Method and device for guiding improvement of reading and writing skills

Abstract

The present invention relates to a method and device for guiding improvement of reading comprehension and writing ability, and the method and device thereof include defining basic sentence components, modeling and storing in advance the type of structure for one sentence or the type of connection structure for at least two sentences, and applying each of the modeled types to reading comprehension and writing to learn types of logical development methods including deduction, induction and dialectic, thereby guiding improvement of a user's reading comprehension and writing ability.

Description

[Technical field]

[0001] The present invention relates to a method and an apparatus for guiding improvement of reading comprehension and writing ability, and more particularly to a method and an apparatus for guiding improvement of a user's reading comprehension and writing ability by defining basic sentence components, modeling and storing in advance a type of structure for one sentence or a connection structure for at least two sentences, and applying each of the modeled types to reading comprehension and writing to learn types of logical development methods including deduction, induction and dialectic. [Background technology]

[0002] Humans are so-called social animals, who communicate with each other and form close relationships through conversation and listening to each other throughout the world. In other words, it is no exaggeration to say that how well a person lives depends on how well they communicate with others.

[0003] However, the main means by which people communicate with others are not only "speaking" and "listening," but also "reading" and "writing." The representative activity of "reading" is reading comprehension, and the representative activity of "writing" is composition. Reading allows people to recognize the general trends of society and acquire knowledge, and by composition, they can convey their thoughts to others accurately and effectively.

[0004] However, because there are no firm principles or standards for these important tools of reading and writing, it is extremely difficult for people to clearly express their intentions in writing, and they often get confused or misunderstand when they are unable to accurately decipher what the other person has written.

[0005] In addition, the educational programs for reading and writing that have been developed and used so far are Japanese language education programs for elementary, middle, and high school students, and since they are simply composed of reading and discussion, they do not have a systematic method or system for evaluating the improvement of accurate reading and writing skills.

[0006] In addition, while there are various gifted programs run and structured in a variety of ways for subjects such as science and mathematics, there are currently no programs or evaluation tools for gifted education in the field of Japanese language, which includes reading comprehension and composition.

[0007] Therefore, the present invention aims to provide a method for improving reading comprehension and writing skills by applying a structure for one sentence or a connection structure for at least two or more sentences that has been modeled and stored in advance to reading comprehension and writing, and learning types of logical development methods including deduction, induction, and dialectic, thereby enabling a user to grasp a subject and write logically on his or her own.

[0008] First, a brief description will be given of prior inventions in the technical field of the present invention, and then a description will be given of the technical matters that the present invention aims to achieve in a differentiated manner compared to the prior inventions.

[0009] Korean Patent Registration No. 1058405 (August 23, 2011) is a prior invention relating to a logic learning support system and method using a logic map, which includes a logic map server consisting of an input window display module that outputs a plurality of text input boxes to a user terminal for the step-by-step input of text to realize a logic map having a logical structure, a text input module that receives via the Internet input of text to be entered in the text input boxes according to the logical stages to realize the logic map, and an argument generation module that combines the text in the order of the logical stages to generate an argument.

[0010] In other words, the prior invention describes a system and method that allows a user to input text into a text input box in stages according to a logical structure to practice writing an argument that realizes a logic map, and that allows the text realized by the logic map to be combined and output as a single plain text, which is effective in training the creation of arguments based on logical errors and can improve logical narrative ability and logical expression ability for logical dialogue / discussion.

[0011] However, the present invention defines basic sentence components, models patterns of how sentences are connected based on the defined basic sentence components, constructs a map, and applies the constructed map to reading comprehension and writing, enabling learning of patterns of logical development methods including deduction, induction, and dialectic. Therefore, there is a significant structural difference between the prior invention and the present invention.

[0012] In addition, Korean Patent Publication No. 2010-0111456 (October 15, 2010) is a prior invention relating to a method and system for teaching composition, which includes a step in which the writing instruction system determines a theme of the composition, a step in which the writing instruction system provides the learner with an example composition that corresponds to the theme of the composition and includes one or more word blanks, and a step in which the writing instruction system receives input of words to be filled in the one or more word blanks from the learner.

[0013] In other words, the prior invention describes a method and system that enables learners to more easily learn the overall structure of a sentence and how to develop a logical progression in the process of practicing writing, even if this is their first time encountering writing or they are not accustomed to writing.

