Information processing system and information processing method

Abstract

Provided is a novel information processing system that exhibits exceptional convenience, usefulness, or reliability. This information processing system includes a first information processing device and a second information processing device. The first information processing device has: a function for receiving input of a first-language document; a function for translating the first-language document to a second language; a function for reverse-translating the second-language document to the first language; a function for comparing the two same-language documents and determining whether the documents are synonymous; a function for transmitting, to the second information processing device, a prompt including a first-language document A that is determined to not be synonymous in the aforementioned determination and instruction text for inducing generation of a first-language document B that is synonymous with the document A, thereby acquiring the document B; and a function for outputting the document B. The second information processing device has a function for generating the document B in accordance with the prompt.

Description

Information processing system and information processing method

[0001] One aspect of the present invention relates to an information processing system and an information processing method.

[0002] Note that one embodiment of the present invention is not limited to the above technical field. The technical field of one embodiment of the invention disclosed in this specification and the like relates to an object, a method, or a manufacturing method. Alternatively, one embodiment of the present invention relates to a process, a machine, manufacture, or a composition of matter. Therefore, more specifically, examples of the technical field of one embodiment of the present invention disclosed in this specification include a semiconductor device, a display device, a light-emitting device, a power storage device, a memory device, a driving method thereof, or a manufacturing method thereof.

[0003] Research and development into machine translation, which uses a computer to translate one natural language into another, is currently underway. Examples of machine translation include rule-based machine translation, which performs translation based on rules, statistical machine translation, which performs translation using a language model and a translation model, and neural machine translation, which performs translation using an artificial neural network (ANN, hereinafter simply referred to as a neural network).

[0004] In machine translation, the source text may be corrected (also called pre-editing) before translation (Patent Document 1).

[0005] In recent years, the development of language models using neural networks has been actively pursued, with large-scale language models (LLMs) attracting particular attention. A large-scale language model is a natural language processing model trained using a large amount of data. A large-scale language model can realize, for example, a dialogue model that responds to user instructions. Non-Patent Document 1 discloses GPT-4 (registered trademark) (Generative Pre-trained Transformer 4) as a large-scale language model, and also discloses ChatGPT as a chat service.

[0006] The use of large-scale language models has significantly increased the capabilities of natural language processing models. However, as language models become larger, it is difficult to incorporate and operate language models in-house due to the equipment and cost involved. Therefore, one way to use language models is to use external services that provide language models.

[0007] JP 2014-229122

[0008] Summary of ChatGPT / GPT-4 Research and Perspective Towards the Future of Large Language Models, Yiheng Liu et al. (Submitted on 4 Apr 2023, [online], Internet <URL: https: / / arxiv.org / abs / 2304.01852>

[0009] Patent Literature 1 discloses a translation device that enables pre-editing when translating an original text, including a function for generating original candidate sentences processed based on the original text using rule-based processing, a function for translating the generated original candidate sentences into another language, a function for further back-translating them back into the original language, and a function for comparing the original candidate sentences and the back-translated sentences to calculate similarities. Here, in order to generate original candidate sentences using rule-based processing, rules for generation must be prepared in advance, and creating these rules requires specialized linguistic knowledge and knowledge of machine translation algorithms. Patent Literature 1 simplifies the generation of original candidate sentences using these rules by defining only simple partial expressions as expression patterns, but the accuracy of generating original candidate sentences ultimately depends on the linguistic knowledge and knowledge of machine translation algorithms of the creator of the rules.

[0010] In view of the above, an object of one embodiment of the present invention is to provide an information processing system capable of generating source candidate sentences that are easy to translate without having specialized knowledge of language and machine translation algorithms.Another object of one embodiment of the present invention is to provide an information processing method capable of generating source candidate sentences that are easy to translate without having specialized knowledge of language and machine translation algorithms.Another object of one embodiment of the present invention is to provide an information processing device capable of generating source candidate sentences that are easy to translate without having specialized knowledge of language and machine translation algorithms.

[0011] An object of one embodiment of the present invention is to provide a novel information processing system with excellent convenience, usefulness, or reliability, or to provide a novel information processing method with excellent convenience, usefulness, or reliability, or to provide a novel information processing system, a novel information processing method, or a novel semiconductor device.

[0012] Note that the description of these problems does not preclude the existence of other problems. Note that one embodiment of the present invention does not necessarily solve all of these problems. Note that problems other than these will become apparent from the description of the specification, drawings, claims, etc., and it is possible to extract other problems from the description of the specification, drawings, claims, etc.

[0013] One aspect of the present invention is an information processing system comprising a first information processing device and a second information processing device, wherein the first information processing device has a function of accepting input of a document in a first language, a function of translating the document in the first language into a second language, a function of reverse-translating the document in the second language into the first language, a function of comparing two documents in the same language and determining whether they are synonymous, a function of acquiring document B by sending a prompt to the second information processing device including document A in the first language that has been determined to be not synonymous, and an instruction sentence for generating document B in the first language that is synonymous, and a function of outputting document B, and the second information processing device has a function of generating document B in accordance with the prompt.

[0014] In the above, it is preferable that the first information processing device has a function of outputting document B when it determines that document B is synonymous with document C obtained by translating document B into a second language and then back-translating it into the first language.

[0015] Alternatively, one aspect of the present invention includes a first information processing device and a second information processing device, wherein the first information processing device has a function of accepting input of document A in a first language, a function of translating document A into document B in a second language, and a function of reverse-translating document B into document C in the first language, and a function of comparing document A with document C as a first determination and determining whether they are synonymous, and the first information processing device has a function of generating document A and document D1 synonymous with document A when the first determination is not synonymous, and to a second information processing device, the second information processing device having a function of generating a document D1 in a first language in accordance with the prompt, the first information processing device having a function of translating the document D1 into a document E1 in a second language, the first information processing device having a function of reverse-translating the document E1 into a document F1 in the first language, the first information processing device having a function of comparing the document D1 with the document F1 and determining whether they are synonymous as a second determination, and the first information processing device having a function of outputting the document D1 if the second determination is that they are synonymous.

[0016] In the above information processing system, the first information processing device preferably has a function of outputting a message indicating that document A does not contain any passages that are difficult to translate when the first determination is synonymous.

