Language Assessment Problem Analysis Method and Language Assessment Problem Analysis Device

Abstract

A method for analyzing language assessment questions and a device for analyzing language assessment questions are disclosed. The present invention is implemented through a process in which an analysis device classifies the question types of language assessment questions, analyzes passages constituting the language assessment questions, and generates correct answer information for the language assessment questions based on the analysis results of the passages. According to the present invention, correct answer information for language assessment questions can be generated with very high accuracy, and in the process, the question types of each language assessment question can be subdivided and classified. Consequently, various language assessment questions of the same question type as the language assessment questions that the learner answered incorrectly can be provided to the learner, thereby significantly increasing the achievement of language learning.

Description

Technology Field

[0001] The present invention relates to a method for analyzing language assessment questions and a device for analyzing language assessment questions, and more specifically, to a method for analyzing language assessment questions and a device for analyzing language assessment questions that can significantly increase the achievement of language learning by enabling the generation of correct answer information in language assessment questions with very high accuracy and, in the process, subdividing and classifying the question types of each language assessment question, thereby providing the learner with various language assessment questions of the same question type as the language assessment questions the learner got wrong. Background Technology

[0002] Conventional language learning and problem-solving methods had limitations in that learning outcomes varied depending on the learner's or teacher's proficiency, as they relied on the learner's or teacher's subjective interpretation or experiential knowledge.

[0003] Furthermore, there is a problem with such conventional language learning methods in that it is difficult for learners to engage in voluntary, repetitive learning of areas where they are weak, such as specific grammatical knowledge like prepositions.

[0004] In this regard, the method of creating and utilizing error notebooks as a learning method to improve test scores for learners preparing for various language exams, such as the Grade 9 Civil Service English exam and TOEIC, has long been widely known.

[0005] Specifically, learners record the questions they got wrong during preliminary assessments, such as TOEIC mock exams, in a separate notebook and study them repeatedly, thereby practicing to avoid making the same or similar mistakes on the actual exam.

[0006] However, such conventional methods of utilizing error notebooks are limited to repeating the study of only the specific problems the learner got wrong, and have the limitation that it is difficult to repeat study problems of the same type as the ones answered incorrectly (for example, another problem evaluating the same grammatical knowledge as the grammar content unknown in the incorrect problem).

[0007] The foregoing description is provided merely as related art to aid in understanding the present invention, and does not imply that the foregoing description can be applied as prior art related to the present invention. Prior art literature

[65535] Published Patent Application No. 10-2022-0131640 (September 29, 2022) The problem to be solved

[0008] Accordingly, the objective of the present invention is to provide a method for analyzing language assessment questions and a device for analyzing language assessment questions that can significantly increase the achievement of language learning by enabling the generation of correct answer information in language assessment questions with very high accuracy, and by subdividing and classifying the question types of each language assessment question in the process, thereby enabling the provision of various language assessment questions of the same question type as the language assessment questions that the learner got wrong.

[0009] The problems to be solved by the present invention are not limited to the problems mentioned, and include other technical problems that can be clearly understood by those skilled in the art to which the present invention belongs from the description below. means of solving the problem

[0010] A method for analyzing language assessment questions according to the present invention for achieving the above objective comprises: (a) a step in which an analysis device analyzes a passage constituting a language assessment question; and (b) a step in which the analysis device generates correct answer information for the language assessment question based on the analysis result of the passage.

[0011] Preferably, prior to step (a), the analysis device further includes a step of classifying the types of questions for the language assessment.

[0012] Meanwhile, the language assessment problem analysis device according to the present invention includes an analysis unit that analyzes a passage constituting a language assessment problem and generates correct answer information for the language assessment problem based on the analysis result of the passage.

[0013] Preferably, the analysis unit is characterized by classifying the types of questions for the language assessment problem. Effects of the invention

[0014] According to the present invention, correct answer information for language assessment questions can be generated with very high accuracy, and in the process, the question types of each language assessment question can be subdivided and classified, thereby enabling the learner to be provided with various language assessment questions of the same question type as the language assessment questions that the learner got wrong, which can significantly increase the achievement of language learning.