[0014] However, the present invention models the types of sentence-to-sentence connection structures through basic sentence components to construct a map, and enables learning of types of logical development methods based on inference, including deduction, induction, and dialectic, using the modeled map. Therefore, there is a clear difference in technical configuration between the prior invention and the present invention. Summary of the Invention [Problem to be solved by the invention]

[0015] The present invention has been created to solve the above-mentioned problems, and has an object to provide a method and an apparatus for guiding a user to improve his / her reading and writing skills.

[0016] Another object of the present invention is to provide a method and apparatus for defining basic components of a sentence, constructing a map by modeling a structure for one sentence or a type of connection structure for at least two or more sentences through the defined basic components of a sentence, and applying the constructed map to reading comprehension and writing, thereby improving reading comprehension and writing abilities.

[0017] Another object of the present invention is to provide a method and apparatus for learning types of logical development methods using inference, including deduction, induction, and dialectics, for explanatory texts that convey information and argumentative texts that persuade through argumentation, using each type that has been modeled and stored in advance.

[0018] Another object of the present invention is to provide a method and apparatus for learning reading and writing through a map that is modeled differently for each level set based on age and ability, thereby enabling each user to improve their ability through evaluation and checking of reading and writing that suits their own level.

[0019] Another object of the present invention is to provide a method and apparatus for objectively measuring the degree of improvement of a user's reading comprehension and writing skills through a verification tool.

[0020] Another object of the present invention is to provide a method and apparatus for improving reading comprehension and writing skills naturally through interactive text checking.

[0021] Another object of the present invention is to provide a method and apparatus for assisting a user in developing insight into a text, enabling the user to grasp a subject and complete a logical composition by himself / herself without the help of others. [Means for solving the problem]

[0022] According to one embodiment of the present invention, there is provided an apparatus for guiding the improvement of reading and writing skills, the apparatus comprising a processor, the processor being configured to analyze a structure for one sentence or a linking structure type for at least two sentences by separating them into a superstructure and a substructure for a given input sentence or a composition created by a user, and to guide the reading or writing.

[0023] The processor is also configured to compare keywords between each sentence of the inputted given sentence or composition created by the user with each of the patterns, sequentially check whether a matching pattern exists up to the last sentence, and if a matching pattern exists, analyze it as the superstructure, and if a matching pattern does not exist, analyze it as an additional model of the superstructure or a substructure, thereby guiding the reading or composition.

[0024] The processor is also characterized by including a pattern modeling unit that models and stores in advance a pattern of a structure for one sentence or a linking structure for at least two or more sentences, and a sentence analysis map construction unit that constructs a map for analyzing the content of the specified sentence required for reading comprehension or a composition created by a user according to each of the modeled patterns.

[0025] The processor further includes a reading processing unit that analyzes the inputted predetermined sentence according to the modeled types and a sentence analysis map and provides the analysis result to a user terminal, and the reading processing unit further includes a sentence analysis unit that analyzes the inputted predetermined sentence according to the modeled types and a sentence analysis map, and a guide information providing unit that provides the analysis result together with the original text of the sentence to the user terminal, thereby enabling an error correction of incorrect reading and accurate reading practice to be performed for the inputted predetermined sentence through type learning of a logical development method including deduction, induction, and dialectic.

[0026] In addition, the sentence analysis unit further includes: reading the inputted predetermined sentence, dividing it into each sentence, dividing each divided sentence into unit components to extract keywords, comparing the extracted keywords between each sentence with each modeled type, sequentially checking whether a matching type exists up to the last sentence, generating an analysis result including a superstructure, a substructure, or a combination thereof including each type of a one-level or additional model, or a combination thereof, if the matching type exists as a result of the checking, generating an analysis result for a superstructure including each type of a two-level or three-level sentence structure model, thereby separating the inputted predetermined sentence into a superstructure and a substructure for interpretation.

[0027] The processor further includes a composition processing unit that analyzes the content of a composition created by a user according to the modeled types and a sentence analysis map, and provides the analysis result to a user terminal, and the composition processing unit further includes a composition analysis unit that analyzes the content of a composition received from the user terminal according to the modeled types and a sentence analysis map, and a guide information providing unit that generates guide information related to the content of the composition based on the analysis result, and provides the generated guide information to the user terminal so that the user can practice writing by correcting and rewriting.