[0017] Alternatively, in the above information processing system, it is preferable that the first information processing device has a function of sending a prompt including document A and an instruction sentence for generating document D2 synonymous with document A to the second information processing device when the second judgment is not synonymous, the second information processing device has a function of generating document D2 in the first language in accordance with the prompt, the first information processing device has a function of translating document D2 into document E2 in the second language, the first information processing device has a function of reverse-translating document E2 into document F2 in the first language, the first information processing device has a function of comparing document D2 and document F2 as a third judgment and determining whether they are synonymous, and the first information processing device has a function of outputting document D2 when the second judgment is synonymous.

[0018] In any of the above information processing systems, it is preferable that the second information processing device includes a large-scale language model, and that the document D1 is generated by the large-scale language model.

[0019] Alternatively, one aspect of the present invention is an information processing method having the steps of: (A) accepting document A in a first language; (B) translating document A into a second language to generate document B; (C) back-translating document B into the first language to generate document C; (D) comparing document A and document C as a first determination to determine whether they are synonymous; (E) if the first determination is not synonymous, causing a large-scale language model to generate at least one document D that is synonymous with document A and is in the first language; (F) translating document D into the second language to generate document E; (G) back-translating document E into the first language to generate document F; (H) comparing document D and document F as a second determination to determine whether they are synonymous; and (I) if the second determination is synonymous, outputting document D.

[0020] Alternatively, one aspect of the present invention includes the steps of: (A) accepting document A in a first language; (B) causing a large-scale language model to translate document A into a second language to generate document B; (C) causing the large-scale language model to back-translate document B into the first language to generate document C; (D) causing the large-scale language model to compare document A and document C as a first determination and determine whether they are synonymous; and (E) causing the large-scale language model to determine whether they are synonymous, if the first determination is not synonymous, that they are synonymous with document A, and This information processing method includes the steps of: (F) generating at least one document D in a first language; (G) causing a large-scale language model to translate document D into a second language to generate document E; (H) causing the large-scale language model to compare document D with document F as a second determination to determine whether they are synonymous; and (I) outputting document D if the second determination is that they are synonymous.

[0021] In the drawings accompanying this specification, components are classified by function and shown as block diagrams that are independent of each other, but in reality, it is difficult to completely separate components by function, and one component may be involved in multiple functions.

[0022] According to one aspect of the present invention, it is possible to provide an information processing system capable of generating source sentence candidate sentences that are easy to translate without requiring specialized knowledge of language and machine translation algorithms. Alternatively, according to one aspect of the present invention, it is possible to provide an information processing method capable of generating source sentence candidate sentences that are easy to translate without requiring specialized knowledge of language and machine translation algorithms. Alternatively, according to one aspect of the present invention, it is possible to provide an information processing device capable of generating source sentence candidate sentences that are easy to translate without requiring specialized knowledge of language and machine translation algorithms.

[0023] According to one aspect of the present invention, a novel information processing system with excellent convenience, usefulness, or reliability can be provided. Alternatively, according to one aspect of the present invention, a novel information processing method with excellent convenience, usefulness, or reliability can be provided. Also, a novel information processing system can be provided. Alternatively, a novel information processing method can be provided.

[0024] Note that the description of these effects does not preclude the existence of other effects. Note that one embodiment of the present invention does not necessarily have all of these effects. Note that effects other than these will become apparent from the description in the specification, drawings, claims, etc., and it is possible to extract other effects from the description in the specification, drawings, claims, etc.

[0025] FIG. 1 is a diagram illustrating the configuration of an information processing system. FIG. 2 is a diagram illustrating an information processing method. FIGS. 3A and 3B are diagrams illustrating the information processing method. FIGS. 4A and 4B are diagrams illustrating the information processing method. FIG. 5 is a diagram illustrating the configuration of an information processing system. FIG. 6 is a diagram illustrating the configuration of an information processing device. FIG. 7A is a diagram illustrating the configuration of an information processing system. FIG. 7B is a diagram illustrating the information processing method. FIG. 8 is a diagram illustrating the information processing method.

[0026] An information processing system according to one aspect of the present invention is a machine translation system that translates one natural language into another natural language.

[0027] For example, an information processing system of one embodiment of the present invention has a function of, when document A (also referred to as original text) expressed in a first language is input, translating document A expressed in the first language into a second language to create document B (also referred to as translated text). The information processing system of one embodiment of the present invention also has a function of translating document B into the first language to create document C (also referred to as reverse-translated text). Note that, in the above description, translating from a first language into a second language and then translating back into the first language is also referred to as reverse translation.

[0028] An information processing system according to one embodiment of the present invention has a function of comparing document A (original text) expressed in a first language with document C (reverse-translated text) and performing a first synonymy evaluation (also referred to as synonymy determination). Furthermore, when it is determined in the synonymy evaluation that document A and document C are not synonymous, the information processing system according to one embodiment of the present invention has a function of using generative artificial intelligence (AI) to generate document D1 (also referred to as a source candidate sentence) expressed in the first language that is synonymous with document A. Note that in this specification, generative AI refers to a dialogue model program and system that has a function of generating documents using a large-scale language model.

[0029] In addition, an information processing system according to one embodiment of the present invention has a function of translating the above-mentioned document D1 (original candidate sentence 1) expressed in a first language into a second language to create document E1, a function of reverse-translating document E1 to create document F1, and a function of comparing document D1 with document F1 and performing a second synonym evaluation.

[0030] Furthermore, the information processing system according to one embodiment of the present invention has a function of generating document D2 (source candidate sentence 2) different from document D1 (source candidate sentence 1) and repeating the above process if the second synonymy evaluation determines that document D1 and document F1 are not synonymous. This process can be repeated until the synonymity of at least one source candidate sentence and its reverse-translated sentence reaches or exceeds a specified value. In this specification, synonymity refers to a numerical value that indicates whether documents are synonymous.

[0031] Furthermore, the information processing system according to one embodiment of the present invention has a function of outputting a set of document A, document C, and the results of the first synonymy evaluation (e.g., synonymity numerical values), and a set of document D1, document F1, and the results of the second synonymy evaluation, side by side. In this case, depending on the results of the synonymy evaluation, for example, a set of documents with a higher degree of synonymity may be displayed at the top (e.g., displayed in descending order).

[0032] A user of an information processing system according to an embodiment of the present invention can check the set of document and synonym evaluation results output as described above, and choose whether to use document A (original text) or adopt document D1 (original text candidate sentence 1) and modify document A to document D1. In this way, using the information processing system according to an embodiment of the present invention, a user can perform pre-editing, modifying the original text to be translated into a form that is easier to translate.

