Method for evaluating sentences

Abstract

To provide a method for evaluating sentences made by a computer capable of evaluating contexts.SOLUTION: The method for evaluating sentences made by a computer includes: a sentence input step in which a computer receives input of a sentence to evaluate; a sentence word extraction step in which the computer extracts sentence words, which are words in the sentence to evaluate; a key sentence extraction step in which the computer extracts a key sentence in the sentence to evaluate, the key sentence including at least one sentence word; and a sentence evaluation step in which the computer evaluates the sentence on the basis of the key sentence.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] This invention relates to a method and a system for evaluating sentences by a computer.

Background Art

[0002] Japanese Patent No. 7049010 describes a presentation evaluation system.

[0003] This evaluation system evaluates the content of a conversation or the person who had the conversation based on the number of keywords, the number of related words, combinations of keywords, or combinations of related words. In this case, there was a problem that even if the conversation was incorrect in terms of context, it would be evaluated as correct if keywords or the like were included in the conversation.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] An object of this invention is to provide a method for evaluating sentences by a computer that can also evaluate context.

[0006] An object of this invention is to provide a learning support method using the above evaluation method.

[0007] An object of this invention is to provide a method capable of obtaining a subsequent sentence of a certain sentence so that role-playing can be performed after evaluating the context.

Means for Solving the Problems

[0008] This method basically relates to a method for evaluating a sentence that can correctly grasp the context by evaluating the sentence using a key sentence including a particle. This method is a method in which a computer evaluates a sentence based on or based on a key sentence.

[0009] Examples of this method include a sentence input step (S101), a sentence word extraction step (S102), a key sentence extraction step (S103), and a sentence evaluation step (S104). The sentence input step (S101) is a step in which a sentence to be evaluated is input to a computer. The sentence word extraction step (S102) is a step in which a computer extracts sentence words that are words included in the sentence to be evaluated. The key sentence extraction step (S103) is a step in which a computer extracts a key sentence included in the sentence to be evaluated. The key sentence includes one or more sentence words and one or more particles. The sentence evaluation step (S104) is a step in which a computer evaluates a sentence based on or based on a key sentence.

[0010] Examples of the sentence input step (S101) include a step in which voice is input to a computer and a step in which the computer analyzes the voice input to the computer to obtain a sentence to be evaluated.

[0011] Examples of the sentence to be evaluated are an explanatory sentence or an answer sentence.

[0012] Examples of the sentence words are words stored in a sentence word storage unit that stores sentence words related to the sentence to be evaluated.

[0013] Examples of the key sentences are those stored in a key sentence storage unit that stores key sentences related to the sentence to be evaluated.

[0014] Examples of key sentences may be groups of words that exist at consecutive positions or at non - consecutive positions in the sentence to be evaluated.

[0015] An example of using the above - mentioned method is a method for assisting learning by a computer.

[0016] Another example of using the above - mentioned method relates to a method for creating a subsequent sentence by a computer. This method further includes a subsequent - sentence creation step (S105). The subsequent - sentence creation step (S105) is a step of obtaining a subsequent sentence, which is a sentence following the sentence to be evaluated, based on the evaluation. The computer has a subsequent - sentence storage unit that stores subsequent sentences corresponding to the evaluations. And the subsequent - sentence creation step (S105) is a step in which the computer reads out a subsequent sentence corresponding to the evaluation using the evaluation of the sentence to be evaluated.

[0017] This specification also discloses a system for evaluating sentences by a computer. This system 1 is a system in which a computer evaluates sentences. This system 1 includes a sentence input unit 3, a sentence - word extraction unit 5, a key - sentence extraction unit 7, and a sentence evaluation unit 9. The sentence input unit 3 is an element for inputting a sentence to be evaluated. The sentence - word extraction unit 5 is an element for extracting sentence words, which are words included in the sentence to be evaluated. The key - sentence extraction unit 7 is an element for extracting key sentences included in the sentence to be evaluated. A key sentence includes one or more sentence words and one or more auxiliary words. The sentence evaluation unit 9 is an element for evaluating a sentence based on the key sentence.

[0018] This specification also discloses a learning support system using the above - described system.

[0019] This specification also discloses a system for creating a subsequent sentence using the above - described system. The system for creating a subsequent sentence has a subsequent - sentence creation unit.

[0020] This specification also discloses a program for causing a computer to execute the above-described method or for causing the computer to function as the above-described system, and a non-transitory information recording medium readable by a computer on which such a program is recorded.

Advantages of the Invention

[0021] This invention can provide a method for a computer to evaluate a sentence that can also evaluate the context.

[0022] This invention can provide a learning support method using the above-described evaluation method.

[0023] This invention can provide a method for obtaining a subsequent sentence of a certain sentence so that role-play can be performed after evaluating the context.