[0028] In addition, the composition analysis unit further includes: reading the content of the received composition, dividing it into sentences, dividing each divided sentence into unit components to extract keywords, comparing the extracted keywords between each sentence with each modeled type, sequentially checking whether a matching type exists up to the last sentence, generating an analysis result including a superstructure, substructure, or a combination thereof including each type of a one-stage or additional model, or a combination thereof, if the matching type exists as a result of the checking, generating an analysis result for a superstructure including each type of a two-stage or three-stage sentence structure model, thereby separating the content of the received composition into a superstructure and a substructure and analyzing it.

[0029] The type modeling unit further includes modeling and storing a type of a structure for one sentence or a connection structure for at least two sentences into a sentence structure model and an additional model, and the sentence structure model is, by referring to a sentence analysis model including a main topic in a one-stage sentence, a development type, an opposition type, and a conclusion type in a two-stage sentence, and a development type, a conclusion type, and an opposition type and a conclusion type in a three-stage sentence, one type of type in a one-stage sentence, four types of types in which a development type conjunction "however" is connected between each sentence in a two-stage sentence, three types of types in which an opposition type conjunction "however" is connected between each sentence in a two-stage sentence, two types of types in which a conclusion type conjunction "therefore" is connected between each sentence in a two-stage sentence, The types of the classes are modeled and stored into a total of 14 types, including two types in which an extension type conjunction "however" is connected between the first sentence and the second sentence in a three-line sentence and a conclusion type conjunction "therefore" is connected between the second sentence and the third sentence, and two types in which an adversative type conjunction "however" is connected between the first sentence and the second sentence and a conclusion type conjunction "therefore" is connected between the second sentence and the third sentence in a three-line sentence, and the additional models are modeled and stored into two types, including a type for conjunctions not mentioned in the sentence structure model and a type for contents unrelated to the sentence structure model.

[0030] In addition, the method for guiding the improvement of reading comprehension and writing ability according to one embodiment of the present invention is characterized in that, in the method for guiding reading comprehension or writing, for a given input sentence or a composition created by a user, a type of structure for one sentence or a linking structure for at least two sentences is separated into a superstructure and a substructure, and analyzed, and the reading comprehension or writing is guided.

[0031] In addition, the method is characterized in that a processor is configured to compare keywords between each sentence of the inputted given sentence or composition created by the user with each of the patterns, sequentially check whether a matching pattern exists up to the last sentence, and if the matching pattern exists, analyze it as the superstructure, and if the matching pattern does not exist, analyze it as an additional model of the superstructure or a substructure, thereby guiding the reading or composition.

[0032] The method further includes a pattern modeling step of modeling and storing a pattern of a structure for one sentence or a linking structure for at least two sentences in advance by a processor, and a sentence analysis map construction step of constructing a map for analyzing the content of the predetermined sentence or a composition created by a user, which is necessary for reading comprehension, according to each of the modeled patterns.

[0033] Also, the method further includes a reading processing step of analyzing the inputted given sentence according to the modeled types and a sentence analysis map by the processor, and providing the analysis result to a user terminal, the reading processing step further including a sentence analysis step of analyzing the inputted given sentence according to the modeled types and a sentence analysis map, and a guide information providing step of providing the analysis result together with the original sentence of the sentence to the user terminal, thereby enabling the user to correct reading errors and practice reading correctly by learning types of logical development methods including deduction, induction, and dialectic for the inputted given sentence.

[0034] In addition, the sentence analysis step further includes reading the inputted predetermined sentence and dividing it into each sentence, dividing each divided sentence into unit components to extract keywords, comparing the extracted keywords between each sentence with each modeled type to check whether a matching type exists up to the last sentence, generating an analysis result including a superstructure, substructure, or a combination thereof including each type of a one-level or additional model, or a combination thereof, if the matching type exists as a result of the check, generating an analysis result for a superstructure including each type of a two-level or three-level sentence structure model, thereby separating the inputted predetermined sentence into a superstructure and a substructure for interpretation.

[0035] In addition, the method further includes a composition processing step of analyzing the content of a composition created by a user according to the modeled types and a sentence analysis map, and providing the analysis result to the user terminal, in which the composition processing step further includes a composition analysis step of analyzing the content of the composition received from the user terminal according to the modeled types and a sentence analysis map, and a guide information providing step of generating guide information related to the content of the composition based on the analysis result, and providing the generated guide information to the user terminal so that the user can practice writing by correcting and rewriting the composition.