[0015] The effects of the present invention are not limited to the effects mentioned, and include other effects that can be clearly understood by those skilled in the art from the description below. Brief explanation of the drawing

[0016] FIG. 1 is a configuration diagram of a learning system in which a method for analyzing language assessment questions according to an embodiment of the present invention is utilized. FIG. 2 is a functional block diagram illustrating the structure of an analysis device for executing a method for analyzing language assessment questions according to an embodiment of the present invention. FIG. 3 is a signal flowchart illustrating the execution process of a learning service utilizing a language assessment problem analysis method according to an embodiment of the present invention, and FIG. 4 is a flowchart illustrating an example of a process for analyzing language assessment questions in a language assessment question analysis device according to the present invention. Specific details for implementing the invention

[0017] The present invention will be described in more detail below with reference to the drawings. It should be noted that identical components in the drawings are denoted by the same reference numerals wherever possible. Furthermore, detailed descriptions of known functions and configurations that could unnecessarily obscure the essence of the invention are omitted.

[0018] FIG. 1 is a configuration diagram of a learning system utilizing a method for analyzing language assessment questions according to an embodiment of the present invention. Referring to FIG. 1, the learning system utilizing a method for analyzing language assessment questions according to an embodiment of the present invention includes a learner terminal (100) and an analysis device (200), and may further include an external server (300) in implementing the present invention.

[0019] The learner terminal (100) is a communication terminal such as a PC, tablet, or smartphone used by a learner who studies various language subjects, such as English, through a learning service utilizing the language evaluation problem analysis method according to the present invention.

[0020] In carrying out the present invention, a predetermined application program for learners necessary to use a learning service utilizing the language evaluation problem analysis method according to the present invention may be installed on the learner terminal (100).

[0021] The analysis device (200) is a device installed and operated by a business operator that provides a learning service utilizing the language evaluation problem analysis method according to the present invention, and in implementing the present invention, the learning device may be implemented in the form of a service provision server.

[0022] Meanwhile, such an analysis device (200) may primarily classify the types of language assessment questions received periodically from an external server (300), analyze the passages constituting each language assessment question, generate correct answer information for the corresponding language assessment question based on the analysis results, and in the process, further subdivide and secondarily classify the types of language assessment questions.

[0023] Various language evaluation questions, such as past exam questions for each round of evaluation tests for various language subjects including English, are stored in an external server (300). In carrying out the present invention, the analysis device (200) may periodically receive language evaluation questions from such an external server (300) and store them in an accumulated manner.

[0024] Meanwhile, the analysis device (200) according to the present invention is configured to communicate with a learner terminal (100) and an external server (300) through a communication network, and the communication network provides a connection path so that the analysis device (200), the learner terminal (100), and the external server (300) can transmit and receive packet data after connecting to each other.

[0025] Communication networks may include wired networks such as LANs (Local Area Networks), WANs (Wide Area Networks), MANs (Metropolitan Area Networks), and ISDNs (Integrated Service Digital Networks), as well as wireless networks such as wireless LANs, CDMA, Bluetooth, and satellite communication, but the scope of the present invention is not limited thereto.

[0026] FIG. 2 is a functional block diagram illustrating the structure of an analysis device (200) for executing a method for analyzing language assessment questions according to an embodiment of the present invention. Referring to FIG. 2, the analysis device (200) for executing a method for analyzing language assessment questions according to an embodiment of the present invention includes a receiving unit (210), a storage unit (230), an analysis unit (250), and a transmitting unit (270).

[0027] The receiving unit (210) of the analysis device (200) receives language evaluation questions from an external server (300) and receives correct answer input information from a learner terminal (100).

[0028] The analysis unit (250) of the analysis device (200) classifies the type of language assessment question, analyzes the passage constituting the language assessment question, and generates correct answer information for the corresponding language assessment question based on the analysis result.

[0029] The transmitting unit (270) of the analysis device (200) transmits a language evaluation question, the analysis process and content of the evaluation question (including the correct answer information of the evaluation question), and another question belonging to the same question type as the question the learner got wrong to the learner terminal (100).

[0030] Meanwhile, various receiving information from the receiving unit (210), various analysis, generation, and judgment information from the analysis unit (250), and various transmission information from the transmission unit (270) may be accumulated and stored in the storage unit (230) of the analysis device (200).

