing unit. A function for extracting keywords contained in the first reference text analysis data and a function for extracting words contained in the first reference text analysis data. The function to extract related words of keywords from the list and to provide a function to extract related words of keywords and related words and a function of assigning a weight to each of the plurality of second reference sentence analysis data. A function for assigning a score based on the weight and a function for generating multiple second reference sentence analysis data based on the score. It has a function to generate ranking data by ranking the data, and a function to output the ranking data. It has the function of:

[0012] In one aspect of the present invention, a keyword included in text data is extracted, and a plurality of first reference texts are generated. Extract related words from the words included in the analysis data, and A weight is assigned to each of the words, and the weight is assigned to each of the plurality of second reference sentence analysis data. A score is assigned to each of the sentences based on the weight, and a plurality of second reference sentence analysis data are selected based on the score. The ranking data is generated by ranking the data, and the ranking data is output. This is a book search method.

[0013] One aspect of the present invention includes a first step of extracting keywords contained in text data, Extracting words related to keywords from words included in the first reference text analysis data. Step 2 and the third step of assigning weights to each keyword and related term. and for each of the plurality of second reference text analysis data, Scoring is based on the weight of the keywords or related words that match the words in the data. A fourth step is to rank the plurality of second reference text analysis data based on the scores. A fifth step of generating ranking data by the above-mentioned method and a program for causing a processor to execute the above-mentioned steps. Another aspect of the present invention is a non-transitory computer having the above program stored therein. It is a readable storage medium.

[0014] The weight of the keyword is determined based on the weight of the keyword based on a plurality of first reference sentence analysis data or a plurality of second reference sentence analysis data. It is preferable that the value is based on the inverse document frequency in the reference text analysis data of 2. The weight of a word is determined by the similarity between the embedded vector of related words and the embedded vector of the keyword. Preferably, is a product of a distance-based value and a keyword weight.

[0015] For a second reference text analysis data having words that match the keyword or related words, It is preferable to assign points.

[0016] The plurality of first reference sentence analysis data are identical to the plurality of second reference sentence analysis data. This is also fine.

[0017] The related words are machine-learned to obtain distributed representations of words included in multiple first reference text analysis data. It is preferable to extract the distributed representation vector obtained by the above method.

[0018] The related words are obtained by extracting the distributed representation of words from among the words included in the first reference text analysis data. Based on the similarity or distance between the vector and the distributed representation vector of the keyword It is preferable to extract the distributed representation vector of words using a neural network. It is preferably a generated vector.

[0019] A morphological analysis of text data is performed to extract keywords contained in the text data. The system has the functions of generating analysis data by analyzing the data and extracting keywords from the analysis data. It is preferable that the keywords are selected from a plurality of first words included in the analysis data. A high inverse document frequency in the reference text analysis data or a plurality of second reference text analysis data It is preferable to extract the images based on their length.

[0020] The weights are preferably user changeable.

[0021] The first reference sentence analysis data is generated by performing a morphological analysis of the first reference sentence data. The second reference sentence analysis data is data obtained by dividing the morpheme The data is preferably generated by performing an analysis.

[0022] The document search system according to one aspect of the present invention preferably includes an electronic device and a server. The electronic device has a first communication unit. The server has the above-mentioned processing unit and a second communication unit. The first communication unit transmits the document to the server by one or both of wired communication and wireless communication. The processing unit has a function of supplying ranking data to the second communication unit. The second communication unit has a function of communicating with the device by one or both of wired communication and wireless communication. It has a function of supplying ranking data to electronic devices.

[0023] The processing section may include a transistor having a metal oxide in a channel forming region, The semiconductor device may have a transistor having silicon in a channel forming region. Effect of the Invention

[0024] According to one aspect of the present invention, a document search system capable of searching documents with high accuracy can be provided. Furthermore, according to one aspect of the present invention, a document retrieval method capable of retrieving documents with high accuracy can be provided. According to one aspect of the present invention, a highly accurate document search, particularly an intellectual property search, can be performed by a simple input method. It is possible to search for documents related to the above.

[0025] The description of these effects does not preclude the existence of other effects. However, it is not necessary to have all of these effects. However, it is possible to extract effects other than these. [Brief description of the drawings]

[0026] [Figure 1] FIG. 1 is a block diagram showing an example of a document search system. [Diagram 2] FIG. 11 is a flow diagram showing an example of a method for generating search data. [Diagram 3] 6A to 6C are diagrams showing an example of a method for generating search data. [Figure 4] FIG. 1 is a flow diagram showing an example of a document search method. [Diagram 5] FIG. 1 is a flow diagram showing an example of a document search method. [Figure 6] 1A to 1C are diagrams showing an example of a document search method. [Figure 7] 1A and 1B are diagrams showing an example of a document search method. [Figure 8] FIG. 1 is a flow diagram showing an example of a document search method. [Figure 9] FIG. 1 is a flow diagram showing an example of a document search method. [Figure 10] 1A and 1B are diagrams showing an example of a document search method. [Figure 11]FIG. 1 is a block diagram showing an example of a document search system. [Figure 12] (A) and (B) Diagram showing an example of a neural network configuration. [Figure 13] 1A and 1B are diagrams illustrating examples of the configuration of a semiconductor device. [Figure 14] FIG. 2 shows an example of the configuration of a memory cell. [Figure 15] FIG. 4 is a diagram showing a configuration example of an offset circuit. [Figure 16] Timing chart. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0027] The embodiment will be described in detail with reference to the drawings. However, the present invention is not limited to the following description. The present invention is not limited to the above embodiment, and various changes and modifications may be made in the form and details without departing from the spirit and scope of the present invention. It will be easily understood by those skilled in the art that the above-mentioned invention can be obtained by the following embodiments. It should not be construed as being limited to the contents described.

[0028] In the configuration of the invention described below, the same parts or parts having similar functions are The same reference numerals are used in common between different drawings, and the repeated explanations are omitted. When referring to a function, the hatch pattern may be the same and no particular symbol may be used.

[0029] In addition, the position, size, range, etc. of each component shown in the drawings are not necessarily the same as in the actual device for ease of understanding. Therefore, the disclosed invention may not necessarily represent the position, size, range, etc. Furthermore, the present invention is not limited to the position, size, range, etc. disclosed in the drawings.

[0030] The words "membrane" and "layer" may be interchangeable depending on the situation. For example, the term "conductive layer" can be used interchangeably with the term "conductive film." ". Alternatively, for example, the term "insulating film" may be changed to It is possible to change the term to "insulating layer."

[0031] (Embodiment 1) In this embodiment, a document retrieval system and a document retrieval method according to one embodiment of the present invention will be described with reference to FIGS. This will be explained using FIG.

[0032] One aspect of the present invention is a document retrieval system having a processing unit. and a function of extracting keywords contained in the first reference text analysis data. The function to extract related words from keywords and to search for keywords and related words. and a function of assigning weights to each of the plurality of second reference text analysis data. and a function of assigning a score based on the score alone and a function of assigning a score based on a plurality of second reference sentence analysis data. A function to generate ranking data by ranking the items and output the ranking data. It has the function.

[0033] In one embodiment of the document search system of the present invention, a keyword is generated by using the first reference text analysis data. The second reference text analysis data is used as a search target to extract words related to the text data. Or similar data can be searched for.

[0034] The first reference text analysis data and the second reference text analysis data may be the same. In this case, in the document search system according to one embodiment of the present invention, the reference text analysis data to be searched is The first reference text analysis data can be used to extract related words of the keyword. may include part or all of the second reference text analysis data.

[0035] In other words, the processing unit of the document search system according to one aspect of the present invention searches for keywords contained in the text data. The function to extract keywords from words contained in multiple reference text analysis data The function to extract related words of a keyword and to assign weights to each of the keywords and related words is also included. The function of assigning points based on the weights to each of multiple reference text analysis data is also included. The function of ranking by ranking multiple reference text analysis data based on the score. The ranking data generating unit 20 may have a function of generating ranking data and a function of outputting ranking data.

[0036] In a document search system, when a user selects keywords to use for a search, the user Not only keywords themselves, but also synonyms, similar words, and variations in spelling of the keywords It is necessary to select keywords to be used in a search while taking into consideration the above. The selection of the model is a burden for the user, and the difference in the user's skill is likely to occur. It is also possible to find the necessary document from the many documents output by the document search system. It is a burden.

[0037] Here, the document search system according to one aspect of the present invention is a system for searching keywords included in text data and The document search function of the present invention is a function to extract related words of the keyword. Users of the search system do not have to select the keywords to be used in the search. A document search system that uses text data, which is larger than keywords, as is. Users can also select keywords and related words themselves. There is no need to select words from scratch. Instead, you can refer to keywords and related words extracted by the document search system. By using the search function, you can add, modify, or delete keywords and related terms. This reduces the burden on users in searching and reduces the difference in search results depending on the user's skill level. can.

[0038] In particular, the document search system according to one embodiment of the present invention is a system for searching a single word contained in a plurality of reference text analysis data. It has the function of extracting words related to keywords from among words. Words included in the existing concept dictionary When extracting related words from a keyword, it is necessary to consider the unique table contained in the data to be searched. However, in one aspect of the present invention, it is difficult to extract keywords as related words. Data prepared for extracting related words of the word (first reference text analysis data), and is a key word that is included in the data to be searched (second reference text analysis data). Extract related words of the word. This allows the unique notation to be extracted as a related word. This is preferable because it makes it easier to search and reduces missed searches.

[0039] Furthermore, the document search system according to one embodiment of the present invention is Each keyword and related word has a weight. Then, the reference text analysis data where the keyword or related word is found is scored based on the weight. The document search system according to one aspect of the present invention can then assign a score based on the score. Based on the ranking, the search target reference text analysis data is ranked to generate ranking data. The search results are ranked according to relevance or similarity. This allows users to easily find the documents they need from the search results, improving work efficiency. In this way, the document search system according to one embodiment of the present invention is simple and This allows documents to be searched for with high accuracy.

[0040] The weight of a keyword is determined based on the weight of the keyword in the first or second reference text analysis data. Inverse Document Frequency (IDF) The IDF is a measure of how unlikely a word is to appear in a document. The IDF of a word that appears in many documents is small, and the IDF of a word that appears in only a few documents is small. The IDF is high. Therefore, words with high IDF are the first or second reference sentence analysis. It can be said to be a characteristic word in the data.