[0036] In addition, the composition analysis step further includes reading the content of the received composition, dividing it into each sentence, dividing each divided sentence into unit components to extract keywords, comparing the extracted keywords between each sentence with each modeled type to check whether a matching type exists up to the last sentence, generating an analysis result including a superstructure, substructure, or a combination thereof including each type of a one-stage or additional model, or a combination thereof, if the matching type exists as a result of the check, generating an analysis result for a superstructure including each type of a two-stage or three-stage sentence structure model, thereby separating the content of the received composition into a superstructure and a substructure and analyzing them. Effect of the Invention

[0037] As described above, according to the method and device for guiding improvement of reading comprehension and writing ability of the present invention, a map is constructed by modeling the structure of one sentence or the type of connection structure of at least two or more sentences, and the constructed map is applied to reading comprehension and writing to naturally learn types of logical development methods including deduction, induction, and dialectic. This allows a user to grasp the subject and complete a logical composition by himself / herself without the help of others, thereby significantly improving reading comprehension and writing ability.

[0038] In addition, the present invention supports learning of reading and writing through a map that is modeled differently for each level set based on age and ability, so that each user can improve their ability through evaluation and checking of reading and writing that suits their own level, and can naturally achieve their learning goals through interactive writing checks. [Brief description of the drawings]

[0039] [Figure 1] 1 is a conceptual diagram illustrating the configuration of an entire system including a device for guiding improvement of reading and writing skills according to an embodiment of the present invention.

[0040] [Diagram 2] 1 is a diagram for explaining a program structure map used in a device for guiding improvement of reading and writing skills according to an embodiment of the present invention.

[0041] [Diagram 3] FIG. 11 is a diagram showing an example in which each type of sentence structure model of the program structure map applied to the present invention is classified by a sentence analysis model.

[0042] [Figure 4] FIG. 2 is a diagram showing an example of a text analysis model used when constructing a text structure model of the present invention.

[0043] [Diagram 5] FIG. 13 is a diagram showing an example of an additive model of a program structure map applied to the present invention.

[0044] [Figure 6] 1 is a block diagram showing in detail the configuration of a device for guiding improvement of reading and writing skills according to an embodiment of the present invention;

[0045] [Figure 7] FIG. 4 is a block diagram showing in detail the configuration of a type modeling unit according to an embodiment of the present invention.

[0046] [Figure 8] 2 is a block diagram showing in detail the configuration of a reading comprehension processing unit according to an embodiment of the present invention; FIG.

[0047] [Figure 9] 1A and 1B are diagrams showing examples of original sentences and sentence analysis results provided to a user terminal performing reading comprehension in accordance with the present invention;

[0048] [Figure 10] 2 is a block diagram showing in detail the configuration of an essay processing unit according to an embodiment of the present invention; FIG.

[0049] [Figure 11] FIG. 1 is a diagram showing an example of a topic provided to a user terminal for writing an essay in accordance with the present invention, and an example of a content analysis result of an essay created by a user in response to the topic.

[0050] [Figure 12] FIG. 13 is a diagram showing an example of estimating each type of sentence structure model using artificial intelligence when guiding the improvement of reading comprehension and writing ability applied to the present invention.

[0051] [Figure 13] 4 is a flowchart showing a process of constructing a program structure map used in a device for guiding improvement of reading and writing skills according to an embodiment of the present invention.

[0052] [Figure 14] 2 is a flowchart illustrating in detail the operation process of a method for guiding improvement of reading and writing skills according to an embodiment of the present invention.

[0053] [Figure 15a] 4 is a detailed flowchart illustrating an analysis process for reading comprehension or writing according to an embodiment of the present invention. [Figure 15b] 4 is a detailed flowchart illustrating an analysis process for reading comprehension or writing according to an embodiment of the present invention.

[0054] [Figure 16] FIG. 2 is a diagram showing a hardware structure of a device for guiding the improvement of reading and writing skills according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0055] Hereinafter, with reference to the accompanying drawings, a preferred embodiment of the present invention for a method and an apparatus for guiding the improvement of reading comprehension and writing ability will be described in detail. The same reference numerals in each drawing indicate the same members. In addition, specific structural or functional descriptions for the embodiments of the present invention are merely illustrative for the purpose of describing the embodiments according to the present invention, and unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person having ordinary skill in the art to which the present invention belongs. Terms commonly used and defined in a dictionary should be interpreted to have a meaning consistent with the meaning in the context of the related art, and it is preferable that they are not interpreted in an ideal or overly formal sense unless clearly defined in this specification.

[0056] FIG. 1 is a conceptual diagram showing the overall configuration of a system including a device for guiding improvement of reading comprehension and writing ability according to an embodiment of the present invention.