Brief Description of the Drawings

[0024]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Best Mode for Carrying Out the Invention

[0025] Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. The present invention is not limited to the embodiments described below, and also includes those appropriately modified by those skilled in the art within an obvious range from the following embodiments.

[0026] Basically, this method relates to a method of evaluating a sentence that can correctly grasp the context by evaluating the sentence using a key sentence including a particle. In this method, a computer evaluates a sentence based on or based on a key sentence.

[0027] "Evaluating a sentence" may be any of interpreting whether the sentence is a correct answer, interpreting to which category the sentence belongs, and determining the degree of goodness of the sentence.

[0028] FIG. 1 is a flowchart for explaining a method of evaluating a sentence using a computer. As shown in FIG. 1, an example of this method includes a sentence input step (S101), a word extraction step in a sentence (S102), a key sentence extraction step (S103), and a sentence evaluation step (S104). This method may further include a subsequent sentence creation step (S105). S means step (process).

[0029] FIG. 2 is a block diagram showing a configuration example of a system for evaluating sentences by a computer. As shown in FIG. 2, this system 1 includes a sentence input unit 3, a sentence word extraction unit 5, a key sentence extraction unit 7, and a sentence evaluation unit 9. This system may further include any one or two or more of a sentence word storage unit 6, a key sentence storage unit 8, and a sentence evaluation information storage unit 10. This system 1 may further include a subsequent sentence creation unit 11. When this system 1 has a subsequent sentence creation unit 11, this system 1 may further include a subsequent sentence storage unit 13. The sentence input unit 3 is an element for inputting a sentence to be evaluated. The sentence word extraction unit 5 is an element for extracting sentence words that are words included in the sentence to be evaluated. The key sentence extraction unit 7 is an element for extracting key sentences included in the sentence to be evaluated. The sentence evaluation unit 9 is an element for evaluating a sentence based on or based on the key sentence. The subsequent sentence creation unit 11 is an element for obtaining a subsequent sentence that is a sentence following the sentence to be evaluated. This system 1 is a computer-based system for executing the above-described method.

[0030] A computer has an input unit, an output unit, a control unit, an arithmetic unit, and a storage unit, and each element is connected by a bus or the like so that information can be exchanged. For example, a control program may be stored in the storage unit, or various types of information may be stored. When predetermined information is input from the input unit, the control unit reads out the control program stored in the storage unit. Then, the control unit appropriately reads out the information stored in the storage unit and transmits it to the arithmetic unit. Also, the control unit appropriately transmits the input information to the arithmetic unit. The arithmetic unit performs arithmetic processing using the received various types of information and stores it in the storage unit. The control unit reads out the arithmetic result stored in the storage unit and outputs it from the output unit. In this way, various processes and steps are executed. The ones that execute these various processes are each unit and each means. The computer may have a processor, and the processor may realize various functions and various steps. The computer may be stand-alone. Part of the functions of the computer may be distributed between a server and a terminal. In that case, it is preferable that the server and the terminal can exchange information via a network such as the Internet or an intranet.

[0031] The sentence input step (S101) is a step in which a sentence to be evaluated is input to the computer. The sentence input unit 3 in system 1 may input the sentence to be evaluated into the system. For example, a sentence may be input to the computer using an input device (e.g., a keyboard or a mouse). Also, the voice that is the basis of the sentence may be input to the computer using a voice input device such as a microphone. The voice input to the computer may be subjected to voice recognition by a known method to obtain the sentence to be evaluated. In this case, the computer performs arithmetic processing for digitizing the voice and appropriately stores the digitized voice in the storage unit. And the computer may have a voice analysis unit that analyzes the words and terms included in the voice. The voice analysis unit can read out the voice analysis program and the digitized voice from the storage unit, analyze the voice, and obtain the sentence to be evaluated. The obtained sentence to be evaluated may be appropriately stored in the storage unit.

[0032] When a sentence to be evaluated is input to a computer, if the sentence is related to something, information about what the sentence is related to may also be input to the computer. Then, this system can perform various arithmetic operations using various information related to what the sentence is related to. Examples of arithmetic operations are various operations in speech analysis, word extraction, sentence extraction, and subsequent sentence analysis. Examples of what the sentence is related to are presentation materials, pages of presentation materials, reports, meeting materials, pharmaceuticals, questions, problems, questionnaires, Q&A collections, telephone response records, and chatbot manuals. For example, when presentation materials are launched on a certain terminal, information related to the presentation materials may be input to the system. In this case, as will be described later, a dictionary or a storage unit related to the presentation materials can be used. The sentence to be evaluated may be any sentence. Examples of the sentence to be evaluated are explanatory sentences or answer sentences. That is, this system can be used to evaluate a sentence when a person explains something to another person or to automatically evaluate an answer sentence related to a certain question. Also, this system can automatically interpret the input question and effectively create an answer to it.