[0033] In particular, by using generation AI to generate candidate source sentences, it is possible to effectively and efficiently modify the source text to be translated into a form that is easier to translate, even if the user does not have specialized linguistic knowledge or knowledge of machine translation algorithms.

[0034] An information processing system according to one aspect of the present invention has a function of repeatedly performing the above series of processes from input to output in units of one sentence, two or more sentences, one paragraph, two or more paragraphs, etc. The user checks and selects a set of output documents and synonym evaluation results, and this checking and selection may be performed for each of the above repetition units, for each of multiple repetition units, or after the entire process is completed.

[0035] In addition, terms using ordinal numbers such as first, second, and third used in this specification are given for convenience in identifying components and do not limit the number.

[0036] The embodiments will be described in detail with reference to the drawings. However, the present invention is not limited to the following description, and it will be readily understood by those skilled in the art that various changes in form and details can be made without departing from the spirit and scope of the present invention. Therefore, the present invention should not be interpreted as being limited to the description of the embodiments shown below. In the configuration of the invention described below, the same parts or parts having similar functions will be denoted by the same reference numerals in different drawings, and repeated explanations will be omitted.

[0037] Embodiment 1 In this embodiment, an information processing system and an information processing method according to one embodiment of the present invention will be described with reference to FIGS.

[0038] FIG. 1 is a diagram illustrating a configuration of an information processing system according to one embodiment of the present invention.

[0039] 1, the information processing system described in this embodiment includes an information processing device 10 and an information processing device 40. The information processing device 10 and the information processing device 40 are connected via a network 30, and can transmit and receive document data and the like.

[0040] Furthermore, the information processing system described in this embodiment may be configured so that a user can input documents, etc. by directly operating the information processing device 10, or may be configured so that a user can input documents, etc. by using an information terminal 20 connected to the information processing device 10 via a network 31, as shown in Figure 1.

[0041] <Example 1 of Information Processing Method> An example of an information processing method used in the information processing system described in this embodiment will be described with reference to FIGS. 2 to 4B.

[0042] 2 to 4B are flowcharts illustrating an example of an information processing method according to one embodiment of the present invention.

[0043] The information processing method according to one aspect of the present invention starts, and in step S101 of FIG. 2, document A (original text) expressed in a first language is input.

[0044] Next, in step S111 of FIG. 2, document A expressed in a first language is translated into a second language to create document B (translated text).

[0045] Next, in step S121 of FIG. 2, document B is translated into the first language to create document C (back-translated text).

[0046] Next, in step S131 of FIG. 2, document A (original text) expressed in a first language is compared with document C (back-translated text), and a first synonymy evaluation (also called synonym determination) is performed.

[0047] For the synonym evaluation, a machine learning model can be used, which is a neural network model, BERT (Bidirectional Encoder Representation from Transformers), which can be fine-tuned.

[0048] When a machine learning model is used for synonym evaluation, an entailment recognition task may be used, replacing "entailed" and "not entailed" with "synonymous" and "not synonymous." For example, two sentences may be compared, and those with equivalent information may be labeled as synonymous, and those with unequal information may be labeled as not synonymous, and then the model may be trained. Furthermore, for example, document A and document C may be compared, and the information in document C may be labeled as insufficient, excessive, equivalent (synonymous), or contradictory to that in document A, and then the model may be trained.

[0049] Alternatively, synonym evaluation may be performed by comparing sentence vectors and evaluating the cosine similarity of the vectors. In this case, for example, it can be defined that a similarity calculated by the cosine similarity is equal to or greater than a certain value, and that a similarity less than the certain value is not synonymous.

[0050] Alternatively, LLM can be used for synonymy evaluation. When LLM is used for synonymy evaluation, a method of outputting a binary choice between whether document A and document C are synonymous or not synonymous, instead of a numerical value of the degree of synonymy, may be used. Furthermore, in synonymy evaluation using LLM, if the information contained in document C is excessive or missing compared to the information contained in document A, such information may be added and output together with the judgment. Note that in synonymy evaluation using LLM, documents A and C may be compared, and learning may be performed by labeling the information in document C with document A as being insufficient, excessive, equivalent (synonymous), or contradictory.

[0051] Alternatively, the synonym evaluation may be performed by combining the cosine similarity of the vectors and the synonym determination label. For example, the synonym evaluation can be performed by the product of the cosine distance converted from the cosine similarity and the label evaluation value. As an example, the label evaluation value can be defined as 1 for an excessive label, -1 for an insufficient label, 0 for a synonymous label, and nan for a contradictory label. In this case, the closer the cosine distance x label evaluation value is to 0, the more synonymous the label is, and the excess or deficiency of information can also be expressed. This is more effective than simply expressing the similarity because it can evaluate the excess or deficiency of information when it is particularly objectionable to add extra information that is not in the original text.

[0052] Next, in step S141 of Fig. 2, if the degree of synonymy is equal to or greater than a specified value, it is determined that document A and document C are synonymous, and the process proceeds to connector A. Here, connector A is a symbol that connects step S141 to the flowchart shown in Fig. 3A or 4A. Alternatively, in step S141 of Fig. 2, if the degree of synonymy is lower than a specified value, it is determined that document A and document C are not synonymous, and the process proceeds to step S201.

[0053] If it is determined that document A and document C are not synonymous, then in step S201 of Fig. 2, a document D1 (source candidate sentence) expressed in the first language and synonymous with document A is generated. Note that it is preferable to use a large-scale language model (sometimes referred to as generation AI) to generate the source candidate sentence.

[0054] When a large-scale language model is used to generate source candidate sentences, a prompt including document A (also referred to as the source sentence or input sentence) and an instruction such as "Please proofread the following input sentence and output a sentence that is easier to understand. However, as a constraint, the meaning of the sentence must not be changed" can be provided to the large-scale language model to output source candidate sentences that are synonymous with the source sentence. Note that the prompt corresponds to an input sentence that causes the language model to perform a desired operation. When a prompt is provided, the language model generates a response content based on the prompt.

[0055] Next, in step S211 of FIG. 2, document D1 (original candidate sentence 1) expressed in the first language is translated into the second language to create document E1.

[0056] Next, in step S221 of FIG. 2, the document E1 expressed in the second language is reverse-translated to generate a document F1 (reverse-translated sentence) expressed in the first language.

[0057] Next, in step S231 of FIG. 2, document D1 (original candidate sentence) and document F1 (backward-translated sentence) are compared to perform a second synonym evaluation.