[0057] As shown in FIG. 1, the present invention includes an apparatus 100 for guiding improvement of reading and writing skills (hereinafter referred to as a reading and writing improvement guide apparatus), a plurality of user terminals 200, a database 300, and the like.

[0058] The reading and writing improvement guide device 100 is a server computer operated by a business that provides language-related reading and writing learning programs for everyone, including children, students, and adults.

[0059] Here, the language used in the reading and composition learning program provided by the present invention will be explained as a Japanese language as an example, but it should be made clear that the present invention is not limited to this and can be expanded and applied to various languages ​​such as English, Japanese, and Chinese.

[0060] The reading and writing improvement guide device 100 defines basic components of a sentence so as to improve the reading and writing ability of each user, models a structure for one sentence or a type of connection structure for at least two sentences based on the defined basic components of a sentence, and constructs a map, and stores the constructed map in the database 300. At this time, the program structure map applied to the present invention will be described in more detail with reference to Figs. 2 to 5.

[0061] In addition, the reading and writing improvement guide device 100 stores a map constructed by modeling the structure of one sentence or the type of connection structure of at least two or more sentences in the database 300, and then applies the constructed map to reading and writing, thereby enabling each user to learn types of logical development methods based on inference, including deduction, induction, and dialectic, thereby improving reading and writing abilities.

[0062] That is, the reading and writing improvement guide device 100 analyzes the specific sentences required for reading and the contents of the writing created by the user by separating them into superstructure and substructure according to each modeled type of the constructed map, and provides the analyzed information to the user terminal 200, thereby playing the role of a guide that allows each user who reads or writes to read or write while recognizing the logical structure of sentence A.

[0063] As a result, each user can naturally improve their reading comprehension and writing skills through the interactive guide of the reading comprehension and writing improvement guide device 100. In addition, it is possible to develop insight into the text, and to grasp the subject matter by themselves and complete logical writing without the help of others.

[0064] Meanwhile, the reading and writing improvement guide device 100 may classify each user who reads or writes into a plurality of levels based on age, ability, etc., and form groups, and model and store a different type of sentence and sentence link structure for each of the formed groups. That is, by modeling a different type of link structure between sentences for each level and storing it as a map, each user can improve their ability through reading and writing study that suits their level.

[0065] In addition, the reading comprehension and writing improvement guide device 100 can objectively measure the improvement level of each user's reading comprehension and writing ability by evaluating the improvement level of the reading comprehension and writing ability according to a predetermined period.

[0066] In this case, the evaluation can be performed by presenting questions related to reading comprehension or composition to the user terminal 200 connected via an application program installed on the user terminal 200 or via the Web, and having each user review the questions and provide an answer sheet, which is then given to an expert for grading.

[0067] The user terminal 200 is a communication device that can use the wired or wireless Internet, such as a smartphone or tablet owned by a user, and establishes a communication connection to the reading and writing improvement guide device 100 via a network according to a user's operation to log in.

[0068] In addition, the user terminal 200 requests the reading and writing practice from the reading and writing improvement guide device 100 in response to a user's operation, and in response to the request, the user can practice reading or writing by receiving data for reading and writing improvement provided by the reading and writing improvement guide device 100. At this time, the reading or writing practice can be performed by an application program installed in the user terminal 200, or by directly connecting to the reading and writing improvement guide device 100 based on the Web.

[0069] For example, when a user requests reading comprehension, the user terminal 200 can receive sentences including explanatory sentences that convey information and argumentative sentences that persuade through arguments from the reading comprehension and writing improvement guide device 100, and information that analyzes each sentence according to each type that has been modeled in advance, and perform reading comprehension practice.

[0070] In addition, when a user requests to write an essay, the user terminal 200 receives a specific topic for writing from the reading comprehension and writing improvement guide device 100, and transmits the content of the essay created by the user according to the specific topic provided to the reading comprehension and writing improvement guide device 100. Next, the user terminal 200 receives information obtained by analyzing the content of the essay created by the user according to each type modeled in advance (for example, data corrected to see whether the content of the essay created by the user conforms to the rules according to the modeled type) from the reading comprehension and writing improvement guide device 100, and if the essay created by the user is not suitable, the user can check which part is lacking and practice writing.

[0071] The database 300 stores and manages information on a map that is constructed by modeling the structure of one sentence or the type of connection structure of at least two or more sentences based on the basic components of the sentence in the reading and writing improvement guide device 100.