[0041] The extraction of keywords from text data itself also involves searching multiple first sets of words contained in the text data. Alternatively, it can be performed based on the IDF in the second reference sentence analysis data. For example, Words with an IDF greater than or equal to a certain value may be extracted as keywords. Any number of words may be extracted as keywords.

[0042] The keyword is either the first reference sentence analysis data or the second reference sentence analysis data. The second reference text analysis data to be searched may be extracted based on the IDF. Extracting keywords from text data based on IDF is more effective in extracting keywords from target documents. However, this method is preferable when there are only a few documents to search. Based on the IDF in the first reference text analysis data, keywords are extracted from the text data. It may be easier to extract keywords from text data by extracting words from the text.

[0043] Or, word part-of-speech information obtained by morphological analysis of text data entered by the user. For example, when analyzing Japanese text using morphological analysis, In addition, when performing morphological analysis on English sentences, it is preferable to extract adjectives, nouns, and It is preferable to extract words and verbs.

[0044] Related words include synonyms, antonyms, hypernyms, and hyponyms. Among the words contained in the multiple reference text analysis data, the distributed representation vector of the word is Extraction is based on the similarity or distance between the distributed representation vector of the keyword. It is preferable that the keyword included in the text data input by the user is compared with the reference text. It is possible to extract synonyms and similar words contained in the analysis data as related words. This can improve search accuracy.

[0045] The weight of related words is the similarity between the distributed representation vector of related words and the distributed representation vector of keywords. Alternatively, it is preferable that the weight is a product of a value based on the distance and the weight of the keyword. The weight is set based on both the relevance of the related words to the keyword and the weight of the keyword itself. By setting the above criteria, the accuracy of the ranking can be improved.

[0046] Document search systems use artificial intelligence (AI) for at least some of their processing. It is preferable to use the Intelligent Intelligence (Intelligence).

[0047] Document retrieval systems, in particular, use artificial neural networks (ANNs). Neural Network (hereafter simply referred to as neural network) is used. It is preferable that the neural network is implemented as a circuit (hardware) or a program. This is realized by a system (software).

[0048] For example, it is preferable to use machine learning to generate distributed representation vectors of words. More preferably, a neural network is used. A distributed representation database is obtained by machine learning distributed representations of words contained in the reference text analysis data. It is preferable to use a vector to extract related words. This makes it easier to extract related words and to assign weights to them. Accuracy can be improved.

[0049] In this specification, the term "neural network" refers to a network that imitates the neural circuitry of living organisms and that learns by learning. This refers to a general model in which the strength of connections between neurons is determined by the neural network, giving the brain problem-solving capabilities. A neural network has an input layer, an intermediate layer (hidden layer), and an output layer.

[0050] In this specification, when discussing neural networks, Determining the connection strengths (also called weight coefficients) between neurons is called "learning." There are cases.

[0051] In this specification, the neural network is constructed using the connection weights obtained by learning. The process of constructing evidence and deriving a new conclusion from it is sometimes called "inference."

[0052] <1. Document search system configuration example 1> In this embodiment, the document search system can be used to search for intellectual property. The document search system according to one embodiment of the present invention is The present invention is not limited to searches for intellectual property, but can also be used for searches for other purposes.

[0053] FIG. 1 shows a block diagram of a document retrieval system 100. The block diagram shows components classified by function as independent blocks. However, in reality, it is difficult to completely separate components by function, and one component may be complex. A function may involve multiple components. For example, the two processes performed by the processing unit 103 may be executed by different servers. This may be the case.

[0054] The document retrieval system 100 includes at least a processing unit 103. The system 100 further includes an input unit 101, a transmission path 102, a storage unit 105, a database 106, and a 07, and output unit 109.

[0055] [Input section 101] Data is supplied to the input unit 101 from outside the document search system 100. The data supplied to the data processing unit 103, the storage unit 105, or the like is transmitted through a transmission line 102 to the data processing unit 103, the storage unit 105, or the data processing unit 106. The data is fed to the database 107.

[0056] [Transmission path 102] The transmission path 102 has a function of transmitting data. Data is transmitted and received between the data input unit 105, the database 107, and the output unit 109 via a transmission line 102. This can be done via:

[0057] [Processing section 103] The processing unit 103 receives information provided from the input unit 101, the storage unit 105, the database 107, etc. The processing unit 103 has a function of performing calculations, inferences, etc. using the data. The results can be supplied to the storage unit 105, the database 107, the output unit 109, etc. Cut.

[0058] The processing section 103 may use a transistor having a metal oxide in a channel forming region. Since the off-state current of the transistor is extremely small, the transistor can be preferably used as a memory element. It is used as a switch to hold the charge (data) that has flowed into the capacitive element that functions as a This allows data to be retained for a long period of time. By using the memory for at least one of the register and the cache memory of the memory unit 103, The processing unit 103 is operated only when necessary, and in other cases, the information of the immediately preceding processing is stored in the memory element In other words, the processing unit 103 can be turned off by shifting the processing unit 103 to the normally Off-chip computing is now possible, and the power consumption of the document search system can be reduced. Cut.

[0059] In this specification and the like, a semiconductor device using an oxide semiconductor or a metal oxide for a channel formation region The transistor is called an Oxide Semiconductor transistor, or OST. The channel formation region of an OS transistor often contains a metal oxide. preferable.

[0060] In this specification, the term "metal oxide" refers to a metal oxide in a broad sense. Metal oxides are oxide insulators and oxide conductors (including transparent oxide conductors). , oxide semiconductor (also called OS) For example, when a metal oxide is used for the semiconductor layer of a transistor, the metal Oxides are sometimes called oxide semiconductors. In other words, metal oxides have amplifying and rectifying properties. and switching action, the metal oxide is Metal oxide semiconductor, abbreviated as OS. This can be done.

[0061] The metal oxide of the channel formation region preferably contains indium (In). When the metal oxide in the panel formation region contains indium, the OS transistor The carrier mobility (electron mobility) of the semiconductor is increased. The oxide is preferably an oxide semiconductor containing an element M. The element M is preferably aluminum. Al, Ga, or Sn. Other elements M may be used. Available elements include boron (B), silicon (Si), titanium (Ti), and iron (Fe). , Nickel (Ni), Germanium (Ge), Yttrium (Y), Zirconium (Zr ), Molybdenum (Mo), Lanthanum (La), Cerium (Ce), Neodymium (Nd), Hf (Hf), Tantalum (Ta), Tungsten (W), etc. However, the element M In some cases, the above elements may be combined. The element M is, for example, oxygen. For example, the bond energy with oxygen is higher than that with indium. In addition, the metal oxide in the channel formation region contains zinc (Zn). Metal oxides containing zinc are preferred. Metal oxides containing zinc may tend to crystallize. .

[0062] The metal oxide contained in the channel formation region is not limited to a metal oxide containing indium. The semiconductor layer may be an indium-free material such as zinc tin oxide or gallium tin oxide. , metal oxides containing zinc, metal oxides containing gallium, metal oxides containing tin, etc. It's okay.

[0063] The processing section 103 may also include a transistor containing silicon in a channel forming region. stomach.

[0064] The processing section 103 includes a transistor having an oxide semiconductor in a channel formation region and a It is preferable to use a transistor including silicon in a panel formation region in combination with the transistor.

[0065] The processing unit 103 is, for example, an arithmetic circuit or a central processing unit (CPU: Central Processing Unit). ocessing Unit) etc.

[0066] The processing unit 103 includes a DSP (Digital Signal Processor), a GP U (Graphics Processing Unit) and other microprocessors The microprocessor may be a Field Programmable Gate Array (FPGA). Gate Array), FPAA (Field Programmable A PLD (Programmable Logic Device) The processing unit 103 may be realized by a processor. It performs various data processing and program control by interpreting and executing instructions from various programs. The program that can be executed by the processor can be executed by the processor. The information is stored in at least one of the memory area and the storage unit 105 .

[0067] The processing unit 103 may have a main memory. The main memory may be a random access memory (RAM). Volatile memory such as volatile memory (M Access Memory) and ROM (Read Only Memory) The optical fiber has at least one of a non-volatile memory such as a volatile memory (VRAM) and a non-volatile memory such as a volatile memory (VRAM).

[0068] As an example of RAM, DRAM (Dynamic Random Access Memory) mory), SRAM (Static Random Access Memory), etc. is used, and a memory space is virtually allocated and used as a working space for the processing unit 103. The operating system and application programs stored in the storage unit 105 , program modules, program data, and lookup tables, etc. These data, programs, and processes loaded into RAM are Each of the program modules is directly accessed and operated by the processing unit 103.

[0069] The ROM contains a BIOS (Basic Input / Output) that does not require rewriting. It can store the ROM, ... ROM, OTPROM (One Time Programmable Read Only Memory), EPROM (Erasable Programmable EPROM is a type of memory that can be read and written by ultraviolet light. UV-EPROM (Ultra-Violet) Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmability) e Read Only Memory, flash memory, etc.

[0070] [Storage section 105] The storage unit 105 has a function of storing the program executed by the processing unit 103. The memory unit 105 stores the calculation results and inference results generated by the processing unit 103, as well as the It may also have a function of storing input data, etc.

[0071] The storage unit 105 includes at least one of a volatile memory and a non-volatile memory. The unit 105 may include a volatile memory such as a DRAM or an SRAM. The unit 105 is, for example, a resistive random access memory (ReRAM). Memory, also called resistive memory), PRAM (Phase change RAM) andom Access Memory), FeRAM (Ferroelectric Random Access Memory), MRAM (Magnetoresis tive Random Access Memory, also known as magnetoresistive memory), Alternatively, the storage unit 105 may have a non-volatile memory such as a flash memory. Hard Disk Drive (HDD) and solid Recording media drives such as solid state drives (SSDs) It may have a live.

[0072] [Database 107] The database 107 includes at least the reference text analysis data to be searched, the IDF data, The database 107 has a function of storing the data and vector data. 103 generates calculation results and inference results, and data input to the input unit 101, etc. The storage unit 105 and the database 107 may have a function of storing the information. For example, the document search system may be configured as a storage unit 105 and a database. The device 105 may have a storage unit having both the functions of the device 107 and the storage unit 108.