[0058] Next, in step S241 of Fig. 2, if the degree of synonymy is equal to or greater than a specified value, it is determined that documents D1 and F1 are synonymous, and the process proceeds to connector B. Here, connector B is a symbol that connects step S241 to the flowchart shown in Fig. 3B or 4B. Alternatively, in step S241 of Fig. 2, if the degree of synonymy is lower than a specified value, it is determined that documents A and C are not synonymous, and the process proceeds to step S201, where the processes from step S201 to step S241 are performed again.

[0059] Next, using Figure 3A, a case will be described where it is determined in step S141 of Figure 2 that document A (original text) and document C (reverse-translated text) are synonymous, and the process proceeds to connector A. In step S301 connected from connector A, the results of the processes in steps S111 to S141 are output. The output results may include one or more selected from document A (original text), document B (translated text), document C (reverse-translated text), the first synonymy, and the like. Note that the output results preferably include at least document A (original text) and the first synonymy, and may further include information such as "documents A and C are synonymous, and document A did not contain any parts that were difficult to translate."

[0060] The flow when proceeding to connector A can end after the processing of step S301 described above. Alternatively, as shown in FIG. 4A, in step S321, if there is no next input sentence, the processing can end, and if there is a next input sentence, the processing can proceed to connector C. Here, connector C is a symbol that connects step S321 to step S101 shown in FIG. 2. When proceeding to connector C, it is possible to proceed to input processing of the next source sentence, and the above processing can be repeated until processing of all source sentences is completed.

[0061] Next, referring to FIG. 3B , a case will be described in which, in step S241 of FIG. 2, document D1 (original candidate sentence) and document F1 (reverse-translated sentence) are determined to be synonymous, and the process proceeds to connector B. In step S311, connected by connector B, the results of the processes in steps S111 through S241 are output. The output results may include one or more selected from document A (original sentence), document B (translated sentence), document C (reverse-translated sentence), the first synonymy degree, document D1 (original candidate sentence), document E1 (translated sentence), document F1 (reverse-translated sentence), the second synonymy degree, etc. The output results may include at least document D1 (original candidate sentence), and may also include document A (original sentence), the first synonymy degree, and the second synonymy degree. It may also include information such as "documents A and C are not synonymous, and document A contained a section that was difficult to translate." One or more documents D1 may be output. For example, the original candidate sentence with the highest synonymy can be output as document D1. If the process proceeds to connector B after repeating the processes of steps S201 to S241 multiple times, the output results may increase according to the number of repetitions. The output may be displayed according to the synonymy, for example, with a set of documents with higher synonymy rankings (for example, in descending order).

[0062] 3B, a document to be adopted is selected from the original sentence and one or more candidate original sentences. At this time, the user of an information processing system capable of implementing the information processing method of one aspect of the present invention may make the selection, or a rule may be set to select the document with the highest degree of synonymy.

[0063] The flow when proceeding to connector B can end after the processing of step S312 described above. Alternatively, as shown in FIG. 4B, in step S321, if there is no next input sentence, the processing can end, and if there is a next input sentence, the processing can proceed to connector C. Here, connector C is a symbol that connects step S321 to step S101 shown in FIG. 2. When proceeding to connector C, it is possible to proceed to input processing of the next original sentence, and the above processing can be repeated until processing of all original sentences is completed.

[0064] In the information processing method of one embodiment of the present invention described using Figures 2 to 4B, which device or terminal performs the processing of each step, information processing device 10, information processing device 40, or information terminal 20, will be described using Figure 5.

[0065] 5 is a relationship diagram showing the processes performed by the information processing device 10, the information processing device 40, and the information terminal 20, using the symbols for each step used in FIGS. 2 to 4B. Note that the block arrows labeled with network 30 indicate data transmission and reception between the information processing device 10 and the information processing device 40, and the block arrows labeled with network 31 indicate data transmission and reception between the information processing device 10 and the information terminal 20.

[0066] 5, information processing device 10 has a function of performing the process of step S111, a function of performing the process of step S121, a function of performing the process of step S131, a function of performing the process of step S141, a function of performing the process of step S211, a function of performing the process of step S221, a function of performing the process of step S231, a function of performing the process of step S241, a function of performing the process of step S301, and a function of performing the process of step S311. Information processing device 40 has a function of performing the process of step S201. Information terminal 20 has a function of performing the process of step S101 and a function of performing the process of step S312.

[0067] An example of the configuration of the information processing device 10, the information processing device 40, the information terminal 20, the network 30, and the network 31 will be described below.

[0068] <<Configuration Example of Information Processing Device 10>> A configuration example of the information processing device 10 included in the information processing system of one embodiment of the present invention will be described with reference to FIG. 6 .

[0069] As shown in FIG. 6, the information processing device 10 includes an input unit 110 , a storage unit 120 , a processing unit 130 , an output unit 140 , and a transmission path 150 .

[0070] [Input Unit 110] The input unit 110 can accept data from outside the information processing device 10. For example, the input unit 110 can accept data from the information terminal 20. The input unit 110 can also accept data from the information processing device 40.

[0071] The input unit 110 can supply the received data to one or both of the storage unit 120 and the processing unit 130 via the transmission path 150 .

[0072] [Storage Unit 120] The storage unit 120 has a function of storing a program executed by the processing unit 130. The storage unit 120 may also have a function of storing data generated by the processing unit 130 (e.g., calculation results, analysis results, inference results), data accepted by the input unit 110, and the like.

[0073] The storage unit 120 may have a database. Furthermore, the information processing device 10 may have a database separate from the storage unit 120. The information processing device 10 may have a function to retrieve data from a database that exists outside the storage unit 120, outside the information processing device 10, or outside the information processing system. Furthermore, the information processing device 10 may have a function to retrieve data from both its own database and an external database.

[0074] Either or both of a storage and a file server can be used as the memory unit 120. Also, a database that records paths of files stored in a file server can be used as the memory unit 120.

[0075] The storage unit 120 includes at least one of a volatile memory and a non-volatile memory. Examples of the volatile memory include a dynamic random access memory (DRAM) and a static random access memory (SRAM). Examples of the non-volatile memory include a resistive random access memory (ReRAM), a phase change random access memory (PRAM), a ferroelectric random access memory (FeRAM), a magnetoresistive random access memory (MRAM), and a flash memory. The storage unit 120 may also include at least one of NOSRAM (registered trademark) and DOSRAM (registered trademark). The storage unit 120 may also include a recording media drive. Examples of recording media drives include a hard disk drive (HDD) and a solid state drive (SSD).