[0031] When the analysis device (200) according to the present invention is implemented as a server, the server may have aspects of being entirely hardware or partially hardware and partially software. For example, the server and each module and unit included therein in this specification may collectively refer to a device for exchanging data of a specific format and content via electronic communication and software related thereto.

[0032] FIG. 3 is a signal flowchart illustrating the execution process of a learning service utilizing a language assessment problem analysis method according to an embodiment of the present invention. Hereinafter, with reference to FIGS. 1 to 3, the execution process of a learning service utilizing a language assessment problem analysis method according to an embodiment of the present invention will be described.

[0033] First, the storage unit (230) of the analysis device (200) has accumulated language evaluation questions, such as past questions from various language tests including the Grade 9 Civil Service English Test and TOEIC.

[0034] Meanwhile, these language evaluation questions may be directly entered into the storage unit (230) by the administrator, but they may also be accumulated and stored in the storage unit (230) by the receiving unit (210) of the analysis device (200) periodically receiving language evaluation questions from an external server (300) (S410).

[0035] The analysis unit (250) of the analysis device (200) may generate classification code information for each type of language evaluation question stored in the storage unit (230) and store the generated type classification code information in association with the corresponding language evaluation question (S420).

[0036] Specifically, the analysis unit (250) of the analysis device (200) may generate code information by classifying the question types of each evaluation question based on the content and format of the answer selection criteria included in the language evaluation question (e.g., Which part among the underlined parts is grammatically incorrect?, Which word is most appropriate to fill in the blank in the following sentence?).

[0037] Meanwhile, the analysis unit (250) of the analysis device (200) may analyze the text constituting the language evaluation question and generate correct answer information for the corresponding language evaluation question based on the analysis result (S430).

[0038] For example, in analyzing an evaluation problem classified as a problem type (e.g., classification code 01) that fills in a blank with a preposition in the aforementioned step S420, the analysis unit (250) of the analysis device (200) may extract a word (keyword) located immediately before the blank in the problem and then search for the extracted word (e.g., responsible) through a predetermined mapping table already stored in the storage unit (230) (S431).

[0039] Meanwhile, for this purpose, the storage unit (230) of the analysis device (200) may store a mapping table in which a predetermined keyword used with a grammatical preposition (e.g., responsible), at least one preposition combined after the keyword (e.g., for, to), and the usage conditions of each preposition (e.g., thing / person) are stored as shown in Table 1 below, with the classification code (e.g., 01) as identification information.

[0040]

[0041] Accordingly, the analysis unit (250) of the analysis device (200) may search a mapping table stored in the storage unit (230) based on the type classification code (e.g., 01) of the evaluation problem, using the type classification code (e.g., 01) as identification information, search for the keyword (e.g., interested) in the searched mapping table, and then generate the preposition (e.g., in) stored in association with the searched keyword as the correct answer information for the problem and store it in association with the problem (S433).

[0042] Meanwhile, when there are multiple prepositions (e.g., in, to) associated with keywords (e.g., reponsible) found in the mapping table, the analysis unit (250) of the analysis device (200) can extract the word located immediately after the blank in the problem and analyze the characteristic information (e.g., object, action, person) of the extracted word through the storage unit (230) (S435).

[0043] To this end, the storage unit (230) of the analysis device (200) may be configured to store characteristic information (e.g., objects, actions, people) associated with each word.

[0044] Next, the analysis unit (250) of the analysis device (200) can generate the correct answer information for the problem by using a preposition (e.g., for) that is stored as an associated condition for use as shown in Table 1, with the characteristic information (e.g., object) of the word analyzed in this way (e.g., object) (S437).

[0045] Meanwhile, in analyzing an evaluation question classified as a question type that selects a grammatically incorrect passage (e.g., classification code 02) or a question type that selects a grammatically incorrect part among the underlined parts (e.g., classification code 03) in the aforementioned S420 step, the analysis unit (250) of the analysis device (200) may extract the subject and verb from the passage such as Table 2 below that constitutes the question, process the words between the extracted subject and verb as noise, and then determine whether the subject and verb agree in number.