[0073] [Output section 109] The output unit 109 has a function of supplying data to the outside of the document search system 100. For example, the calculation results or inference results in the processing unit 103 can be supplied to the outside.

[0074] <2. Document search method> First, referring to FIG. 2 and FIG. 3, in order to perform a search using the document search system 100, The process to be performed will be described. A flowchart is shown in FIG. 2, and each step shown in FIG. The diagram of each data shown in FIG. 3 is an example, and the present invention is not limited to this example. .

[0075] In the following, we will discuss the reference text data used to extract related words of keywords and the search The target reference text data is the same (reference text data TD ref In the case of As mentioned above, even if these two reference sentence data are different, For example, vector data VD (described later) may be generated using the first reference text data. In addition, the second reference text data may be used as the search target. In this case, the IDF data (described later) may be used. The ID may be generated using the first reference sentence data, or may be generated using the second reference sentence data. It may be generated by

[0076] [Step S1] First, reference text data TD is input to the input unit 101. ref Enter multiple (Figure 2 and Figure 3 (A )).

[0077] In the document search method of the present embodiment, a plurality of documents prepared in advance are used as search targets. It is possible to search for documents related to or similar to the input document. Reference Document Data TD ref is text data of a document to be searched in the document search system 100. Reference Text Data TD ref The data used as the search result depends on the purpose of the document search system 100. The appropriate selection can be made according to the circumstances.

[0078] Reference Text Data TD ref is input to the input unit 101 from outside the document search system 100. And the reference text data TD ref is transmitted from an input unit 101 via a transmission line 102. and supplied to the processing unit 103. Alternatively, the reference text data TD ref is the input section 10 1, the data is stored in the storage unit 105 or the database 107 via the transmission line 102, and The data is supplied from the memory unit 105 or the database 107 to the processing unit 103 via the transmission line 102. This may be done.

[0079] In FIG. 3(A), n pieces of reference text data TD (n is an integer equal to or greater than 2) are ref Illustrated below is a Data TD respectively ref It is written as (x) (x is an integer greater than or equal to 1 and less than or equal to n).

[0080] In this embodiment, an example is shown in which the object of the search is a document related to intellectual property. T.D. ref is text data of a document related to intellectual property.

[0081] As described above, the reference text data input in step S1 is the data to be searched. The reference text data entered in step S1 is the document related to intellectual property. The text data input in step S1 is preferably, but is not limited to, the text data. In some cases, part of the reference text data is the data to be searched. The reference text data entered in step S1 is text data from patent documents and papers. The image may be only the text data of the patent document.

[0082] Here, the document related to intellectual property specifically includes patent documents (publication patent publications, patent publications, etc.) These include publications such as patent gazettes, design gazettes, and papers. The use of publications issued in countries around the world as intellectual property documents is not limited to those published in Japan. It is possible.

[0083] The specification, claims, and abstract contained in a patent document may each be used in whole or in part. Reference text data TD ref For example, the implementation of a particular invention The embodiments, examples or claims for the purpose of the present invention are referred to in the reference text data TD ref It can also be used as Similarly, texts contained in other publications, such as papers, may be used in whole or in part as references. Chapter Data TD ref It can be used as:

[0084] Documents related to intellectual property are not limited to publications. For example, users of document search systems or Document files owned by user organizations are also included as reference text data TD ref Can be used as can.

[0085] In addition, intellectual property documents include documents describing inventions, devices, or designs, or industrial products. Examples include sentences that clarify the above.

[0086] Reference Text Data TD ref For example, patent documents of a particular applicant or a particular technical field The present invention may have patent documents in the field.

[0087] Reference Text Data TD ref Not only the description of the intellectual property itself (e.g., specification, etc.), , may contain various information related to the intellectual property (e.g., bibliographic information, etc.). Such information may include, for example, the patent applicant, the technical field, the application number, the publication number, and the status. (Pending, registered, withdrawn, etc.)

[0088] Reference Text Data TD ref It is preferable that the date information related to the intellectual property is included. For example, if the intellectual property is a patent document, the information may include the application date, publication date, and registration date. For example, if the intellectual property is technical information of an industrial product, the release date may be included.

[0089] In this way, the reference text data TD ref has various information related to intellectual property With the document retrieval system, various search scopes can be selected.

[0090] For example, the document search system of this embodiment can be used to search for patents related to or similar to inventions that have not yet been filed. You can search patent documents, papers, or industrial products. This allows you to find inventions before filing. By identifying and reviewing the relevant prior art, This will strengthen the clarity of your invention and turn it into a strong patent that will be difficult for other companies to circumvent.

[0091] In addition, for example, the document search system of this embodiment can be used to search for documents related to industrial products that have not yet been released. can search for similar patent documents, papers, or industrial products. T.D. ref If a company has its own patent documents, the technology related to the industrial product before its release is sufficiently developed within the company. You can check whether a patent application has been filed. Or, you can use the reference text data TD ref but If you have information on other companies' intellectual property, you may be able to determine whether your pre-release industrial products infringe on the intellectual property rights of other companies. It is possible to check whether the relevant prior art is available or not. By reconsidering the technology, we can discover new inventions and develop strong patents that contribute to our company's business. In addition, the inspection can be carried out not only on industrial products before they are released, but also on industrial products after they are released. You can search for it.

[0092] In addition, for example, the document search system of this embodiment can be used to search for documents related to or similar to a specific patent. You can search for patents, papers, or industrial products that are related to the patent. By using this as a basis for the search, we can easily and accurately investigate whether the patent in question contains grounds for invalidation. It is possible.

[0093] [Step S2] Next, multiple reference text data TD ref Perform morphological analysis of each and analyze the sentence for reference Data AD ref (FIGS. 2 and 3(A)).

[0094] The processing unit 103 generates n pieces of reference text data TD ref Perform morphological analysis on each of them, and Reference text analysis data AD ref (Reference text analysis data AD, respectively ref (x) (x is an integer between 1 and n) For example, re f By performing morphological analysis of (n), reference sentence analysis data AD ref Generate (n) .

[0095] In morphological analysis, text written in natural language is divided into morphemes (the smallest units that have meaning in language). This allows us to distinguish the parts of speech of morphemes, etc. Data TD ref Only nouns can be extracted from

[0096] In FIG. 3(A), the reference text data TD to be input is ref (1) contains a long sentence and is output as Reference text analysis data AD ref In (1), the sentence is divided into several words.

[0097] In FIG. 3(A), the generated reference text analysis data AD ref is output to the outside of the processing unit 103. For example, the processing unit 103 stores a plurality of reference sentence analysis data in the database 107. Data AD ref In addition, the processing unit 103 can provide a plurality of reference sentence analyses. Data AD ref It is also possible to generate a corpus that compiles all of the above into a single data set and output it.

[0098] In addition to Japanese text, texts in various languages ​​(e.g. English, Chinese, Korean, etc.) are also supported. It is possible to perform chapter analysis. Various methods can be applied to analyze text depending on the language. .

[0099] [Step S3] After step S2, a plurality of reference sentence analysis data AD refCalculate the IDF of words in The IDF data ID is generated (FIGS. 2 and 3(B)). It is preferable to do so.

[0100] Reference Text Analysis Data AD ref is transmitted from the database 107 via the transmission line 102 as follows: The signal is supplied to the processing unit 103 .

[0101] The processing unit 103 generates n pieces of reference text analysis data AD ref Calculate the IDF of words in By standardizing the data, an IDF data ID is generated.

[0102] As shown in Figure 3(B), the IDF data ID is a word and a standardized IDF Includes.

[0103] The IDF(t) of a certain word t is calculated by normalizing the idf(t) in equation (1). The normalization method is not particularly limited. For example, idf(t) can be normalized by the formula (2). In formula (1), N is the total number of documents (reference text analysis data AD ref The number of n and above df(t) is the number of documents in which a certain word t appears (reference text analysis data AD r ef In equation (2), the idf MAX Reference text analysis data AD re f is the maximum value of idf(t) of words in MIN is a reference text analysis data AD ref is the minimum value of idf(t) of the words contained in

[0104]

number

[0105] For example, the normalized IDF of word A is 0.868, and the normalized IDF of word B is The normalized IDF for Word C is 0.642. Therefore, Word A has a higher IDF than Word B and Word C. Reference Text Analysis Data AD ref It can be said that these are characteristic words that rarely appear in the (B) shows an example of sorting words by alphabetical order, but this is not limited to this, and can be sorted by IDF, etc. You can also line up the words.

[0106] FIG. 3B shows an example in which the generated IDF data ID is output to the outside of the processing unit 103. For example, the processing unit 103 transmits the IDF data I to the database 107 via the transmission line 102. D can be supplied.

[0107] [Step S4] After step S2, a plurality of reference sentence analysis data AD ref Distributed representations of words in A vector is generated and vector data VD is generated (FIGS. 2 and 3(C)).

[0108] Note that step S3 and step S4 may be performed in either order, or may be performed in parallel. .

[0109] The processing unit 103 generates n pieces of reference text analysis data AD ref A distributed representation vector of words in A torque is generated and vector data VD is generated.

[0110] The word embedding vector is a vector that represents a word as a unique representation of each word. It is a vector that expresses a quantified continuous value for each characteristic element (dimension). The vectors will also be closer together.

[0111] As shown in FIG. 3C, the processing unit 103 uses a neural network NN to It is preferable to generate a distributed representation vector of

[0112] Here, we will explain how to generate distributed representation vectors of words using a neural network (NN). An example of this is explained below. The neural network NN is trained using supervised learning. The neural network is created by inputting a word to the input layer and outputting the surrounding words of the word. The hidden layer is 10 It is preferable to have a relatively low-dimensional vector with dimensions of 1000 or more. This vector is the distributed representation vector of the word.

[0113] Distributed representations of words are, for example, found in the open-source Word2ve algorithm. c. Word2vec uses the same context to distinguish words from each other. Based on the hypothesis that words have meaning, vectorize words including their features and semantic structure. .

[0114] In word vectorization, we generate distributed representation vectors of words to perform vector-to-vector deduction. It is possible to calculate the similarity and distance between words by using the same algorithm. When the two vectors are close to each other, the two vectors are said to be highly related. In this case, the two vectors can be said to be highly related.

[0115] In addition, while one-hot representation assigns one dimension to one word, distributed representation assigns words Since it can be expressed as a low-dimensional real-valued vector, it can be expressed with a small number of dimensions even if the vocabulary size increases. Therefore, even if the corpus contains a large number of words, the amount of calculation does not increase. It is possible to process huge amounts of data in a short amount of time.