[0076] NOSRAM is an abbreviation for "Nonvolatile Oxide Semiconductor Random Access Memory." NOSRAM refers to a memory in which memory cells are two-transistor (2T) or three-transistor (3T) gain cells, and the transistors are transistors (also called OS transistors) that use metal oxide in their channel formation regions. OS transistors have an extremely small leakage current, i.e., a current that flows between the source and drain in an off state. NOSRAM can be used as a nonvolatile memory by retaining a charge corresponding to data in the memory cell using its extremely small leakage current characteristic. In particular, NOSRAM can read stored data without destroying it (nondestructive read), making it suitable for arithmetic processing in which only data read operations are repeated a large number of times. NOSRAM can increase its data capacity by stacking layers, and therefore can be used as a large-scale cache memory, main memory, or storage memory to improve the performance of semiconductor devices.

[0077] DOSRAM is an abbreviation for "Dynamic Oxide Semiconductor RAM" and refers to a RAM having 1T (transistor) 1C (capacitor) type memory cells. DOSRAM is a DRAM formed using OS transistors, and is a memory that temporarily stores information sent from an external device. DOSRAM is a memory that takes advantage of the low off-state current of OS transistors.

[0078] In this specification and the like, a metal oxide refers to an oxide of a metal in a broad sense. Metal oxides are classified into oxide insulators, oxide conductors (including transparent oxide conductors), oxide semiconductors (also referred to as oxide semiconductors or simply as OSs), and the like. For example, when a metal oxide is used for a semiconductor layer of a transistor, the metal oxide may be referred to as an oxide semiconductor.

[0079] The metal oxide included in the channel formation region preferably contains indium (In). When the metal oxide included in the channel formation region contains indium, the carrier mobility (electron mobility) of the OS transistor is increased. Furthermore, the metal oxide included in the channel formation region is preferably an oxide semiconductor containing element M. The element M is preferably at least one of aluminum (Al), gallium (Ga), and tin (Sn). Other elements applicable to element M include boron (B), silicon (Si), titanium (Ti), iron (Fe), nickel (Ni), germanium (Ge), yttrium (Y), zirconium (Zr), molybdenum (Mo), lanthanum (La), cerium (Ce), neodymium (Nd), hafnium (Hf), tantalum (Ta), and tungsten (W). However, a combination of two or more of the above elements may be used as element M. The element M is, for example, an element having a high bond energy with oxygen. For example, it is an element having a higher bond energy with oxygen than indium. The metal oxide contained in the channel formation region is preferably a metal oxide containing zinc (Zn), since zinc-containing metal oxides may be easily crystallized.

[0080] The metal oxide contained in the channel formation region is not limited to a metal oxide containing indium, but may be, for example, a metal oxide containing zinc but not indium, such as zinc tin oxide or gallium tin oxide, a metal oxide containing gallium, or a metal oxide containing tin.

[0081] [Processing Unit 130] The processing unit 130 has a function of performing processes such as calculation, analysis, and inference using data supplied from one or both of the input unit 110 and the storage unit 120. The processing unit 130 can supply generated data (e.g., calculation results, analysis results, and inference results) to one or both of the storage unit 120 and the output unit 140.

[0082] The processing unit 130 has a function of acquiring data from the storage unit 120. The processing unit 130 may also have a function of recording or registering data in the storage unit 120.

[0083] The processing unit 130 may include, for example, an arithmetic circuit, a central processing unit (CPU), and a graphics processing unit (GPU).

[0084] The processing unit 130 may have a microprocessor such as a DSP (Digital Signal Processor). The microprocessor may be implemented by a PLD (Programmable Logic Device) such as an FPGA (Field Programmable Gate Array) or an FPAA (Field Programmable Analog Array). The processing unit 130 may also have a quantum processor. The processing unit 130 can perform various data processing and program control by interpreting and executing instructions from various programs using the processor. Programs that can be executed by the processor are stored in at least one of the memory area of ​​the processor and the storage unit 120.

[0085] The processing unit 130 may include a main memory. The main memory may include at least one of a volatile memory such as a random access memory (RAM) and a non-volatile memory such as a read-only memory (ROM). The main memory may also include at least one of the above-mentioned NOSRAM and DOSRAM.

[0086] The RAM may be, for example, a DRAM or an SRAM, and a virtual memory space is allocated and used as a working space for the processing unit 130. The operating system, application programs, program modules, program data, lookup tables, and the like stored in the storage unit 120 are loaded into the RAM for execution. The data, programs, and program modules loaded into the RAM are each directly accessed and operated by the processing unit 130.

[0087] The ROM can store a BIOS (Basic Input / Output System), firmware, etc., which do not require rewriting. Examples of ROM include mask ROM, OTPROM (One Time Programmable Read Only Memory), and EPROM (Erasable Programmable Read Only Memory). Examples of EPROMs include UV-EPROMs (Ultra-Violet Erasable Programmable Read Only Memories), which allow stored data to be erased by exposure to ultraviolet light, EEPROMs (Electrically Erasable Programmable Read Only Memories), and flash memories.

[0088] The processing section 130 can include one or both of an OS transistor and a transistor having silicon in a channel formation region (a Si transistor).

[0089] The processing unit 130 preferably includes an OS transistor. Because an OS transistor has an extremely small off-state current, using the OS transistor as a switch for retaining charge (data) flowing into a capacitor functioning as a memory element can ensure a long data retention period. By using this characteristic in at least one of the register and cache memory of the processing unit, the processing unit can be operated only when necessary, and can be turned off in other cases by saving information from the previous processing in the memory element. In other words, normally-off computing is possible, and the power consumption of the information processing system can be reduced.

[0090] It is preferable that the information processing device 10 uses AI for at least some of its processing.

[0091] In particular, it is preferable that the information processing device 10 uses an artificial neural network, which is realized by a circuit (hardware) or a program (software).

[0092] In this specification, a neural network refers to a general model that mimics the neural circuit network of a living organism, determines the connection strength between neurons through learning, and has problem-solving capabilities. A neural network has an input layer, an intermediate layer (hidden layer), and an output layer.

[0093] In this specification and the like, when discussing neural networks, determining the connection strengths (also called weighting coefficients) between neurons from existing information may be referred to as "learning."