[0033] For example, a certain MR (Medical Representative / pharmaceutical information officer) inputs information about a certain pharmaceutical A into the system or launches explanatory materials about a certain pharmaceutical A and displays the explanatory materials on a display unit. Then, the system stores in a storage unit information regarding that subsequent utterances and conversations are related to pharmaceutical A. Then, the voice of the MR, "Please provide guidance when it is suspected that a pregnant woman or a woman who may be pregnant is at risk of uterine inertia during delivery," is input to the microphone.

[0034] The system analyzes the terms of the above voice with reference to a term conversion dictionary for Medicine A (a memory unit for conversion terms related to Medicine A). In the term conversion dictionary for Medicine A, words including "pregnant woman", "pregnant", "possible", "female", "treatment", "therapeutically", "benefit", "risk", "exceed", "judgment", "case", and "administration" are stored as words with high priority. For example, since "female" is stored as a word with higher priority than "assist" in this dictionary, for "josei" in the above input voice, "female" is preferentially read out.

[0035] In this way, the sentence "Please administer to pregnant women or women who may be pregnant when it is judged that the therapeutic benefit exceeds the risk" is input to the system for Medicine A.

[0036] The in - sentence word extraction step (S102) is a step in which a computer extracts in - sentence words that are words included in the sentence to be evaluated. The in - sentence word extraction unit 5 extracts in - sentence words that are words included in the sentence to be evaluated. The system has a dictionary related to in - sentence words. The system may read the sentence to be evaluated from the memory unit and compare it with the words in the dictionary to extract in - sentence words from the sentence to be evaluated. Also, when information related to things related to the sentence is input to the computer, by using a dictionary related to things related to the sentence, which is a dictionary for extracting in - sentence words (in - sentence word memory unit 6), the in - sentence words can be easily extracted. In this case, examples of in - sentence words are words stored in the in - sentence word memory unit that stores in - sentence words related to the sentence to be evaluated. The system (in - sentence word extraction unit 5) can extract in - sentence words included in the sentence to be evaluated by reading the in - sentence words stored in the in - sentence word memory unit 6 and comparing them with the sentence to be evaluated. The in - sentence words extracted by the system may be appropriately stored in the memory unit by the system. The system can obtain in - sentence words in this way.

[0037] The system has a sentence word memory unit 6 regarding Medicine A. In the sentence word memory unit 6 regarding Medicine A, words including "pregnant woman", "pregnancy", "female", "benefit", "risk", "exceed", and "administration" are stored as sentence words with high priority.

[0038] Therefore, from the sentence "Please administer when it is determined that the therapeutic benefit exceeds the risk for pregnant women or women who may be pregnant regarding Medicine A" input into the system regarding Medicine A, the sentence words "pregnant woman", "pregnancy", "female", "benefit", "risk", "exceed", and "administration" are extracted. When voice is input into the system, the term conversion dictionary and the sentence word memory unit 6 may be the same, but it is preferably different. That is, the terms used for term conversion and the words extracted as sentence words may be the same, but they may also be different because their purposes are different.

[0039] The key sentence extraction step (S103) is a step in which a computer extracts key sentences included in a sentence to be evaluated. For example, the key sentence extraction unit 7 extracts key sentences included in a sentence to be evaluated. A key sentence includes one or more sentence words and one or more auxiliary words. A key sentence may include one or more sentence words. And when two or more sentence words are included in one key sentence, those sentence words may be sentence words that continuously exist in the sentence to be evaluated, or one or more other sentence words or nouns may exist before those sentence words appear.

[0040] In the above sentence word extraction step (S102), the words included in the sentence to be evaluated were extracted. However, just using words alone may not be able to determine whether it is a correct sentence (explanation sentence or answer sentence). Therefore, this system 1 extracts key sentences through the key sentence extraction step (S103). A key sentence means a clause used to evaluate whether a sentence is a correct explanation sentence or a correct answer sentence. A key sentence usually contains two or more words. The key sentence may be stored in the key sentence storage unit 8 that stores key sentences related to the sentence to be evaluated, or may be extracted from the sentence to be evaluated as including a word A in a certain sentence and a particle related to the word A included in the sentence to be evaluated. Note that the sentence word extraction step (S102) and the key sentence extraction step (S103) are for convenience, and they may be performed simultaneously. In this case, sentence words are also extracted when performing the key sentence extraction step (S103).

[0041] In the former case, the system (key sentence extraction unit 7) reads one or more key sentences stored in the key sentence storage unit 8 in association with the sentence to be evaluated, performs an operation of comparing with the sentence to be evaluated, and may extract the key sentences included in the sentence to be evaluated. The system may appropriately store the extracted key sentences included in the sentence to be evaluated in the storage unit.