[0116] As shown in FIG. 3(C), the vector data VD consists of words and vectors. For example, the Vector for Word A is (0.12, 0.90, 0.3 2, ), and the vector for word B is (0.88, 0.10, 0.29, ), and the vector for word C is (0.23, 0.56, 0.47, . ·).

[0117] FIG. 3C shows an example in which the generated vector data VD is output to the outside of the processing unit 103. For example, the processing unit 103 may provide the vector data VD to the database 107. can.

[0118] The above steps are carried out in advance to create reference text analysis data AD ref , IDF data ID, and By generating the vector data VD, the document can be searched for by using the document search system 100. It can be searched.

[0119] In this embodiment, the document retrieval system 100 is used to retrieve the reference text analysis data AD ref We have shown an example of generating IDF data ID and vector data VD, but document search Outside the system 100, reference text analysis data AD ref , IDF data ID, and In this case, at least one of the vector data VD generated externally may be generated. The data is input to the input unit 101 of the document search system 100 and stored in the database 107. This enables searches using the document search system 100.

[0120] Next, a search method using the document search system 100 will be described. 9 shows a flowchart, and FIG. 6, FIG. 7(A), and FIG. The diagrams show the steps shown in FIG. 6, FIG. 7(A), and FIG. The illustration is merely an example and is not limiting.

[0121] [Step S11] First, text data TD is input to the input unit 101 (FIGS. 4, 5, and 6(A)).

[0122] In the document retrieval method of this embodiment, reference text data TD ref From the text data T It is possible to search for text data related to or similar to D.

[0123] Text data TD is input to an input unit 101 from outside the document search system 100. The text data TD is supplied from an input unit 101 to a processing unit 103 via a transmission line 102. Alternatively, the document data TD is transmitted to the storage unit 105 or the data storage unit 106 via the transmission path 102. The data is stored in the base 107, and the data is read from the storage unit 105 or the database 107 via the transmission line 102. The signal may be supplied to the processing unit 103 via the

[0124] In this embodiment, an example is shown in which the object of the search is a document related to intellectual property. The text data TD is , data of documents relating to intellectual property. Examples of documents relating to intellectual property are as described above.

[0125] The text data TD may be, for example, inventions, devices or designs before application, industrial products before release, technology It may contain information or text explaining technical ideas.

[0126] In particular, as text data TD, the scope of claims, abstract, or text explaining the outline of the invention A chapter or the like can be used suitably. The amount of text in this case is relatively small (compared to the full specification). The text data TD is a set of characteristic keywords that are included in the text data TD. The document search system of the present embodiment is preferable because it is easy to extract characteristic keywords. Therefore, even if the text data TD has a small vocabulary, it is possible to extract related words with high accuracy. A search can be performed.

[0127] [Step S12] Next, morphological analysis of the text data TD is performed to generate text analysis data AD (Fig. 4, Fig. 5 , and Figure 6(A)).

[0128] The processing unit 103 performs a morphological analysis of the text data TD to generate text analysis data AD.

[0129] In FIG. 6(A), the input text data TD includes a long sentence, and the output text analysis data A In D, the sentence is split into multiple words.

[0130] FIG. 6A shows an example in which the generated text analysis data AD is output to the outside of the processing unit 103. For example, the processing unit 103 stores text analysis data in the storage unit 105 or the database 107. AD can be provided.

[0131] [Steps S13 and S33] Next, the text analysis data AD is matched with the IDF data ID to generate the keyword data KD (FIGS. 4, 5, and 6(B)).

[0132] The text analysis data AD is transmitted from the storage unit 105 or the database 107 via the transmission path 102. The IDF data ID is sent from the database 107 to the processor 103. The signal is supplied to a processing unit 103 via a line 102 .

[0133] The processing unit 103 compares the text analysis data AD with the IDF data ID, and obtains the text analysis data A. Keyword data KD is generated by sorting the words contained in D in order of highest IDF.

[0134] The keyword data KD includes keywords KW and their IDFs. Here is an example using IDF.

[0135] Keywords KW can be said to be characteristic words contained in the text analysis data AD. For example, the keyword KW is a word that is included in both the text analysis data AD and the IDF data ID. It may be all words, or words with IDF equal to or greater than a given value. Depending on the amount of text in the text data TD, the keyword KW It is preferable to set the extraction criteria as follows. For example, the number of keywords KW should be between 2 and 100. Preferably, the number is 1 or less, and more preferably, 5 to 30.

[0136] Here, the normalized IDF of the keyword KW is used later as the reference text analysis data AD ref This corresponds to the weight of the keyword KW used when assigning a score to.

[0137] In Figure 6(B), the weight of Word D is 0.873, the weight of Word A is 0.868, An example is shown where word E has a weight of 0.867.

[0138] FIG. 6B shows an example in which the keyword data KD is output to the outside of the processing unit 103. For example, the processing unit 103 may transmit the information to the storage unit 105 or the database 107 via the transmission line 102. , keyword data KD can be provided.

[0139] [Steps S14 and S34] Next, using the text analysis data AD or keyword data KD and the vector data VD, Then, related word data RD is generated (FIGS. 4, 5, and 6(C)).

[0140] The text analysis data AD or the keyword data KD is stored in the storage unit 105 or the database 106. 07 to a processing unit 103 via a transmission line 102. The data is supplied from the database 107 to the processing unit 103 via the transmission line 102 .

[0141] In step S14, the processing unit 103 calculates the distributed representation vector of the keyword KW and the reference Text Analysis Data AD ref The similarity between the distributed representation vector of the words in and or Based on the closeness of the distance, related words RW of the keyword KW are extracted. The related word data RD is generated by arranging the words in order of similarity or distance. In general, for one keyword, 1 to 10 related words should be extracted. It is preferable to extract 2 to 5 related words. The word may be one whose similarity is equal to or greater than a predetermined value, or one whose distance is equal to or less than a predetermined value. The word may be a predetermined number of words with high similarity, or a predetermined number of words with close distance. By using keywords, synonyms, synonyms, antonyms, hypernyms, and hyponyms can be used. Since the number of related words varies depending on the keyword KW, the number of related words RW may differ. Reference text analysis data AD ref Among the words contained in By extracting the above, reference text analysis data AD ref The keyword KW is displayed in a unique way. Even if the expression is written in a different language, it is possible to extract the expression as a related word RW. This is preferable because it reduces the chance of missing a search result due to variations in notation.

[0142] In step S34, the processing unit 103 calculates the distributed representation of the words included in the text analysis data AD. Vector and reference text analysis data AD ref The distributed representation vector of the words in and Based on the similarity or distance, related words RW are extracted. Otherwise, This is the same as step S14.

[0143] The similarity between two vectors can be measured using cosine similarity, covariance, unbiased covariance, and Pearson product-moment correlation. It can be obtained by using a coefficient, etc. In particular, it is preferable to use the cosine similarity.

[0144] The distance between two vectors is Euclidean distance, standard (average) Euclidean distance, Using Mahalanobis distance, Manhattan distance, Chebyshev distance, Minkowski distance, etc. can be found by

[0145] The related word data RD includes related words RW and their degrees of relevance RS.

[0146] Related words RW are words contained in the text analysis data AD or words related to keywords KW. It can be said that.

[0147] The relevance score RS is a value that indicates the level of similarity or the closeness of distance, or a standardized version of these. The relevance score RS is the value obtained by the reference sentence analysis data AD. ref When assigning points to The weight of related words is calculated by using the standardized I The product of DF and the relevance score RS of the related word RW corresponds to the weight of the related word.

[0148] In Figure 6(C), the related words RW of word D are sorted into word X in descending order of relevance RS. (Relevance RS is 0.999), Word Y (Relevance RS is 0.901), Word Z The following is an example of the extracted result (with relevance RS of 0.712).

[0149] FIG. 6C shows an example in which the related word data RD is output to the outside of the processing unit 103. For example, The processing unit 103 transmits the related information to the storage unit 105 or the database 107 via the transmission path 102. Collocation data RD can be provided.

[0150] In addition, when extracting related words using keyword data KD, as shown in FIG. After step S13, step S14 is performed. On the other hand, related words are extracted using the text analysis data AD. As shown in FIG. 5, when the Alternatively, they may be performed in parallel.

[0151] Furthermore, it is determined whether the related word RW is a word included in the concept dictionary, and The method may further include a step of determining a weight for the related word RW when the related word RW is included in the concept dictionary. Therefore, the related word RW is considered to have a high degree of relevance to the keyword KW. If the word is included, the weight of the related word RW is set to be larger than if it is not included. For example, a value indicating the degree of similarity or the closeness of distance may be set according to the result of the determination. The value added or subtracted may be used as the weight of the related word RW. The degree of similarity between the word RW in the concept dictionary and the word RW not in the concept dictionary is A predetermined value may be used as the weight of the related word RW regardless of the distance or the distance between the related words. For example, If the related word RW is included in the concept dictionary, the weight of the related word RW is set to the same weight as the keyword KW. It can also be.

[0152] [Step S15] Next, reference text analysis data AD ref Keywords that match the words in KW or Based on the weight of related words RW, reference text analysis data AD ref (Figure 4) , Figure 5, and Figure 7(A).

[0153] Reference Text Analysis Data AD ref is transmitted from the database 107 via the transmission line 102 as follows: The keyword data KD and the related word data RD are supplied to the processing unit 103. 5 or database 107 to the processing unit 103 via the transmission line 102. The processing unit 103 transmits the result of the scoring via the transmission line 102. The information can be provided to the storage unit 105 or the database 107 .

[0154] First, a concrete example of scoring will be described with reference to FIG. 7(B). In FIG. 7(B), three types of scoring are Here is an example of using a keyword KW and four related words RW for each keyword KW. .

[0155] The denominator of the score is the sum of the weights of the keywords KW. In the case of Figure 7(B), it is 0.9 + 0.9 +0.8=2.6.

[0156] The numerator of the score is the reference text analysis data AD ref Keywords K that match words contained in The sum of the weights of W or related words RW. In the case of Figure 7(B), Word D, Word The sum of the weights of word e and word f is 1.95.

[0157] From this, the score can be calculated as 1.95 / 2.6=0.75 (75%). .