[0094] In this specification and the like, the act of constructing a neural network using connection strengths obtained by learning and deriving a new conclusion from it may be referred to as "inference."

[0095] [Output Unit 140] The output unit 140 can output at least one of the calculation result, analysis result, and inference result in the processing unit 130 to the outside of the information processing device 10.

[0096] For example, the output unit 140 can transmit data to the information processing device 40. The output unit 140 can also transmit data to the information terminal 20.

[0097] [Transmission Path 150] The transmission path 150 has a function of transmitting data. Data can be transmitted and received between the input unit 110, the storage unit 120, the processing unit 130, and the output unit 140 via the transmission path 150.

[0098] <<Configuration Example of Information Processing Device 40>> The information processing device 40 can process received data and transmit the processing results. For example, the information processing device 40 can perform processing such as calculations using data received from the information processing device 10. Furthermore, the information processing device 40 can transmit the processing results to the information processing device 10. This can reduce the calculation load on the information processing device 10.

[0099] The information processing device 40 can perform processing using a natural language processing model that uses AI. For example, the information processing device 40 can execute processing using a natural language processing model that uses AI, such as BERT (Bidirectional Encoder Representations from Transformers) or T5 (Text-to-Text Transformer).

[0100] Furthermore, the information processing device 40 can perform processing using a model (such as a sentence generation model or a dialogue model) that utilizes a large-scale language model. It is preferable that the generation of source candidate sentences in step S201 described in FIG. 2 be performed using a model that utilizes a large-scale language model. For example, it is possible to perform processing using a large-scale language model such as GPT-3, GPT-3.5, GPT-4 (registered trademark), LaMDA (Language Model for Dialogue Applications), PaLM (Pathways Language Model), or Llama2. It is particularly preferable to use GPT-4 (registered trademark).

[0101] For example, the information processing device 40 can translate a document written in language A into a document written in language B. Furthermore, the information processing device 40 can translate a document written in language A into a document written in language B in accordance with a given directive. Furthermore, constraints can be included in the directives. This makes it possible to control the degree of freedom in translation by giving constraints.

[0102] Furthermore, the information processing device 40 can execute processing using a general-purpose language processing model that can perform various natural language processing tasks.

[0103] The information processing device 40 is a large-scale computer such as a server computer or a supercomputer. Preferably, the information processing device 40 has a function as a parallel computer. By using the information processing device 40 as a parallel computer, it is possible to perform large-scale calculations required for AI learning and inference, for example.

[0104] The information processing device 40 is a computer with higher processing power than the information processing device 10. For example, if both the information processing device 10 and the information processing device 40 have functions as parallel computers, the information processing device 40 has higher processing power than the information processing device 10 and can perform large-scale calculations. Also, for example, if both the information processing device 10 and the information processing device 40 can perform processing using a model that utilizes a large-scale language model, the information processing device 40 can execute processing using a large-scale AI model compared to the information processing device 10.

[0105] It should be noted that the service provider does not necessarily have to own the information processing device 40. For example, the service provider can use part of the service provided by another business or the like using the information processing device 40.

[0106] <<Configuration Example of Information Terminal 20>> The information terminal 20 can accept data input by a user of the information processing system of one aspect of the present invention. The information terminal 20 can also provide the user with data output by the information processing system of one aspect of the present invention.

[0107] Furthermore, the information terminal 20 can transmit data received from the user to the information processing device 10. Furthermore, the information terminal 20 can provide data received from the information processing device 10 to the user.

[0108] Furthermore, the information terminal 20 can transmit data generated based on data received from the user to the information processing device 10. Furthermore, the information terminal 20 can provide data generated based on data received from the information processing device 10 to the user.

[0109] For example, dedicated application software or a web browser is installed on the information terminal 20. A user can access the information processing device 10 via either of these. This allows the user to enjoy services using the information processing system according to one aspect of the present invention, for example, using a computer with lower processing power than the information processing device 10.

[0110] The information terminal 20 may also be called a client computer, etc. Each of the information terminals 20 is an information terminal device used by a user of the information processing system according to one aspect of the present invention.

[0111] For example, a desktop computer 20a, a notebook computer 20b, a smartphone 20c, and a tablet computer 20d can be used as the information terminal 20. The tablet computer 20d can also be used as a notebook computer by connecting it to a housing 21 having a keyboard.

[0112] <<Network 30>> The network 30 connects the information processing device 10 and the information processing device 40. This allows input data and processed data to be transmitted and received between the two. Furthermore, the load related to information processing can be distributed.

[0113] In this embodiment, the network 30 is mainly described as a computer network that is larger in scale than the network 31. For example, a global network can be used for the network 30. Specifically, the Internet, which is the foundation of the World Wide Web (WWW), can be used.

[0114] <<Network 31>> The network 31 connects a plurality of information terminals 20 and information processing devices 10. This allows data to be transmitted and received between them. It also allows the load related to information processing to be distributed. Furthermore, a service provider can provide a service using an information processing method according to one aspect of the present invention to a user via the network 31, for example.

[0115] For example, a local network can be used for the network 31. Also, an intranet or an extranet can be used for the network 31. Also, a personal area network (PAN), a local area network (LAN), a campus area network (CAN), a metropolitan area network (MAN), a wide area network (WAN), a global area network (GAN), etc. can be used for the network 31.

[0116] When a provider of a service using an information processing method according to one embodiment of the present invention and a user who receives the service belong to the same organization, such as a company, it is preferable that data be transmitted and received between the information terminal 20 and the information processing device 10 using, for example, a network 31 established within the organization. This allows data to be transmitted and received between the information terminal 20 and the information processing device 40 more securely than when data is transmitted over the Internet. Furthermore, confidential information within the organization can be prevented from leaking to the outside.

[0117] When wireless communication is performed, communication standards such as the fourth generation mobile communication system (4G), fifth generation mobile communication system (5G), and sixth generation mobile communication system (6G), or specifications standardized by the IEEE such as Wi-Fi (registered trademark) and Bluetooth (registered trademark), can be used as communication protocols or communication technologies.

[0118] <Configuration Example 2 of Information Processing System> An example of an information processing system different from the above-described configuration example 1 of the information processing system will be described with reference to FIGS. 7A and 7B.

[0119] The information processing system described here differs from information processing system configuration example 1 in that the processes performed by the information processing device 10 and the information processing device 40 are different, as explained in the relationship diagram of Fig. 5. Here, the different parts will be explained in detail, and the above explanation will be referred to for parts having the same configuration.