[0042] In the latter case, the system (key sentence extraction unit 7) stores, for example, the sentence words for the key sentences stored in the key sentence storage unit 8 in association with the sentence to be evaluated. Then, the system may read out the sentence words for the key sentences from the key sentence storage unit 8, and at the same time read out the sentence words for the key sentences and the subsequent particles from the sentence to be evaluated, and extract the key sentences from the sentence to be evaluated. The system may appropriately store the extracted key sentences included in the sentence to be evaluated in the storage unit.

[0043] For example, Medicine A is administered to pregnant women or women who may be pregnant when the benefits outweigh the risks. However, just extracting the words in the text, the sentence to be evaluated may also be a sentence stating that the risks outweigh the benefits. Such a sentence is not correct as an explanatory sentence regarding Medicine A. Therefore, the key sentence storage unit 8 stores the key sentence "(The therapeutic benefit is... greater than the risk)". Also, the key sentence storage unit 8 may store the key sentences "(The benefit is greater)", "(The risk is lower)", and "(The risk is low)". The key sentence storage unit 8 may store key sentences including any of the above for the administration target with respect to the in-text words "pregnant women" and "pregnancy" and for the drug efficacy.

[0044] Therefore, from the sentence "Please administer to pregnant women or women who may be pregnant when it is determined that the therapeutic benefit outweighs the risk", for example, as the subject, (either or both of "pregnant women" and "pregnancy") and for the drug efficacy, the key sentences "(The therapeutic benefit is)" and "(greater than the risk)" are extracted.

[0045] The sentence evaluation step (S104) is a step in which a computer evaluates a sentence based on a key sentence. For example, the sentence evaluation unit 9 evaluates the sentence to be evaluated based on the key sentence. The system 1 may further include a sentence evaluation information storage unit 10. The sentence evaluation information storage unit 10 stores, for example, a key sentence and evaluation information regarding the key sentence. Examples of the evaluation information are information regarding whether the key sentence is correct as a sentence to be evaluated, information regarding which category the key sentence belongs to, and an evaluation value associated with the key sentence. The system (sentence evaluation unit 9) reads out the evaluation regarding the key sentence extracted from the sentence evaluation information storage unit 10 using the key sentence extracted in the key sentence extraction step (S103). In this way, the system 1 can evaluate the sentence to be evaluated.

[0046] For example, the sentence evaluation information storage unit 10 associates with a key sentence including (either or both of "pregnant woman" and "pregnancy"), (the therapeutic benefit is), and (exceeds the risk) in relation to Medicine A, and stores "correct" as an evaluation. Then, the sentence evaluation unit 9 reads out the evaluation of "correct" from the sentence evaluation information storage unit 10 using the above key sentence. The sentence evaluation unit 9 may store the read evaluation of "correct" in the storage unit as the evaluation of the sentence to be evaluated. Moreover, the system 1 may output the evaluation of "correct". Further, assume that a category of "administration prescription" is stored in the sentence evaluation information storage unit 10 in association with a key sentence including (either or both of "pregnant woman" and "pregnancy"), (the therapeutic benefit is), and (exceeds the risk). Then, the system 1 reads out the category of "administration prescription" as one of the evaluations in relation to Medicine A using the extracted key sentence. In this case, the system 1 may obtain an evaluation of "correct" for the category of "administration prescription" in relation to Medicine A.

[0047] The in-text word memory unit 6 may store the classification of words in relation to the in-text words. Examples of classification are affirmative, negative, and ambiguous. Table 1 shows the classification and examples of in-text words for each classification. Also, when an in-text word is included in the key sentence, the key sentence memory unit 8 may store the classification related to that in-text word. The sentence evaluation information memory unit 10 may store the classification of the in-text words included in the key sentence. The system (sentence evaluation unit 9) reads out the evaluation related to the key sentence extracted from the sentence evaluation information memory unit 10 using the key sentence extracted in the key sentence extraction step (S103). For example, the system 1 uses the extracted key sentence to obtain information (e.g., affirmative, negative, and ambiguous) related to the classification of the in-text words included in that key sentence from the sentence evaluation information memory unit 10 (key sentence memory unit 8, or in-text word memory unit 6). And when the classification of the in-text words included in the read key sentence is negative (or ambiguous), the evaluation of the evaluation target sentence may be evaluated as incorrect (inappropriate). By doing so, it is possible to prevent the situation of making ambiguous sentence expressions and improve the answering ability.

[0048] Note that an invention that extracts in-text words from the sentence to be evaluated instead of the key sentence and evaluates the sentence to be evaluated based on the classification of the extracted in-text words is also the invention described in this specification. In this invention, the key sentence extraction step (S103) and the key sentence extraction unit 7 become unnecessary.