[0158] Step S15 will be described in detail with reference to FIG. 8. As shown in FIG. 8, step S15 includes the following steps: The steps include steps S21 to S27. The number of keywords KW is p, and each keyword If the number of related words RW of a word KW is q, then x is an integer between 1 and p, and y is 1. Represents an integer greater than or equal to q, inclusive.

[0159] [Step S21] First, the unscored reference text analysis data AD ref Select one item.

[0160] [Step S22] Next, reference text analysis data AD ref In this case, the keyword KW x Determine if it will be a hit If there is a hit, proceed to step S25. If there is no hit, proceed to step S23. nothing.

[0161] [Step S23] Next, reference text analysis data AD ref In this case, the keyword KW x Related words for RW xy but It is judged whether there is a hit. If there is a hit, proceed to step S25. If there is no hit, Proceed to step S24.

[0162] [Step S24] Next, it is determined whether all related words RW of the keyword KWx have been searched. If so, the step If no keyword has been searched, the process proceeds to step S23. W x There are two related words RW, and the related word RW x1 Determine if it will be a hit If so, return to step S23 and x2 Determine whether it is a hit.

[0163] [Step S25] In step S25, the weight corresponding to the hit word is added to the score. If 2 hits, the keyword KW x The IDF of is added to the score. If there is a hit, the keyword KW x IDF and related words RW xy The product of the relevance RS of and It is added to the score. In the above scoring example, it is added to the numerator of the score.

[0164] [Step S26] Next, it is determined whether or not all keywords KW have been searched. If so, the process proceeds to step S27. If no search has been performed, the process proceeds to step S22. For example, if the keyword KW x There are two If it has been determined in step S22 whether the keyword KW1 is a hit, Go back to 22 and determine whether keyword KW2 is a hit.

[0165] [Step S27] Next, all reference text analysis data AD refDetermine whether you have scored all the points. If the attachment is complete, proceed to step S16. If not, proceed to step S21. nothing.

[0166] [Step S16] And the reference text analysis data AD ref Ranking is performed to generate ranking data LD. , and output (Figures 4, 5, and 7(A)).

[0167] The processing unit 103 transmits the ranking data LD to the storage unit 105 or The processing unit 103 can provide the ranking data to the database 107. The LD can be supplied to the output section 109 via the transmission line 102. The input unit 109 can supply the ranking data LD to the outside of the document search system 100. can.

[0168] Ranking data LD is the ranking (Lank), and reference text data TD ref Information (Name It can include ID numbers (Doc), scores, etc. Reference text data TD for database 107 etc. ref If saved, ranking Data LD is reference text data TD ref It is preferable to include the file path to the This allows users to easily access the desired document from the ranking data LD. can be done.

[0169] Reference Text Analysis Data AD ref The higher the score, the better the reference text analysis data AD r ef Reference text data TD corresponding to ref is related or similar to the text data TD It can be said that.

[0170] In the example shown in FIG. 7A, n pieces of reference sentence data TD are selected from the ranking data LD. re f Among them, the data most related or similar to the text data TD is the reference text data T D ref (7) and the second most related or similar data is the reference text data T D ref (4) The third related or similar data is the reference text data T D ref It turns out to be (13).

[0171] As described above, a search can be performed using the document search system 100.

[0172] In addition, the keywords KW, the weights (IDFs) of the keywords KW, and the relations The weights (IDF×RS) of the collocation RW and related term RW are manually edited, and then step You may proceed to S15.

[0173] FIG. 9 shows a flowchart including the editing steps, and FIG. 10 shows a schematic diagram of the steps shown in FIG. The data illustrated in FIG. 10 is merely an example, and the present invention is not limited to this.

[0174] [Step S41] After step S14, in step S15, the keyword data KD and related word data to be used are A list of RDs is output (FIGS. 9 and 10(A)).

[0175] From Figure 10(A), the keywords KW are Word D, Word A, and Word E. The weights (IDF) of each are 0.9, 0.9, and 0.8. Light.

[0176] The related words RW of Word D are Word X, Word Y, Word Z, and Word a is listed, and the weights (IDF×RS) are 0.9, 0.8, and 0 We can see that the values ​​are .6 and 0.5.

[0177] The related words RW of Word A are Word b, Word c, Word d, and Word e is listed, and the weights (IDF × RS) are 0.5, 0.5, and 0 We can see that the values ​​are .45 and 0.3.

[0178] The related words RW of Word E are Word f, Word g, Word h, and Word i is listed, and the weights (IDF×RS) are 0.75, 0.75 , 0.75, 0.75.

[0179] [Step S42] Next, the keyword data KD and related word data RD are edited (FIGS. 9 and 10(B)). .

[0180] FIG. 10B shows an example where three edits have been made. Specifically, Word A and its Removal of related words RW and changing word a (weight 0.5) to word x (weight 0.8) and a change in the weight of word f (from 0.75 to 0.8).

[0181] In this way, the user can edit at least one of the words and weights to improve search accuracy. It may also be possible to do so.

[0182] Then, in step S15, the edited keyword data KD and related word data RD are Reference text analysis data AD refIn step S16, the ranking is determined. The data LD is generated and output.

[0183] [Step S43] Next, check whether the ranking data LD is the expected result (Figure 9). If there is, the search ends. If the expected result is not obtained, the process returns to step S41. Then, a list of the edited keyword data KD and related word data RD is outputted, and step S4 You can edit it again in step 2.

[0184] In addition, editing of words and weights is not limited to manual work, but can be done using dictionary data and natural language processing. This may be done automatically using the analysis data obtained by editing. This can be done.

[0185] <3. Document Search System Configuration Example 2> Next, a document retrieval system 150 shown in FIG. 11 will be described.

[0186] FIG. 11 shows a block diagram of the document retrieval system 150. The document retrieval system 150 includes The system includes a server 151 and a terminal 152 (such as a personal computer).

[0187] The server 151 includes a communication unit 161a, a transmission path 162, a processing unit 163a, and a database 1 67. Although not shown in FIG. 11, the server 151 further includes a storage unit, an input / output unit, etc. etc.

[0188] The terminal 152 includes a communication unit 161b, a transmission path 168, a processing unit 163b, a storage unit 165, and an input The terminal 152 further includes a database (not shown in FIG. 11). and the like.

[0189] A user of the document retrieval system 150 inputs document data TD from a terminal 152 to a server 151. The text data TD is transmitted from the communication section 161b to the communication section 161a.

[0190] The text data TD received by the communication unit 161a is transmitted to the database 16 via the transmission path 162. 7 or a storage unit (not shown). Alternatively, the text data TD is stored in the communication unit 161 a directly to the processing unit 163a.

[0191] The various processes described in <2. Document search method> above are performed by the processing unit 163a. Since this process requires high processing power, it is performed by the processing unit 163a of the server 151. It is preferable.

[0192] Then, the processing unit 163a generates ranking data LD. D is stored in a database 167 or a storage unit (not shown) via a transmission line 162. Alternatively, the ranking data LD is directly transmitted from the processing unit 163a to the communication unit 161a. After that, the server 151 transmits the ranking data LD to the terminal 152. The ranking data LD is transmitted from the communication unit 161a to the communication unit 161b. .

[0193] [I / O section 169] Data is supplied to the input / output unit 169 from outside the document search system 150. 169 has a function of supplying data to the outside of the document search system 150. As in the search system 100, the input section and the output section may be separate.

[0194] [Transmission path 162 and transmission path 168] The transmission path 162 and the transmission path 168 have a function of transmitting data. Data is transmitted and received between the management unit 163a and the database 167 via a transmission line 162. The communication unit 161b, the processing unit 163b, the storage unit 165, and the input / output unit 16 Data transmission and reception between the devices 9 can be performed via a transmission line 168.

[0195] [Processing unit 163a and processing unit 163b] The processing unit 163a receives data from the communication unit 161a and the database 167. The processing unit 163b has a function of performing calculations, inferences, etc. using the information. A function of performing calculations using data supplied from the input / output unit 169 and the like. The processing unit 163a and the processing unit 163b can refer to the description of the processing unit 103. The processing unit 163a can perform various processes described in <2. Document search method> above. Therefore, it is preferable that the processing unit 163a has a higher processing capacity than the processing unit 163b. stomach.

[0196] [Storage section 165] The storage unit 165 has a function of storing the program executed by the processing unit 163b. The storage unit 165 stores the calculation results generated by the processing unit 163b and the data input to the communication unit 161b. The input / output unit 169 has a function of storing data input thereto, and data input to the input / output unit 169.

[0197] [Database 167] Database 167 is a reference text analysis data AD ref , IDF data ID, and vector The database 167 also has a function of storing the torque data VD. and a function for storing the calculation results generated by the communication unit 161a and the data input to the communication unit 161a. Alternatively, the server 151 may have a storage unit separate from the database 167. The storage unit stores the calculation results generated by the processing unit 163a and the data input to the communication unit 161a. The device may have a function of storing data, etc.

[0198] [Communication Unit 161a and Communication Unit 161b] Using the communication units 161a and 161b, data is exchanged between the server 151 and the terminal 152. The communication unit 161a and the communication unit 161b can be a hub, a router, or the like. Data can be sent and received using wired or wireless (e.g. For example, radio waves, infrared rays, etc. may be used.

[0199] As described above, in the document search system of the present embodiment, documents prepared in advance are used as the object of search. As an example, it is possible to search for documents related or similar to the input document. Users do not need to select keywords, and the amount of text data is larger than that of keywords. Since it is possible to search using The document search system of this embodiment can search for documents with high accuracy. Related words are extracted from a pre-prepared document, and the unique expressions contained in the document are used It is also possible to extract related words, thereby reducing missed searches. A search system can output search results ranked by relevance or similarity. This makes it easier for users to find the documents they need from the search results, and also reduces the chance of them being overlooked. It becomes hard.

[0200] This embodiment mode can be appropriately combined with other embodiment modes. In the case where a plurality of configuration examples are shown in one embodiment, the configuration examples may be appropriately combined. It is possible to do this.

[0201] (Embodiment 2) In this embodiment, a configuration example of a semiconductor device that can be used in a neural network is shown. We will explain about this.

[0202] The semiconductor device of this embodiment is used, for example, in a processing unit of a document search system according to one aspect of the present invention. There can be.