[0120] Figure 7A is a relationship diagram showing the processes performed by the information processing device 10, the information processing device 40, and the information terminal 20 using the symbols for each step used in Figures 2 to 4B, and shows an example configuration different from that of Figure 5.

[0121] 7A, information processing device 10 has a function to perform the processing of step S141, a function to perform the processing of step S241, a function to perform the processing of step S301, and a function to perform the processing of step S311. Information processing device 40 has a function to perform the processing of step S111, a function to perform the processing of step S121, a function to perform the processing of step S131, a function to perform the processing of step S201, a function to perform the processing of step S211, a function to perform the processing of step S221, and a function to perform the processing of step S231. Information terminal 20 has a function to perform the processing of step S101 and a function to perform the processing of step S312.

[0122] In the first example of the information processing system, the information processing device 40 is only responsible for "Step S201: Generate a source sentence candidate based on the source sentence." In contrast, in the example of the configuration shown in FIG. 7A , in addition to "Step S201: Generate a source sentence candidate based on the source sentence," the information processing device 40 is also responsible for "Step S111: Create a translation of the source sentence," "Step S121: Create a backward-translated sentence," "Step S131: Compare the source sentence with the backward-translated sentence and calculate the degree of synonymy," "Step S211: Create a translation of the source sentence candidate," "Step S221: Create a backward-translated sentence," and "Step S231: Compare the source sentence candidate with the backward-translated sentence and calculate the degree of synonymy." This configuration makes it possible to reduce the processing power required of the information processing device 10, and thus reduces the equipment installation costs when multiple users use this information processing system.

[0123] Fig. 7B is a diagram illustrating a part of the flow of an information processing method in the configuration example shown in Fig. 7A, and is a modified example of Fig. 2. The information processing method used in configuration example 2 of the information processing system has step S151 between step S101 and step S111, and also has step S152 between step S111 and step S121.

[0124] 7B, a prompt including document A and an instruction sentence instructing translation is created by information processing device 10, and the prompt is transmitted to information processing device 40. Upon receiving the prompt, information processing device 40 creates document B (translated text) and transmits it to information processing device 10.

[0125] Next, in step S152 of Fig. 7B, information processing device 10 creates a prompt including document B and an instruction sentence instructing reverse translation, and transmits the prompt to information processing device 40. Upon receiving the prompt, information processing device 40 creates document C (reverse-translated sentence) and transmits it to information processing device 10. Subsequent processing can be performed in the same manner as the processing method described with reference to Fig. 2 etc.

[0126] <Configuration Example 3 of Information Processing System> An example of an information processing system different from the above-described configuration examples 1 and 2 of the information processing system will be described with reference to FIG. 8 .

[0127] <Information Processing Method Example 2> Configuration Example 3 of the information processing system differs from the above-described information processing system configuration examples 1 and 2 in that it uses an information processing method different from information processing method example 1. Here, using Fig. 8 , parts that differ from information processing method example 1 described in the flow charts of Fig. 2 to Fig. 4B will be described in detail, and the above description will be referred to for parts that have the same configuration.

[0128] Fig. 8 is a modified example of the process from step S141 to step S241 shown in the flowchart of Fig. 2, and shows a flow in which the processes of step S141, step S401, step S411, step S421, step S431, and step S441 are performed in order. Note that Fig. 8 omits the description of steps before step S131 in Fig. 2.

[0129] 8, if the degree of synonymy is equal to or greater than a specified value, it is determined that document A and document C are synonymous, and the process proceeds to connector A. Here, connector A is a symbol that connects step S141 to the flowchart shown in FIG. 3A or 4A. Alternatively, if the degree of synonymy is lower than a specified value in step S141 of FIG. 8, it is determined that document A and document C are not synonymous, and the process proceeds to step S401.

[0130] If it is determined that document A and document C are not synonymous, then in step S401 of Fig. 8, multiple documents expressed in the first language that are synonymous with document A are generated. Fig. 8 shows an example in which five documents (original candidate sentences) that are different from document A (original sentence) and different from each other are generated in step S401.

[0131] 8, documents D1, D2, D3, D4, and D5 are generated. It is preferable to use a large-scale language model to generate source candidate sentences.

[0132] When a large-scale language model is used to generate source candidate sentences in step S401, a prompt including document A (also referred to as the source sentence or input sentence) and an instruction such as "Please suggest five sentences that are easy to translate from the input sentence below. Please do not change the meaning or style of the sentences," can be provided to the large-scale language model to output multiple source candidate sentences that are synonymous with the source sentence. Here, an example is shown in which five source candidate sentences are generated, but the number of source candidate sentences generated may be two, three, four, six or more.

[0133] 8, documents D1, D2, D3, D4, and D5 are translated to generate documents E1, E2, E3, E4, and E5, respectively. These processes may be performed in parallel or sequentially.

[0134] Next, in step S421 (S421A, S421B, S421C, S421D, and S421E) of FIG. 8, documents E1, E2, E3, E4, and E5 are reverse-translated to create documents F1, F2, F3, F4, and F5 expressed in the first language.

[0135] Next, in step S431 of FIG. 8, documents D1 to D5 (original candidate sentences) are compared with documents F1 to F5 (backward-translated sentences), respectively, to perform a second synonym evaluation.

[0136] In this way, in step S411A, document D1 expressed in the first language is translated into the second language to create document E1, in step S421A, document E1 expressed in the second language is reverse-translated to create document F1 (reverse-translated sentence) expressed in the first language, and in step S231, document D1 (original candidate sentence) and document F1 (reverse-translated sentence) are compared to obtain the result of second synonym evaluation 1.

[0137] Similarly, document D2 can also perform steps S411B through S431B to obtain the result of second synonym evaluation 2. Similarly, document D3 can also perform steps S411C through S431C to obtain the result of second synonym evaluation 3. Similarly, document D4 can also perform steps S411D through S431D to obtain the result of second synonym evaluation 4. Similarly, document D5 can also perform steps S411E through S431E to obtain the result of second synonym evaluation 5.