[0046]

[0047] For example, the analysis unit (250) of the analysis device (200) may determine that the passage (or underlined part) is grammatically incorrect if the extracted subject is a third-person singular but the present form of the extracted verb does not have an 's' attached to it, and may generate the passage (or underlined part) as correct answer information and store it in association with the corresponding evaluation question.

[0048] Meanwhile, the analysis unit (250) of the analysis device (200) may generate an updated type classification code (02_a) for the evaluation problem and store it in association with the problem by associating a sub-classification code (a) that signifies subject-verb agreement with the type classification code (02) that is initially assigned to the evaluation problem in such cases.

[0049] In addition, when analyzing an evaluation question classified as a question type for selecting a grammatically incorrect passage (e.g., classification code 02) or a question type for selecting a grammatically incorrect part among underlined parts (e.g., classification code 03) in the aforementioned S420 step, the analysis unit (250) of the analysis device (200) may extract and select a word identical to a predetermined coordinating conjunction (and, but, or) stored in the storage unit (230) from the passage constituting the question as shown in Table 3 below, extract words connected by the selected word (coordinating conjunction), determine whether the grammatical forms (part of speech, tense, number, case) of the extracted words are identical, and if a word with an identical grammatical form is included (e.g., two of the three words connected by the coordinating conjunction are nouns and one is an adjective), determine that the word (or the passage containing the word) is grammatically incorrect and generate this as correct answer information.

[0050]

[0051] To this end, the storage unit (230) of the analysis device (200) may be configured to store grammatical form information (part of speech, tense, number, case) for each word in association with the corresponding word.

[0052] Meanwhile, the analysis unit (250) of the analysis device (200) may, in such cases, generate an updated type classification code (02_b) for the evaluation problem and store it in association with the problem by associating a sub-classification code (b) that signifies a grammatical form match in a coordinating conjunction with the type classification code (02) that is initially assigned to the evaluation problem.

[0053] As such, according to the present invention, not only can correct answer information for language assessment questions be generated with very high accuracy, but the types of questions for language assessment can also be further subdivided and classified in the process.

[0054] In addition, when analyzing an evaluation problem classified as a problem type (e.g., classification code 04) that fills in a blank with an appropriate word in the aforementioned S420 step, the analysis unit (250) of the analysis device (200) analyzes whether the sentence following the blank is a clause or a phrase when there are multiple words located after the blank, and if it is determined that the sentence following the blank is a clause because it contains both a subject and a verb as shown in Table 4 below, a word (e.g., since) that matches the predetermined conjunction information (i.e., a conjunction that can be placed before a clause) stored in the storage unit (230) among the words presented in the options may be generated as the correct answer information for the problem.

[0055]

[0056] To this end, the storage unit (230) of the analysis device (200) may be configured to store word information constituting the subject (e.g., you, she, he) and part of speech information of each word (e.g., verb, noun).

[0057] In addition, when analyzing an evaluation problem classified as a problem type (e.g., classification code 04) that fills in a blank with an appropriate word in the aforementioned S420 step, the analysis unit (250) of the analysis device (200) analyzes whether the sentence following the blank is a clause or a phrase when there are multiple words located after the blank, and if it is determined that the sentence following the blank is a phrase because it does not contain both a subject and a verb as shown in Table 5 below, the word that matches the predetermined conjunction information (i.e., a conjunction that can be placed before a clause) stored in the storage unit (230) among the words presented in the options is classified as an incorrect answer, and the word that matches the predetermined preposition information stored in the storage unit (230) (e.g., despite) may be generated as the correct answer information.

[0058]

[0059] In addition, in the aforementioned S420 step, when analyzing an evaluation problem classified as a complex text inference type problem (e.g., classification code 05) as shown in Table 6 below, the analysis unit (250) of the analysis device (200) may extract a word combined with an interrogative word or an interrogative adjective (e.g., which) from the text that asks the learner a question in the problem (e.g., text containing a question mark) as a first keyword (e.g., color), then extract a word that is commonly included in the first text and the second text in the complex text presented in the problem (e.g., SP 722) as a second keyword (i.e., common keyword), and generate a word that has the highest association with the first keyword and the second keyword in the second text (or first text) (e.g., the word located closest to the first keyword and the second keyword: Stormy Blue) as correct answer information.