[0203] As shown in Fig. 12(A), the neural network NN has an input layer IL, an output layer OL, and a middle layer OL. It can be composed of an input layer IL, an output layer OL, and an intermediate layer H Each L has one or more neurons (units). It may be a single layer or two or more layers. The network can also be called a DNN (deep neural network), Learning using neural networks can also be called deep learning.

[0204] Input data is input to each neuron in the input layer IL, and previous data is input to each neuron in the hidden layer HL. The output signal of the neurons in the layer OL or the next layer is input, and each neuron in the output layer OL receives the output signal of the neurons in the previous layer. The output signal of the neuron is input. Each neuron is connected to all the neurons in the previous and next layers. It may be connected to all neurons (full connection) or it may be connected to only a portion of neurons.

[0205] FIG. 12(B) shows an example of a computation by a neuron. Here, we consider a neuron N and The figure shows two neurons in the front layer that output signals to neuron N. Neuron N has a The output of the neuron in the previous layer, x1, and the output of the neuron in the previous layer, x2, are input. In Ron N, the multiplication result of output x1 and weight w1 (x1w1) and the multiplication result of output x2 and weight w2 After the sum of the calculation results (x2w2) x1w1+x2w2 is calculated, the bias b is applied if necessary. are added together to obtain the value a=x1w1+x2w2+b. The value a is then applied to the activation function h. Thus, the neuron N outputs the output signal y=h(a).

[0206] In this way, the operation of a neuron involves adding up the products of the outputs of the neurons in the previous layer and the weights. This multiplication and addition operation is called multiplication and addition (x1w1+x2w2 above). This may be done on software using a program, or on hardware. When the multiplication and accumulation is performed by hardware, a multiplication and accumulation circuit can be used. The multiplication and accumulation circuit may be a digital circuit or an analog circuit. When an analog circuit is used for the sum-of-products operation circuit, the circuit scale of the sum-of-products operation circuit can be reduced. Alternatively, the number of times memory is accessed can be reduced, thereby improving processing speed and reducing power consumption. This can be done.

[0207] The multiply-and-accumulate circuit is a transistor that contains silicon (such as single crystal silicon) in the channel formation region. The channel forming region may be made of a gold. Transistors that contain oxide semiconductors (also called OS transistors) In particular, since the off-state current of an OS transistor is extremely small, It is suitable as a transistor that constitutes the memory of an addition circuit. A multiply-and-accumulate circuit may be configured using both the OS transistor and the MOS transistor. A configuration example of a semiconductor device having the above functions will be described.

[0208] <Example of semiconductor device configuration> FIG. 13 shows an example of the configuration of a semiconductor device MAC having a function of performing neural network calculations. The semiconductor device MAC stores first data corresponding to the connection strength (weight) between neurons. The first data corresponds to the input data, and the second data corresponds to the input data. The first data and the second data are analog data or multi-value digital data (discrete The semiconductor device MAC can be a multiplication and accumulation operation. It has the function of transforming data using an activation function.

[0209] The semiconductor device MAC includes a cell array CA, a current source circuit CS, a current mirror circuit CM, and a circuit WDD, circuit WLD, circuit CLD, offset circuit OFST, and activation function circuit ACT Has V.

[0210] The cell array CA has a plurality of memory cells MC and a plurality of memory cells MCref. In the cell array CA shown in FIG. 13, memory cells MC (MC [1,1] to MC[m,n]), and m memory cells MCref (MCref[1] to The memory cell MC stores the first data. The memory cell MCref stores reference data used in the multiplication and accumulation operations. The reference data can be analog data or multi-value digital data. This can be used as data.

[0211] A memory cell MC[i,j] (i is an integer between 1 and m, and j is an integer between 1 and n) is arranged as It is connected to line WL[i], wiring RW[i], wiring WD[j], and wiring BL[j]. In addition, the memory cell MCref[i] is connected to the wiring WL[i], the wiring RW[i], the wiring WDr ef and the wiring BLref. Here, the memory cell MC[i,j] and the wiring BL [j] is the current flowing between MC[i,j] and the memory cell MCref[i] and the wiring The current flowing between BLref and I MCref[i] It is written as follows.

[0212] A specific example of the configuration of the memory cell MC and the memory cell MCref is shown in FIG. As representative examples, memory cells MC[1,1], MC[2,1] and memory cell MCref[ 1] and MCref[2] are shown, but other memory cells MC and memory cells MCref The memory cell MC and the memory cell MCref can be configured in the same manner. Each of them has a transistor Tr11, a transistor Tr12, and a capacitance element C11. The transistors Tr11 and Tr12 are n-channel transistors. This section explains what happens when

[0213] In the memory cell MC, the gate of the transistor Tr11 is connected to the line WL, and the source One of the drains is connected to the gate of the transistor Tr12 and the first potential of the capacitance element C11. The other of the source and drain is connected to the wiring WD. Either the source or drain of Tr12 is connected to the wiring BL. The other electrode of the capacitance element C11 is connected to the wiring VR. The wiring VR has a function of supplying a predetermined potential. A case where a low power supply potential (such as a ground potential) is supplied from the wiring VR will be described below.

[0214] One of the source or drain of the transistor Tr11, the gate of the transistor Tr12, A node connected to the first electrode of the capacitor C11 is referred to as a node NM. The nodes NM of the resellers MC[1,1] and MC[2,1] are respectively the nodes NM[1,1] , denoted as NM[2,1].

[0215] The memory cell MCref has a similar configuration to the memory cell MC. Cref is connected to the wiring WDref instead of the wiring WD, and the wiring BL is connected to the wiring BL ref. Also, in memory cells MCref[1] and MCref[2], The source or drain of the transistor Tr11 and the gate of the transistor Tr12 are connected to The node connected to the first electrode of the capacitance element C11 and the node connected to the first electrode of the capacitance element C12 are referred to as node NMref. It is abbreviated as [1] and NMref[2].

[0216] The nodes NM and NMref are the storage nodes of the memory cells MC and MCref, respectively. The node NM holds the first data, and the node NMref holds the second data. The reference data is stored in the memory cell MC[1,1] from the wiring BL[1], The transistor Tr12 in [2,1] has a current I MC[1,1] , I MC[2, 1] Also, the memory cells MCref[1] and MCref[2 ] transistor Tr12 has a current I MCref[1] , IMCref[2] flows.

[0217] The transistor Tr11 has the function of holding the potential of the node NM or the node NMref. Therefore, it is preferable that the off-state current of the transistor Tr11 is small. It is preferable to use an OS transistor with extremely low off-state current as the transistor Tr11. This makes it possible to suppress the fluctuation of the potential of the node NM or the node NMref, The calculation accuracy can be improved. This makes it possible to reduce the frequency of refreshing operations, thereby reducing power consumption. Cut.

[0218] The transistor Tr12 is not particularly limited and may be, for example, a Si transistor or an OS transistor. When an OS transistor is used for the transistor Tr12, Transistor Tr12 can be manufactured using the same manufacturing equipment as transistor Tr11. This makes it possible to reduce manufacturing costs. It may be either a p-channel type or a n-channel type.

[0219] The current source circuit CS is connected to the lines BL[1] to BL[n] and the line BLref. The current source circuit CS supplies current to the wirings BL[1] to BL[n] and the wiring BLref. Note that the current value supplied to the wirings BL[1] to BL[n] and the current value supplied to the wirings BLr The current value supplied to ef may be different. Here, the current source circuit CS is connected to the wiring BL The current supplied to BL[1] through BL[n] is I C , the current source circuit CS is supplied to the wiring BLref. The supplied current is I CrefIt is written as follows.

[0220] The current mirror circuit CM has lines IL[1] to IL[n] and a line ILref. The wirings IL[1] to IL[n] are connected to the wirings BL[1] to BL[n], respectively. The wiring ILref is connected to the wiring BLref. The connection points of L[n] and wiring BL[1] to BL[n] are nodes NP[1] to NP[n] The connection point between the wire ILref and the wire BLref is represented as node NPref. Please write down.

[0221] The current mirror circuit CM generates a current I CM Wiring ILre f and this current I CM Also has the function of passing the signal to the wiring IL[1] to IL[n]. In Figure 13, a current I CM is discharged, and wiring BL[ A current I flows from BL[1] to BL[n] to wiring IL[1] to IL[n]. CM Here is an example of the discharge of In addition, the current mirror circuit CM is connected to the sensor through the wiring BL[1] to BL[n]. The current flowing through the lumen array CA is I B [1]~I B It is written as [n]. Also, the current The current flowing from the control circuit CM to the cell array CA via the wiring BLref is I Bref and Write.

[0222] The circuit WDD is connected to the wirings WD[1] to WD[n] and the wiring WDref. The line WDD supplies a potential corresponding to the first data stored in the memory cell MC to the line WD[1 The circuit WDD has a function of supplying data to the memory cells MCref The circuit has a function of supplying a potential corresponding to the reference data stored in the circuit to the wiring WDref. The circuit WLD is connected to the wirings WL[1] to WL[m]. A signal for selecting the memory cell MC or memory cell MCref to be written is arranged The circuit CLD has a function of supplying power to the wirings RW[1] to WL[m]. The circuit CLD is connected to the wiring R It has the function of supplying to W[1] to RW[m].

[0223] The offset circuit OFST includes wirings BL[1] to BL[n] and wirings OL[1] to OL The offset circuit OFST is connected to the wirings BL[1] to BL[n]. The amount of current flowing from the offset circuit OFST to the wirings BL[1] to BL[n It also has the function of detecting the amount of change in the current flowing from the offset The offset circuit OFST has a function of outputting the detection result to the wirings OL[1] to OL[n]. The offset circuit OFST outputs a current corresponding to the detection result to the wiring OL. Alternatively, a current corresponding to the detection result may be converted into a voltage and output to the wiring OL. The current flowing between the line CA and the offset circuit OFST is I α [1]~I α [n] and Please write down.

[0224] A configuration example of the offset circuit OFST is shown in FIG. The circuit OC[1] to OC[n] includes circuits OC[1] to OC[n]. The transistors Tr21, Tr22, and Tr23 are capacitors. The connections of the elements are as shown in FIG. A node connected to the first electrode of the capacitance element C21 and the first terminal of the resistance element R1 is defined as The node Na is connected to the second electrode of the capacitance element C21 and the source of the transistor Tr21. Alternatively, a node connected to one of the drains and the gate of the transistor Tr22 is connected to a node Let Nb.