[0138] Next, in step S441 of FIG. 8, a determination is made based on the results of second synonymy evaluation 1 to second synonymy evaluation 5. For example, if the results of second synonymy evaluation 1 to second synonymy evaluation 5 include results with a synonymity degree equal to or greater than a specified value, the process proceeds to connector B. Here, connector B is a symbol that connects step S241 to the flowchart shown in FIG. 3B or 4B. Alternatively, if the results of second synonymy evaluation 1 to second synonymy evaluation 5 do not include results with a synonymity degree equal to or greater than a specified value, the process proceeds to step S401, and the processes of steps S401 to S441 are performed again.

[0139] In this way, in information processing method example 2, multiple source candidate sentences are generated simultaneously, translations of each source candidate sentence are created, and reverse-translations of each translation are created, and synonym evaluation can be performed. When multiple source candidate sentences are generated at once in this way, they are influenced by the source candidate sentence that was output earlier, making it possible to write source candidate sentences more effectively.

[0140] <Other Configurations> Although the information processing system and the information processing method described in this embodiment are directed to machine translation from a first language to a second language, the information processing system and the information processing method of one embodiment of the present invention are not limited to machine translation between two languages. For example, they can be used for machine translation between three languages ​​(also referred to as machine translation between two pairs of languages), which involves machine translation from a first language to a second language and machine translation from the first language to a third language. Similarly, they can also be used for machine translation between multiple languages ​​(also referred to as machine translation between multiple pairs of languages), which involves machine translation between four or more languages.

[0141] When an information processing system and an information processing method according to an embodiment of the present invention are used for machine translation between two or more pairs of languages, there may be cases where the processing results do not match between the languages, for example, when processing a first language as the source text, translation candidate sentence A is determined to be the easiest document to translate into the second language, and another translation candidate sentence B is determined to be the easiest document to translate into the third language. In such cases, it is advisable to determine in advance the priority of which language should be given priority in terms of ease of translation.

[0142] The information processing system, information processing method, and information processing device described above can provide an information processing system capable of generating source sentence candidate sentences that are easy to translate without requiring specialized knowledge of language and machine translation algorithms. Alternatively, according to one aspect of the present invention, an information processing method can be provided that can generate source sentence candidate sentences that are easy to translate without requiring specialized knowledge of language and machine translation algorithms. Alternatively, according to one aspect of the present invention, an information processing device can be provided that can generate source sentence candidate sentences that are easy to translate without requiring specialized knowledge of language and machine translation algorithms.

[0143] 10: Information processing device, 20a: Desktop computer, 20b: Notebook computer, 20c: Smartphone, 20d: Tablet computer, 20: Information terminal, 21: Housing, 30: Network, 31: Network, 40: Information processing device, 110: Input unit, 120: Storage unit, 130: Processing unit, 140: Output unit, 150: Transmission path

Claims

A first information processing device; A second information processing device, The first information processing device, accepting input of a document in a first language; a function for translating the document in the first language into a second language; back-translating the second language document into the first language; The ability to compare two documents in the same language and determine whether they are synonymous; a function of acquiring document B by transmitting a prompt including document A in the first language determined to be non-synonymous in the determination and an instruction for generating document B in the first language and having the same meaning to document A to the second information processing device; a function of outputting the document B; The second information processing device has a function of generating the document B in accordance with the prompt. 　 In claim 1, When the first information processing device determines that a document C obtained by translating the document B into the second language and back-translating it into the first language is synonymous with the document B, the first information processing device outputs the document B. Information processing system. 　 A first information processing device; A second information processing device, the first information processing device has a function of accepting input of a document A in a first language; the first information processing device has a function of translating the document A into a document B in a second language; the first information processing device has a function of reverse-translating the document B into the document C in the first language; The first information processing device has a function of comparing the document A with the document C and determining whether they are synonymous as a first determination, the first information processing device has a function of transmitting, when the first determination is not synonymous, a prompt including the document A and an instruction for generating a document D1 synonymous with the document A to the second information processing device; the second information processing device has a function of generating the document D1 in the first language in accordance with the prompt; the first information processing device has a function of translating the document D1 into a document E1 in the second language; the first information processing device has a function of reverse-translating the document E1 into a document F1 in the first language; The first information processing device has a function of comparing the document D1 with the document F1 and determining whether they are synonymous as a second determination, the first information processing device has a function of outputting the document D1 when the second determination is synonymous; Information processing system.

4. The information processing system according to claim 3, The first information processing device has a function of outputting a message indicating that the document A does not include any part that is difficult to translate when the first determination is "synonymous." Information processing system.

4. The information processing system according to claim 3, the first information processing device has a function of transmitting, when the second determination is not synonymous, a prompt including the document A and an instruction for generating a document D2 synonymous with the document A to the second information processing device; the second information processing device has a function of generating the document D2 in the first language in accordance with the prompt; the first information processing device has a function of translating the document D2 into a document E2 in the second language; the first information processing device has a function of reverse-translating the document E2 into a document F2 in the first language; The first information processing apparatus has a function of comparing the document D2 with the document F2 and determining whether they are synonymous as a third determination, the first information processing device has a function of outputting the document D2 when the second determination is synonymous; Information processing system. 　 An information processing system according to any one of claims 3 to 5, The second information processing device includes a large-scale language model; The document D1 is generated by the large-scale language model. Information processing system. 　 (A) accepting a document A in a first language; (B) translating the document A into a second language to generate a document B; (C) back-translating the document B into the first language to generate a document C; (D) as a first determination, a step of comparing the document A with the document C and determining whether they are synonymous; (E) if the first determination is not synonymous, generating at least one document D that is synonymous with the document A and is in the first language using a large-scale language model; (F) translating the document D into the second language to generate a document E; (G) back-translating said document E into said first language to generate document F; (H) a step of comparing the document D with the document F as a second determination as to whether they are synonymous; (I) outputting the document D if the second determination is that the two documents are synonymous; An information processing method comprising the steps of: 　 (A) accepting a document A in a first language; (B) using a large-scale language model to translate document A into a second language to generate document B; (C) causing the large-scale language model to back-translate the document B into the first language to generate a document C; (D) causing the large-scale language model to compare the document A with the document C as a first determination as to whether they are synonymous; (E) causing the large-scale language model to generate at least one document D that is synonymous with the document A and is in the first language if the first determination is not synonymous; (F) causing the large-scale language model to translate the document D into the second language to generate a document E; (G) causing the large-scale language model to back-translate the document E into the first language to generate a document F; (H) a step of causing the large-scale language model to compare the document D with the document F as a second determination as to whether they are synonymous; (I) outputting the document D if the second determination is that the two documents are synonymous; An information processing method comprising the steps of:

Images