[0049]

Table 1

[0050] The subsequent sentence creation process (S105) is a process of obtaining a subsequent sentence that is a sentence following the sentence to be evaluated based on the evaluation. The subsequent sentence creation unit 11 may obtain a subsequent sentence that is a sentence following the sentence to be evaluated. The computer has, for example, a subsequent sentence storage unit 13 that stores subsequent sentences corresponding to the evaluation. Then, in the subsequent sentence creation process (S105), the computer reads out a subsequent sentence corresponding to the evaluation using the evaluation of the sentence to be evaluated. Information regarding what the sentence is related to is input to the computer, and an appropriate subsequent sentence may be read out from the subsequent sentence storage unit 13 corresponding to the information regarding what the sentence is related to. For example, when the information regarding what the sentence is related to is "a certain question" and the evaluation is "correct" (correct answer), the subsequent sentence storage unit 13 stores an evaluation such as "Well done." and an explanation regarding that question. The system (subsequent sentence creation unit 11) reads out the evaluation from the storage unit and reads out from the subsequent sentence storage unit 13 the evaluation "Well done.", which is a subsequent sentence corresponding to that evaluation, and an explanation regarding that question, and stores them in the storage unit as appropriate. The read-out subsequent sentence may be output as appropriate. When the information regarding what the sentence is related to is a chatbot regarding a certain product and the evaluation is "problem identification" (category), based on that evaluation, an answer to the sentence to be evaluated may be read out from the subsequent sentence storage unit 13 to obtain an appropriate answer. In this way, the system can obtain a subsequent sentence. For example, when the evaluation of the sentence is "interrogative sentence" or "question", an answer to the sentence can be obtained.

[0051] For System 1 regarding Medicine A, an input is made stating "Please administer when it is determined that the therapeutic benefit outweighs the risk for pregnant women or women who may be pregnant." System 1 extracts words and key sentences from this sentence and, in relation to Medicine A, obtains an evaluation of "correct" for the category of "administration prescription". For example, in the subsequent sentence memory unit 13, the subsequent sentence "The administration frequency of Medicine A is one tablet per day." is stored in relation to the above evaluation of "correct" for the category of "administration prescription" in relation to Medicine A. The system (subsequent sentence creation unit 11) may read the above evaluation from the memory unit, read the above subsequent sentence from the subsequent sentence memory unit 13, and store it in the memory unit. Also, the system may output the above subsequent sentence. Then, as the subsequent sentence of the sentence, "The administration frequency of Medicine A is one tablet per day." will be output.

[0052] Learning support by computer The usage examples of the above method and system are a computer-assisted learning method. This specification also discloses a learning support system using the above system. The invention in this aspect may be provided in a form that can be downloaded as a learning support application. In this case, various dictionaries for each question are stored in the server or installed in the mobile terminal. For example, in this system, a certain question A is stored from the storage unit and displayed on the display unit of the terminal. The user answers the question by voice. The user's voice is input into the terminal via the input unit of the terminal. The input voice (the user's answer) is stored in the terminal as a sentence to be evaluated. The system including either or both of the terminal and the server has any one or two or more of a term conversion dictionary, a sentence word storage unit 6, a key sentence storage unit 8, a sentence evaluation information storage unit 10, and a subsequent sentence storage unit 13 for the question A. Therefore, the system can evaluate the user's answer. In addition, the system can read out a subsequent sentence corresponding to the user's answer and display it on the display unit of the terminal. As will be described later, storing using sentences rather than simply storing words can not only fix the memory but also improve the communication ability. Therefore, the learning support system and learning support method using this invention can give a high learning effect to the user.

[0053] Chatbots and role-playing systems The usage example of the above system is a chatbot or a role-playing system. Chatbot and role-playing systems have one or more of a term conversion dictionary for a certain product or service, etc., a sentence word memory unit 6, a key sentence memory unit 8, a sentence evaluation information memory unit 10, and a subsequent sentence memory unit 13. Then, when there is an input to the system regarding a voice input by phone or a question automatically input via the Internet, etc., the system can request a key sentence, obtain an evaluation regarding the product or service, etc., and appropriately obtain a subsequent sentence. Then, an appropriate answer can be automatically output for the question. In this case, for example, in response to an input such as "The wheels of product AA have stopped moving.", the system may use the memory unit regarding product AA to obtain a subsequent sentence such as "Is the power on?" and output it to the questioner. Also, when the user inputs "Yes" in response to this answer, the system may obtain a subsequent sentence such as "Please try cleaning around the wheels with a brush. If the wheels still don't move, if you can inform the service center, we will come to collect it." and output it to the questioner.