[0225] The wiring VrefL has a function of supplying a potential Vref, and the wiring VaL supplies a potential Va. The wiring VbL has a function of supplying a potential Vb. The wiring VSSL has a function of supplying a potential VSS. Now, let us consider the case where the potential VDD is the high power supply potential and the potential VSS is the low power supply potential. The wiring RST supplies a potential for controlling the conduction state of the transistor Tr21. The transistor Tr22, the transistor Tr23, the wiring VDDL, the wiring The line VSSL and the wiring VbL form a source follower circuit.

[0226] Next, an example of the operation of the circuits OC[1] to OC[n] will be described. An example of the operation of the circuit OC[1] will be described, but the circuits OC[2] to OC[n] also operate in the same manner. First, when a first current flows through the wiring BL[1], the potential of the node Na becomes The potential is determined according to the first current and the resistance value of the resistor R1. r21 is in an on state, and the potential Va is supplied to the node Nb. Then, the transistor T r21 is in the off state.

[0227] Next, when a second current flows through the wiring BL[1], the potential of the node Na increases by a factor of the second current and the resistance The potential changes according to the resistance value of element R1. At this time, transistor Tr21 is in the off state. Since node Nb is in a floating state, the potential of node Na changes. The potential of the node Nb changes due to capacitive coupling. Here, the change in the potential of the node Na is Δ V Na If the capacitance coupling coefficient is 1, the potential of node Nb is Va+ΔV Na It becomes. Then, the threshold voltage of the transistor Tr22 is V th Then, the potential from the wiring OL[1] Va+ΔV Na -V th is output, where Va=V th By doing so, the wiring Potential ΔV from L[1] Na can be output.

[0228] Potential ΔV Na is the change amount from the first current to the second current, the resistance value of the resistive element R1, and Here, the resistance value of the resistor element R1 and the potential Vref are known. Therefore, the potential ΔV Na From this, the amount of change in the current flowing through the wiring BL can be obtained.

[0229] The amount of current detected by the offset circuit OFST as described above and / or the change in current A signal corresponding to the amount of the activation is input to the activation function circuit ACTV via wiring OL[1] to OL[n]. is entered into

[0230] The activation function circuit ACTV includes wirings OL[1] to OL[n] and wirings NIL[1] to The activation function circuit ACTV is connected to the offset circuit OFST The calculation to transform the input signal from according to a predefined activation function is The activation function can be, for example, a sigmoid function, a tanh function, or a so You can use the ftmax function, ReLU function, threshold function, etc. The signal converted by the circuit ACTV is outputted as output data through wirings NIL[1] to NIL[ n].

[0231] <Example of operation of semiconductor device> Using the semiconductor device MAC, a multiplication and accumulation operation of first data and second data can be performed. An example of the operation of the semiconductor device MAC when performing a product-sum operation will be described below.

[0232] FIG. 16 shows a timing chart of an operation example of the semiconductor device MAC. In the diagram, wiring WL[1], wiring WL[2], wiring WD[1], wiring WDref, and node NM [1,1], node NM[2,1], node NMref[1], node NMref[2] , the transition of the potential of the wire RW[1] and the wire RW[2], and the current I B [1]-I α [1], and current I Bref The graph shows the transition of the current I B [1]-I α [1] is wiring BL It corresponds to the sum of the currents flowing from [1] to memory cells MC[1,1] and MC[2,1].

[0233] As a representative example, memory cells MC[1,1], MC[2,1] and The operation will be explained focusing on memory cells MCref[1] and MCref[2]. The memory cell MC and the memory cell MCref can be operated in the same manner.

[0234] [Storage of the first data] First, during the period from time T01 to time T02, the potential of the wiring WL[1] is at a high level (H igh), and the potential of the wiring WD[1] becomes V higher than the ground potential (GND). PR -V W[1, 1] The potential of the wiring WDref becomes V higher than the ground potential. PR It becomes a large electric potential In addition, the potential of the wiring RW[1] and the wiring RW[2] becomes the reference potential (REFP). Oh, potential V W[1,1] corresponds to the first data stored in the memory cell MC[1,1]. The potential V PR is the potential corresponding to the reference data. The transistor Tr11 in the rechargeable cell MC[1,1] and the memory cell MCref[1] When the power supply is turned on, the potential of the node NM[1,1] becomes V PR -V W[1,1] , node NMr The potential of ef[1] is V PR It becomes.

[0235] At this time, a current flows from the wiring BL[1] to the transistor Tr12 of the memory cell MC[1,1]. The current I MC[1,1],0 can be expressed as follows, where k is the number of transistors A constant determined by the channel length, channel width, mobility, and capacitance of the gate insulating film of Tr12 Also, V th is the threshold voltage of transistor Tr12.

[0236] I MC[1,1],0 =k(V PR -V W[1,1] -V th ) 2 (E1)

[0237] Also, current flows from the wiring BLref to the transistor Tr12 of the memory cell MCref[1]. current IMCref[1],0 can be expressed by the following formula:

[0238] I MCref[1],0 =k(V PR -V th ) 2 (E2)

[0239] Next, during the period from time T02 to time T03, the potential of the wiring WL[1] becomes low level (L ow). As a result, the memory cell MC[1,1] and the memory cell MCref[1] The transistor Tr11 in the The potential of ref[1] is maintained.

[0240] As described above, it is preferable to use an OS transistor as the transistor Tr11. This makes it possible to suppress the leakage current of the transistor Tr11, and the node NM[ 1,1] and the potential of node NMref[1] can be accurately maintained.

[0241] Next, during the period from time T03 to time T04, the potential of the line WL[2] becomes high level. The potential of the wire WD[1] is V higher than the ground potential. PR -V W[2,1] This results in a large potential. The potential of the wiring WDref is V higher than the ground potential. PR The potential V W[2, 1] is a potential corresponding to the first data stored in the memory cell MC[2,1]. As a result, the transistors of the memory cell MC[2,1] and the memory cell MCref[2] Tr11 turns on, and the potential of node NM[2,1] becomes V PR -V W[2,1] , The potential of node NMref[2] is V PR It becomes.

[0242] At this time, a current flows from the wiring BL[1] to the transistor Tr12 of the memory cell MC[2,1]. The current I MC[2,1],0 can be expressed by the following formula:

[0243] I MC[2,1],0 =k(V PR -V W[2,1] -V th ) 2 (E3)

[0244] In addition, current flows from the wiring BLref to the transistor Tr12 of the memory cell MCref[2]. current I MCref[2],0 can be expressed by the following formula:

[0245] I MCref[2],0 =k(V PR -V th ) 2 (E4)

[0246] Next, during the period from time T04 to time T05, the potential of the line WL[2] becomes low level. As a result, the transistors of the memory cell MC[2,1] and the memory cell MCref[2] The transistor Tr11 is turned off, and the nodes NM[2,1] and NMref[2 ] potential is maintained.

[0247] Through the above operations, the first data is stored in the memory cells MC[1,1] and MC[2,1]. Then, the reference data is stored in memory cells MCref[1] and MCref[2].

[0248] Here, in the period from time T04 to time T05, Consider the current that flows. A current is supplied to the wiring BLref from the current source circuit CS. Also, The current flowing through the wiring BLref is transmitted through the current mirror circuit CM and the memory cell MCref[1 ], and is discharged to MCref[2]. The current supplied from the current source circuit CS to the wiring BLref Flow I Cref The current flowing from the wiring BLref to the current mirror circuit CM is I CM ,0 Then, the following equation holds:

[0249] I Cref -I CM,0 =I MCref[1],0 +I MCref[2],0 (E5)

[0250] A current is supplied to the wiring BL[1] from the current source circuit CS. The current flows through the current mirror circuit CM, memory cells MC[1,1] and MC[2,1]. Also, a current flows from the wiring BL[1] to the offset circuit OFST. The current supplied from the circuit CS to the wire BL[1] is I C,0 , offset from wiring BL[1] The current flowing through the OFST circuit is I α,0 Then, the following equation holds:

[0251] I C -I CM,0 =I MC[1,1],0 +I MC[2,1],0 +I α,0 (E6)

[0252] [Multiply and add the first data and the second data] Next, during the period from time T05 to time T06, the potential of the wiring RW[1] becomes lower than the reference potential. V X[1] At this time, the memory cell MC[1,1] and the memory cell M Each capacitance element C11 of Cref[1] is connected to a potential V X[1] is supplied to the capacitive coupling This causes the potential of the gate of the transistor Tr12 to rise. X[1] The memory , which corresponds to the second data supplied to the memory cell MC[1,1] and the memory cell MCref[1]. It is the electric potential.

[0253] The change in the potential of the gate of the transistor Tr12 is proportional to the change in the potential of the wiring RW. The capacitance coupling coefficient is determined by the capacitance element. It is calculated based on the capacitance of C11, the gate capacitance of transistor Tr12, and the parasitic capacitance. For the sake of convenience, in the following, the amount of change in the potential of the wiring RW and the amount of change in the potential of the gate of the transistor Tr12 are The explanation will be given assuming that the amount of change is the same, that is, the capacitive coupling coefficient is 1. Considering the number, the potential V X It is sufficient to determine the following.

[0254] A potential V X [1] When the potential of the node NM[1,1] and the node NMref[1] is Each V X[1] to rise.

[0255] In the period from time T05 to time T06, the line BL[1] is connected to the memory cell MC[ Current I flows through transistor Tr12 at [1,1] MC[1,1],1 can be expressed by the following formula: This can be done.

[0256] I MC[1,1],1 =k(V PR -V W[1,1] +V X[1] -V th ) 2 (E7)

[0257] That is, the potential V X[1] By supplying The current flowing through the transistor Tr12 of the memory cell MC[1,1] is ΔI MC[1,1] =I MC[1,1],1 -I MC[1,1],0 Increase.

[0258] In addition, during the period from time T05 to time T06, the memory cell MCre Current I flowing through transistor Tr12 of f[1] MCref[1],1 is expressed by the following formula: It is possible.

[0259] I MCref[1],1 =k(V PR +V X[1] -V th ) 2 (E8)

[0260] That is, the potential V X[1] By supplying The current flowing through the transistor Tr12 of the memory cell MCref[1] is ΔI MCref[ 1] =I MCref[1],1 -I MCref[1],0 Increase.