[0060]

[0061] In addition, in the aforementioned S420 step, when analyzing an evaluation problem classified as a problem type (e.g., classification code 07) that selects the underlined part that is unnatural in context as shown in Table 7 below, the analysis unit (250) of the analysis device (200) may extract all words excluding the underlined word among the words constituting the passage in the problem, and when the extracted words are compared with words of positive meaning and words of negative meaning already stored in the storage unit (230), if there are more words of positive meaning than words of negative meaning, a positive symbol (+) may be generated as symbol information for the passage, and if there are more words of negative meaning than words of positive meaning, a negative symbol (-) may be generated as symbol information for the passage.

[0062]

[0063] In addition, the analysis unit (250) of the analysis device (200) may compare the underlined words in the corresponding text with words of positive meaning and words of negative meaning already stored in the storage unit (230), and if the underlined word matches a word of positive meaning, generate a positive symbol (+) as symbol information for the underlined word, and if the underlined word matches a word of negative meaning, generate a negative symbol (-) as symbol information for the underlined word.

[0064] As a result, the analysis unit (250) of the analysis device (200) may generate a word among the underlined words whose generated symbol information does not match the symbol information generated for the passage of the problem as the correct answer information for the problem.

[0065] To this end, the storage unit (230) of the analysis device (200) may be configured to store classification information associated with each word, classifying it into a positive meaning, a negative meaning, or a neutral meaning.

[0066] In addition, in the aforementioned S420 step, when analyzing an evaluation problem classified as a problem type (e.g., classification code 08) that selects an appropriate text for a situational description of a photo as shown in Table 8 below, the analysis unit (250) of the analysis device (200) may generate an image that visually represents the text for each text based on text information constituting the text (voice script) in the problem, using image generation technology in generative AI, and may generate the text with the highest similarity as the correct answer information by calculating the similarity between the image generated for each text and the photo image included in the problem.

[0067]

[0068] In carrying out the present invention, the analysis device (200) may store the analysis process and content in the process of generating correct answer information for each language evaluation question as described above in association with the corresponding language evaluation question.

[0069] Meanwhile, when the learner subsequently requests the sending of language evaluation questions through the learner terminal (100) (e.g., selecting past questions from the Grade 9 Civil Service English exam, TOEIC exam, etc., and then selecting the question number to start learning), the transmitting unit (270) of the analysis device (200) sends the corresponding language evaluation questions to the learner terminal (100) (S440).

[0070] Meanwhile, in carrying out the present invention, the transmitting unit (270) of the analysis device (200) may also transmit the analysis process and content of the evaluation problem together when transmitting the language evaluation problem to the learner terminal (100) (S450).

[0071] Accordingly, when solving language assessment problems, the learner may refer to the above analysis process and content to select the correct answer information for the problem and then input it into the learner terminal (100).

[0072] Accordingly, the receiving unit (210) of the analysis device (200) receives the learner's correct answer input information from the learner terminal (100) (S460), and accordingly, the analysis unit (250) of the analysis device (200) determines whether the learner's correct answer input information matches the correct answer information (or previously stored correct answer information) generated for the problem in the aforementioned step S430, thereby enabling the evaluation of the learner's solution result for the problem.

[0073] Meanwhile, if it is determined that the learner has entered an incorrect answer because the correct answer information entered by the learner and the correct answer information stored in the storage unit (230) are inconsistent, the analysis unit (250) of the analysis device (200) may store the problem in which the learner entered an incorrect answer in association with the learner's account provided in the storage unit (230) (S470).

[0074] In addition, the analysis unit (250) of the analysis device (200) may search through the storage unit (230) for language assessment questions that have the same type classification code as the type classification code generated as described above for language assessment questions stored in association with the learner's account (i.e., other questions of the same question type), and may construct a customized question pool selected for the learner by additionally storing the searched questions in association with the learner's account (S480).

[0075] In this case, the analysis device (200) may increase the language learning effect by transmitting language assessment questions to the learner terminal (100) of the learner in the future, by transmitting language assessment questions included in the customized question pool, so that the learner can repeatedly solve various questions of the same question type as the questions they got wrong in the past (S490).