[0054] Conversation or presentation support system The usage example of the above system is a conversation or presentation support system. This system has, for example, a term conversion dictionary for explanatory materials, presentation materials, each page of presentation materials, a sentence word memory unit 6, a key sentence memory unit 8, a sentence evaluation information memory unit 10, and a subsequent sentence memory unit 13, either one or two or more of them. Then, when a conversation or presentation is conducted based on the materials or presentation materials, a key sentence can be obtained, and using that key sentence, an appropriate evaluation can be obtained from the sentence evaluation information memory unit 10 and an appropriate subsequent sentence can be obtained from the subsequent sentence memory unit 13. In that case, the evaluation may be something like "Very good", "Excellent", "Good". An example of a subsequent sentence may be an explanation that leads to an improvement in the conversation or presentation, such as "Using BBB instead of AAA for the explanation will result in a higher evaluation."

[0055] This specification also discloses a program for causing a computer to execute the above-described method and for causing the computer to function as the above-described system, and a non-transitory information recording medium (such as a CD-ROM, DVD, SD card, and USB memory) readable by a computer on which such a program is recorded.

Example

[0056] Consideration of a learning model using speech learning It is considered that an easy-to-understand explanation can be made through three steps: "words and explanations are fixed in memory", "the fixed memory is replaced with one's own words", and "an easy-to-understand explanation can be given". In this example, the verification results regarding "words and explanations are fixed in memory" and "the fixed memory is replaced with one's own words" are considered. In this study, the following three verification items were verified. In the following experiment, an application implementing the computer-based learning support system in this specification was used as a role-playing app. The outline is shown in Figure 3.

[0057] (i) Verification of the degree of memory fixation in rote word learning (ii) Measurement of the fixation rate of word explanations using speech learning (Regarding pharmaceutical-related questions, 10 men and women in their 20s to 50s were selected as subjects. They were divided into two groups and the same content was implemented.) (iii) Measurement of the fixation rate and learning time in an iPad (registered trademark) role-playing app (Regarding network-related questions, 6 men and women in their 30s to 50s were selected as subjects and the experiment was carried out.)

[0058] Verification of the degree of memory fixation in rote word learning First, in order to verify how much rote memorization of words is possible through silent reading, 21 technical terms related to the pharmaceutical industry, which are words not commonly used on a daily basis, were picked up and questions were created. An overview of the corresponding words and explanations is shown in Table 2.

[0059]

Table 2

[0060] On the first day of the validation experiment, the subjects were given an A4 sheet with words and explanations written on it, and they were asked to memorize the words in a way other than by vocalizing them (e.g., by visual inspection, writing on paper, etc.) for 10 minutes on the spot. After 10 minutes, they were asked questions about the explanations of the words in the form of interrogative sentences and read out to the subjects. In case of incorrect answers, the correct answers were verbally provided as feedback to the subjects on the spot. On the second and third days, only tests were conducted to observe how the fixation of memory changed and to tabulate the number of correct answers of the subjects.

[0061] Measurement of the fixation rate of word explanations using speech learning After the three-day word test was completed, a test was conducted to measure how well the subjects could recall the explanations of the words. As the content of the test, 21 words were verbally asked of the subjects as "Please explain ○○" and the subjects' answers were recorded (using the voice recognition function of iPad (registered trademark) for work efficiency). Thereafter, as shown in Table 3, the same test was repeatedly conducted at regular intervals over a total of 19 days (please note that due to business reasons, the implementation schedule is not evenly distributed), and the results were recorded to measure the memory fixation rate of the word explanations.

[0062]

Table 3

[0063] Also, in order to achieve the memory fixation effect by speech learning, the subjects were given three speech learning sessions at intervals of approximately 3 to 4 days from the first day of measurement. The content of the speech learning was the same as in the above verification, where an A4 sheet with the words and explanations shown in Table 1 was given, and the subjects were asked to memorize the word explanations by vocalizing them for 10 minutes on the spot. The "memory test" in Table 2 means a test of how well the word explanations can be spoken, and the "confirmation test" refers to a test to confirm how well the word explanations can be spoken immediately after the speech learning.

[0064] Measurement of Fixation Rate and Learning Time in iPad (Registered Trademark) Role-Playing Apps Using the iPhone (Registered Trademark) and iPad (Registered Trademark) app "iRolePlay (Registered Trademark)" developed by Interactive Solutions (Registered Trademark), we measured the memorization rate and learning time of words and explanatory texts. "iRolePlay (Registered Trademark)" is a new era reskilling tool that can "acquire explanatory power and proposal power" through an objective analysis and evaluation & advice function of AI that utilizes voice recognition and the Neural Engine installed in iPhone (Registered Trademark) and iPad (Registered Trademark) for the user's speech content. It consists of two types of modes (learning mode / challenge mode). In the former, answers to questions read aloud are displayed and speech practice can be done (Figure 4). In the latter, practical explanatory power can be acquired through speech practice of answers to questions read aloud.