[0261] Let us now consider the current flowing through the wiring BL[1] and the wiring BLref. The current I Cref Also, the current flowing through the wiring BLref The current is drained to the current mirror circuit CM and the memory cells MCref[1] and MCref[2]. The current discharged from the wiring BLref to the current mirror circuit CM is I CM,1 and Then, the following equation holds:

[0262] I Cref -I CM,1 =I MCref[1],1 +I MCref[2],0 (E9)

[0263] The wire BL[1] carries the current I C Also, wiring BL[1] The current flowing through the current mirror circuit CM, memory cells MC[1,1] and MC[2,1] In addition, a current also flows from the wiring BL[1] to the offset circuit OFST. The current flowing from the wiring BL[1] to the offset circuit OFST is I α,1 Then, the following formula is It works.

[0264] I C -I CM,1 =I MC[1,1],1 +I MC[2,1],1 +I α,1 (E10)

[0265] Then, from equations (E1) to (E10), the current I α,0 and current I α,1 (Differential current ΔI α ) can be expressed by the following formula:

[0266] ΔI α =I α,1 -I α,0 =2kV W[1,1] V X[1] (E11)

[0267] Thus, the differential current ΔI α is the potential V W[1,1] and V X[1] The value depends on the product of .

[0268] After that, during the period from time T06 to time T07, the potential of the line RW[1] becomes the reference potential. The potentials of the nodes NM[1,1] and NMref[1] are The period will be the same as

[0269] Next, during the period from time T07 to time T08, the potential of the line RW[1] becomes lower than the reference potential. V X[1] The potential of the wiring RW[2] becomes V higher than the reference potential. X[2] big This causes the memory cell MC[1,1] and the memory cell MCref[1] The potential V X[1] is supplied, and the node NM[ 1,1] and node NMref[1] are V X[1] Also, memo The capacitance element C11 of the rechargeable cell MC[2,1] and the memory cell MCref[2] Potential V X[2] is supplied, and the nodes NM[2,1] and NMref [2] has a potential of V X[2] to rise.

[0270] In the period from time T07 to time T08, the line BL[1] is connected to the memory cell MC[ Current I flows through transistor Tr12 at [2,1] MC[2,1],1 can be expressed by the following formula: This can be done.

[0271] I MC[2,1],1 =k(V PR -V W[2,1] +V X[2] -V th ) 2 (E12)

[0272] That is, the potential V X[2] By supplying The current flowing through the transistor Tr12 of the memory cell MC[2,1] is ΔI MC[2,1] =I MC[2,1],1 -I MC[2,1],0 Increase.

[0273] In addition, during the period from time T07 to time T08, the memory cell MCre Current I flowing through transistor Tr12 of f[2] MCref[2],1 is expressed by the following formula: It is possible.

[0274] I MCref[2],1 =k(V PR +V X[2] -V th ) 2 (E13)

[0275] That is, the potential V X[2] By supplying The current flowing through transistor Tr12 of memory cell MCref[2] is ΔI MCref[ 2] =I MCref[2],1 -I MCref[2],0 Increase.

[0276] Let us now consider the current flowing through the wiring BL[1] and the wiring BLref. The current I Cref Also, the current flowing through the wiring BLref The current is drained to the current mirror circuit CM and the memory cells MCref[1] and MCref[2]. The current discharged from the wiring BLref to the current mirror circuit CM is I CM,2 and Then, the following equation holds:

[0277] I Cref -I CM,2 =I MCref[1],1 +I MCref[2],1 (E14)

[0278] The wire BL[1] carries the current I C Also, wiring BL[1] The current flowing through the current mirror circuit CM, memory cells MC[1,1] and MC[2,1] In addition, a current also flows from the wiring BL[1] to the offset circuit OFST. The current flowing from the wiring BL[1] to the offset circuit OFST is I α,2 Then, the following formula is It works.

[0279] I C -I CM,2 =I MC[1,1],1 +I MC[2,1],1 +I α,2 (E15)

[0280] From the formulas (E1) to (E8) and the formulas (E12) to (E15), the current I α,0 and current I α,2 Difference between the current (ΔI α ) can be expressed by the following formula:

[0281] ΔI α =I α,2 -I α,0 =2k(V W[1,1] V X[1] +V W[2,1] V X[ 2] ) (E16)

[0282] Thus, the differential current ΔI α is the potential V W[1,1] and potential V X[1] and the potential V W [2,1] and potential V X[2] The value is determined by adding the product of and .

[0283] After that, during the period from time T08 to time T09, the potential of the wirings RW[1] and RW[2] is The nodes NM[1,1], NM[2,1] and the nodes NMref[1], NM The potential of ref[2] becomes the same as in the period from time T04 to time T05.

[0284] As shown in equations (E11) and (E16), the offset signal is input to the offset circuit OFST. Differential current ΔIα is the potential V corresponding to the first data (weight) W and the second data (input data The potential V X It can be calculated from an equation that has the product terms of Minute current ΔI α The offset circuit OFST measures the first data and the second data. The result of the multiplication and accumulation of data can be obtained.

[0285] In the above, the memory cells MC[1,1], MC[2,1] and MCre Focusing on f[1] and MCref[2], the memory cells MC and MCref The number of rows m of the memory cells MC and the memory cells MCref can be set arbitrarily. The differential current ΔIα for any number i can be expressed by the following equation:

[0286] ΔI α =2kΣ i V W[i,1] V X[i] (E17)

[0287] In addition, by increasing the number of columns n of the memory cells MC and the memory cells MCref, The number of multiply-add operations performed can be increased.

[0288] As described above, by using the semiconductor device MAC, the multiplication of the first data and the second data can be performed. The memory cell MC and the memory cell MCref shown in FIG. By using the configuration shown in 4, a multiply-and-accumulate circuit can be constructed with a small number of transistors. Therefore, the circuit scale of the semiconductor device MAC can be reduced.

[0289] When the semiconductor device MAC is used for calculations in a neural network, the memory cell MC The number of rows m corresponds to the number of input data supplied to one neuron, and the number of columns of memory cells MC The number n can correspond to the number of neurons. For example, the hidden layer H Consider a case where a multiply-and-accumulate operation is performed using a semiconductor device MAC in L. The number of rows m of the model MC is the number of input data provided by the input layer IL (the number of neurons in the input layer IL). The number of columns n of the memory cells MC is set to the number of neurons in the hidden layer HL. This can be done.

[0290] The structure of the neural network to which the semiconductor device MAC is applied is not particularly limited. For example, the semiconductor device MAC is a convolutional neural network (CNN), a recurrent neural network (RNN), Recurrent neural networks (RNNs), autoencoders, Boltzmann machines (restricted Boltzmann machines) It can also be used for other purposes, such as

[0291] As described above, by using the semiconductor device MAC, the product-sum operation of the neural network can be realized. Furthermore, the cell array CA includes the memory cells MC and memory By using the reseller MCref, it is possible to improve the calculation accuracy, reduce the power consumption, or reduce the circuit regulation. It is possible to provide an integrated circuit that allows for a reduced size.

[0292] This embodiment mode can be combined with other embodiment modes as appropriate. [Explanation of symbols]

[0293] AD: Text analysis data, AD ref : Reference text analysis data, C11: Capacitive element, C21 : Capacitive element, ID: IDF data, KD: Keyword data, KW: Keyword, KW1 :Keyword, KW2:Keyword, KWx :Keywords, LD:Ranking data, N N: neural network, R1: resistor element, RD: related word data, RS: relevance, R W: Related words, RW x1 :Related words, RW x2 :Related words, RW xy :Related words, TD:Text data T, T.D. ref :Reference text data, Tr11: transistor, Tr12: transistor , Tr21: transistor, Tr22: transistor, Tr23: transistor, VD: Vector data, 100: document search system, 101: input unit, 102: transmission line, 103 : Processing unit, 105: Storage unit, 107: Database, 109: Output unit, 150: Document search System, 151: server, 152: terminal, 161a: communication unit, 161b: communication unit, 16 2: transmission path, 163a: processing unit, 163b: processing unit, 165: storage unit, 167: database 168: transmission line, 169: input / output section

Claims

1. A document search device for searching for documents related to or similar to an input document, comprising: The document search device has a processing unit, the processing unit has a function of extracting a first distributed representation vector and a first weight of the first distributed representation vector from a document input to the processing unit; the processing unit has a function of extracting a second distributed representation vector by using a similarity of the first distributed representation vector; The processing unit has a function of acquiring a second weight of the second distributed representation vector, the second weight being a product of the first weight and a similarity; The processing unit outputs the first weight and the second weight; The processing unit accepts a change to at least one of the first weighting and the second weighting, and performs a document search using the first weighting and the second weighting after the acceptance.

2. In claim 1, A document search device in which the first distributed representation vector and the second distributed representation vector are obtained by machine learning distributed representations of words contained in multiple reference text analysis data.

3. In claim 1, A document search device, wherein the first distributed representation vector is a vector obtained from a keyword extracted from a document input to the processing unit.

4. In claim 2, A document search device, wherein the plurality of reference text analysis data are data generated by performing morphological analysis on a document input to the processing unit.

5. In claim 2, The machine learning is a document search device that uses a neural network.

6. A document search device having a storage unit and a processing unit, The storage unit has a function of storing a program to be executed by the processing unit, The processing unit has a function of searching for a document from an input document, The document search device includes: an operation of extracting a first distributed representation vector and a first weight of the first distributed representation vector from the document input to the processing unit; an operation of extracting a second distributed representation vector using a similarity of the first distributed representation vector; obtaining a second weight of the second distributed representation vector, the second weight being a product of the first weight and a similarity; outputting the first weight and the second weight; an operation of executing a document search using one or more of the changed first weight and second weight values, and the values ​​of the keyword data and related word data changed by editing; A document search device having the above structure.

7. In claim 6, A document search device in which the first distributed representation vector and the second distributed representation vector are obtained by machine learning distributed representations of words contained in multiple reference text analysis data.

8. In claim 6, A document search device, wherein the plurality of reference text analysis data are data generated by performing morphological analysis on a document input to the processing unit.

9. In claim 6, The machine learning is a document search device that uses a neural network.

10. In claim 6, A document search device, wherein the first distributed representation vector is a vector obtained from a keyword extracted from a document input to the processing unit.

Images