[0076] As such, according to the present invention, by providing learners with various questions of the same question type as the language assessment questions they answered incorrectly, the achievement level of language learning can be significantly improved.

[0077] In addition, in carrying out the present invention, a program for executing the method for analyzing language assessment problems according to the present invention may be installed in an analysis device (200) according to the present invention, recorded on various computer-readable recording media, or stored on a server that transmits the program via a network.

[0078] In addition, the program according to the present invention may be distributed across networked computer systems, and computer-readable code may be stored and executed in a distributed manner; furthermore, the functional program, code, and code segments for implementing the present invention will be easily understood by a person skilled in the art to which the present invention pertains.

[0079] Meanwhile, the order of the steps described above (S410–S490, S431–S437) in the present invention is merely an example and is not limited thereto. That is, the order of the steps described above (S410–S490, S431–S437) may vary, and some of these steps may be executed simultaneously or deleted.

[0080] The terms used in this invention are used merely to describe specific embodiments and are not intended to limit the invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "having" are intended to indicate the presence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0081] Although preferred embodiments and applications of the present invention have been illustrated and described above, the present invention is not limited to the specific embodiments and applications described above. Furthermore, various modifications can be made by those skilled in the art without departing from the essence of the invention as claimed in the claims, and such modifications should not be understood individually from the technical spirit or perspective of the present invention. Explanation of the symbols

[0082] 100: Learner terminal, 200: Analysis device, 210: Receiver, 230: Storage, 250: Analysis unit, 270: Transmitter unit, 300: External server.

Claims

Claim 1 (a) a step in which an analysis device classifies the question types of language assessment questions; (b) a step in which the analysis device analyzes a passage constituting the language assessment question to generate correct answer information for the language assessment question; and (c) a step in which the analysis device generates correct answer information for the language assessment question based on the analysis results of the passage, wherein step (b) comprises: (b1) a step in which the analysis device selects a predetermined coordinating conjunction from the passage when the language assessment question is classified in step (a) as a question type for selecting grammatically incorrect words; (b2) a step in which the analysis device extracts words connected by the selected coordinating conjunction; (b3) a step in which the analysis device selects words among the extracted words that do not have the same grammatical form and determines them to be grammatically incorrect; and (b4) a method for analyzing language assessment questions, wherein the analysis device, in step (a), if the language assessment question is classified into a question type that selects an appropriate passage as a situational description of a photograph, generates an image that visually represents the passage based on text information for each passage constituting the language assessment question, and calculates the similarity between the generated image and the image of the photograph for each passage, and (d) further comprises the step of generating an updated type classification code for the language assessment question by associating a sub-classification code meaning a grammatical form match in a coordinating conjunction with the type classification code primarily assigned to the language assessment question in step (a). Claim 2 A method for analyzing language assessment problems according to claim 1, wherein (e) the analysis device further comprises the step of storing the updated type classification code associated with the language assessment problem. Claim 3 A language assessment problem analysis device comprising: a unit that classifies the question types of language assessment problems, analyzes passages constituting the language assessment problems to generate correct answer information for the language assessment problems, and generates correct answer information for the language assessment problems based on the analysis results of the passages; wherein, when the language assessment problems are classified as a question type for selecting grammatically incorrect words, the unit selects a predetermined coordinating conjunction from the passages, extracts words connected by the selected coordinating conjunctions, selects words among the extracted words that do not have the same grammatical form and determines them to be grammatically incorrect; wherein, when the language assessment problems are classified as a question type for selecting a passage appropriate for describing a situation regarding a photograph, the unit generates an image that visually represents the passage for each passage based on text information for each passage constituting the language assessment problems, and calculates the similarity between the generated image and the image of the photograph for each passage; and wherein, the unit generates an updated type classification code for the language assessment problems by associating a sub-classification code signifying the match of grammatical forms in the coordinating conjunctions with a type classification code primarily assigned to the language assessment problems. Claim 4 A language assessment problem analysis device according to paragraph 3, further comprising a storage unit that stores the updated type classification code associated with the language assessment problem.

Images