[0065] The target questions were 21 words related to unfamiliar networks. Two courses, one for word learning and one for learning word explanations, were prepared, and iPads (Registered Trademark) were distributed to the subjects. Self-study during the gaps between work was promoted for three days. As shown in Figure 4, the subjects learned in a role-playing format where the question text and answers were displayed on the iPad (Registered Trademark) in the learning mode. In this app, one can proceed to the next question by correctly speaking the correct word or explanatory text. The test method used the challenge mode where questions asking for words or turning word explanations into interrogative sentences were presented by voice, and it was determined by voice recognition whether the answers to the questions were correct.

[0066] Experimental Results and Discussion Verification of the Memory Fixation Degree of Word Memorization Learning As mentioned above, it is effective to speak out words for memorization learning. Even with methods such as visual inspection and writing on paper for 10 minutes of memory, the correct answer rate was 78.6%. Also, the correct answer rate on the second day without a dedicated memorization time was 81.9%, and on the third day, it was 87.6%, showing an upward trend (Figure 5).

[0067] Discussion Generally, as is known as "Ebbinghaus Forgetting Curve", memory is said to be forgotten over time. However, in this verification, the fact that the subjects were taught the correct answers to the wrong answers on the spot also made it easier to remain in memory, and as a result, it is considered to be a factor in the increase in test scores. In fact, in the interviews with the subjects, three people reported that the correct answers conveyed for the questions they got wrong in the previous test remained in their memory. In word memory, it has been shown that immediate feedback of the correct answer for incorrect answers is effective for memory consolidation.

[0068] Measurement of the fixation rate of word explanations using speech learning The correct answer rate for the word test was 87.6%, while the correct answer rate for the word explanation test was 43.3%, less than half. This result proved that even if one can answer a word in a question-and-answer format, it does not mean that one can explain the word.

[0069] Now, let's present the results regarding the memory of word explanations. In the word explanation test, when comparing the results of visual learning and speech learning, the latter had a 64% higher correct answer rate than the former. Also, the transition of the correct answer rate in the word explanation test, that is, the transition of the memory fixation rate, is as shown in Figure 6. By performing speech learning regularly, the correct answer rate over the entire schedule has been on an upward trend. Even after 19 days have passed since the first explanation test, the response rate exceeds 70%.

[0070] Discussion By repeating the speech learning, the subject was able to speak an answer close to the example answer (the explanatory example shown in Figure 3). When analyzing the answer sentences recorded by the subject's speech recognition, it was found that the proportion of technical terms in the answer sentences increased by repeating the speech learning. In the hearing of the subject, there were voices saying that even technical terms not usually used were effective when speaking them out or taking in the voice through the ears when answering out loud. As the number of tests was repeated, the subject was able to speak smoothly with technical terms, and at the same time, the correct answer rate gradually increased. To cultivate explanatory power, it is necessary to speak to the other person using appropriate expressions and sometimes technical terms. In that regard, this verification is considered to have enabled the technical terms to be fixed in memory by speaking, and to be able to explain smoothly as one's own words because one got used to speaking the words. Also, it was clarified in this verification that the explanatory text once memorized is likely to become long-term memory. Looking at the answer results of 21 questions each, the questions that could be answered once did not show a decrease in the correct answer rate even after the passage of days. In the hearing of the subject, two people gave the impression that they could correctly explain during the word explanation test by memorizing the words that are keywords in the explanatory text.

[0071] Measurement of the fixation rate and learning time in the iPad (registered trademark) role-playing app Although the verification using the iPad (registered trademark) role-playing app is still in progress, the transition of the correct answer rate of the word explanatory text shows a similar trend. The biggest advantage is that the time taken for measurement as an educator is only the distribution of the app. In the above verification, it takes 2480 minutes to improve the scores of 10 people considering all of the test question posing, the feedback of the correct answer for wrong answers, and the scoring of the test results. Also, it is an advantage that the subject can learn at any time during the gap time. Since it is possible to obtain the app's log, it is possible to accurately measure the learning time, which is presumed to lead to advantageous learning guidance as an educator. How effectively the subject can utilize the gap time to increase the correct answer rate of the word explanatory text and shorten the period is an issue to be verified in the future.

[0072] In this study, explanatory power was defined as "the ability to explain to others using appropriate expressions," and it was considered that there are three steps to establish explanatory power. In the verification of the degree of memory fixation in rote learning, it has been shown that immediate feedback of the correct answer for incorrect answers is effective for memory fixation. On the other hand, it has been proven that even if one can answer a word in a question-and-answer format, one may not be able to explain that word. In the measurement of the fixation rate of word explanations using speech learning, it has been clarified that repeating speech learning improves the fixation rate of word explanations. Also, it has been shown that by following three steps based on the content of the word explanation test of the subjects, the correct answer rate increases.

Example

[0073] Role-play Figure 7 is a conceptual diagram explaining the concept of role-play. Among them, "parentheses" are categories for evaluation.

[0074] Figure 8 is a conceptual diagram showing an example of role-play in which the response of the AI doctor branches depending on the answer of the MR.

[0075] The actually developed application was installed on a computer, and commands based on the application were executed on the computer. The following Q is the sentence of the evaluation target input, and A is the subsequent sentence for it. Q What about the patient's medical history? A Had a myocardial infarction three years ago Q What about other symptoms? A Seems to have headache and dizziness Q What were the findings before administration? A It was thought to be stress-related Q What is the current treatment method? A Dietary therapy and taking antihypertensive drugs Q What is the relationship with EyePro tablets? A Since EyePro tablets have few side effects on the respiratory system, they are easy to prescribe

[0076] In this example, when a doctor enters a question, first, a term conversion dictionary related to the medical field and the target patient, a sentence word memory unit 6, a key sentence memory unit 8, a sentence evaluation information memory unit 10, and a subsequent sentence memory unit 13 are referred to, and an appropriate subsequent sentence is displayed on the display unit. During this process, "EyePro tablets" appeared as a sentence word. For this reason, in this system, the key sentence memory unit 8, the sentence evaluation information memory unit 10, and the subsequent sentence memory unit 13 related to EyePro tablets are referred to, and an appropriate subsequent sentence is obtained. In order to perform key sentence analysis, this system makes chatbots and role-playing become interactive, and it becomes possible to have a natural conversation even using a computer. Also, different from conventional conversation systems using deep learning, it becomes possible to construct a system for conversations in a narrow area very simply and without spending time. Role-playing using this system was effective for cultivating advanced conversation skills.

Industrial Applicability

[0077] This invention can be used in the information industry, the education industry, etc.

Explanation of Signs

[0078] 1 System 3 Sentence Input Unit 5 Sentence Word Extraction Unit 6 Sentence Word Memory Unit 7 Key Sentence Extraction Unit 8 Key Sentence Memory Unit 9 Sentence Evaluation Unit 10 Sentence Evaluation Information Memory Unit 11 Subsequent Sentence Creation Unit 13 Subsequent Sentence Memory Unit

Claims

1. A method for evaluating a sentence by a computer, comprising: a sentence input step of inputting a sentence to be evaluated into the computer; a sentence word extraction step of extracting a sentence word, which is a word included in the sentence to be evaluated, by the computer; a key sentence extraction step of extracting a key sentence included in the sentence to be evaluated by the computer, where the key sentence includes one or more of the sentence words and one or more auxiliary words; a sentence evaluation step of evaluating the sentence by the computer based on the key sentence; a subsequent sentence creation step of obtaining a subsequent sentence corresponding to the evaluation of the sentence, where the subsequent sentence is a sentence following the sentence; The method comprising the above steps.

2. The method according to claim 1, wherein the sentence evaluation step includes a step of reading evaluation information regarding the key sentence from a storage unit, and the evaluation information includes information regarding whether the key sentence is correct as a sentence to be evaluated or information regarding to which category the key sentence belongs.

3. The method according to claim 2, further comprising a step of obtaining an explanation regarding a certain question when the sentence is an answer to the certain question.

4. The method according to claim 2, wherein the sentence input step includes a step of inputting voice into the computer, and a step of the computer analyzing the voice to obtain the sentence to be evaluated.

5. The method according to claim 2, wherein the evaluation information is information regarding to which category the key sentence belongs.

6. A learning support method by a computer, comprising a step of evaluating the sentence to be evaluated based on the method according to claim 1.

7. A system (1) for evaluating a sentence by a computer, comprising: a sentence input unit (3) for inputting a sentence to be evaluated; a sentence word extraction unit (5) for extracting a sentence word, which is a word included in the sentence to be evaluated; a key sentence extraction unit (7) for extracting a key sentence included in the sentence to be evaluated, where the key sentence includes one or more of the sentence words and one or more auxiliary words. A sentence evaluation unit (9) that reads evaluation information regarding the key sentence from a storage unit based on or in accordance with the key sentence and evaluates the sentence to be evaluated A system including a subsequent sentence creation unit (11) that obtains a subsequent sentence corresponding to the evaluation of the sentence, where the subsequent sentence is a sentence that follows the sentence

8. A program for causing a computer to execute a method for evaluating a sentence, the program wherein the program includes a sentence input step in which a sentence to be evaluated is input to the computer; a sentence word extraction step in which the computer extracts sentence words that are words included in the sentence to be evaluated; a key sentence extraction step in which the computer extracts a key sentence included in the sentence to be evaluated, the key sentence including one or more of the sentence words and one or more auxiliary words; a sentence evaluation step in which the computer evaluates the sentence to be evaluated based on the key sentence; a subsequent sentence creation step in which a subsequent sentence corresponding to the evaluation of the sentence is obtained, the subsequent sentence being a sentence that follows the sentence; and causes the computer to execute the method

9. A non-transitory information recording medium readable by a computer storing the program according to Claim 8

Images