[0072] In addition, the database 300 stores and manages information on sentences, including various explanatory and expository sentences, used in reading comprehension learning, and various topics used in composition learning. At this time, the sentences and topics can be provided by education-related research institutes, educational institutions, question banks, etc., and can be classified and stored according to each level.

[0073] In addition, the database 300 stores and manages membership information for each user, reading comprehension and composition learning information for each user, and the like.

[0074] Next, a map constructed by modeling a plurality of types of sentences and sentence link structures that are constructed in advance for reading comprehension and writing provided by the present invention will be described in detail with reference to FIG. 2 to FIG. 5.

[0075] FIG. 2 is a diagram for explaining a program structure map used in the device for guiding the improvement of reading comprehension and writing ability according to an embodiment of the present invention.

[0076] As shown in Figure 2, all texts are made up of larger sentences which then become paragraphs, and paragraphs which then become sentences. However, sentences, paragraphs, and sentences all have a logical structure. In addition, non-literary texts generally consist of two types of texts: explanatory texts for conveying information and argumentative texts for persuasive argumentation, and the methods of logical development are divided into three types: deductive reasoning, inductive reasoning, and dialectical reasoning.

[0077] In the present invention, all such sentences are defined by three elements: a superstructure consisting of a premise or a topic, and a substructure consisting of a support. Based on the above-defined sentence elements, a map can be constructed by modeling the sentence into 14 types of sentence structure models and 2 types of additional models.

[0078] In this case, the substructure can have 14 types of sentence structure models and 2 types of additional models, just like the superstructure. In other words, just like the superstructure connected by premise or topic, the substructure consisting of the premise or topic support of the superstructure can also be connected to each other and have each type of sentence structure model.

[0079] FIG. 3 is a diagram showing an example in which each type of text structure model of a program structure map applied to the present invention is classified by a text analysis model.

[0080] As shown in FIG. 3, in the present invention, the sentence structure model is classified into 14 types of sentences and sentence linkage structures, including one type of one-stage type, nine types of two-stage types, and four types of three-stage types, by referring to a sentence analysis model, and a map can be constructed and stored in the database 300. Here, modeling is to generate a specific model defined in the present invention, and the specific model can be defined and identified as a data structure or a template.

[0081] In this case, the sentence analysis model may include a topic in a one-line sentence, an extension-type conjunction, an adversative conjunction and a conclusion-type conjunction in a two-line sentence, an extension-type and conclusion-type conjunction, and an adversative and conclusion-type conjunction in a three-line sentence, or a combination thereof.

[0082] For example, in the case of a one-paragraph text, the text structure model may be modeled and stored as one type of topic type.

[0083] In addition, in the case of a two-line sentence, the sentence structure model can be stored in a total of nine types, such as four types for the conjunction linkage of "however", three types for the conjunction linkage of "but", and two types for the conjunction linkage of "therefore". In this case, A, B, C, etc. on the drawing indicate similar keywords and contexts extracted from each line, "-" indicates a sentence with an opposite concept to the keyword or context of the previous sentence, and "!" indicates that if the previous sentence is a sentence with a positive meaning, a sentence containing a keyword or context that maintains the meaning of the sentence is connected, and if the previous sentence is a sentence with an incorrect meaning, a sentence containing a keyword or context that improves the negative meaning of the sentence is connected.

[0084] For example, if keywords A and B are extracted from each paragraph of the first sentence, and keywords B and C are extracted from each paragraph of the second sentence, they are modeled and stored as a type linked by the development conjunction "however." Also, if keywords A and B are extracted from each paragraph of the first sentence, and keywords A and -B are extracted from each paragraph of the second sentence, they are modeled and stored as a type linked by the adversative conjunction "however." Also, if keywords A and B (affirmative) are extracted from each paragraph of the first sentence, and keywords A and B! (maintained) are extracted from each paragraph of the second sentence, they are modeled and stored as a type linked by the conclusion conjunction "therefore."

[0085] In addition, in the case of a three-line sentence, the sentence structure model can be stored in a total of four types: two types in which the development-type and conclusion-type conjunctions "however" and "therefore" are sequentially connected between the first sentence and the second sentence, and between the second sentence and the third sentence; and two types in which the opposition-type and conclusion-type conjunctions "however" and "therefore" are sequentially connected between the first sentence and the second sentence, and between the second sentence and the third sentence.

[0086] For example, if keywords A and B are extracted from each paragraph of the first sentence, keywords B and C (affirmative) are extracted from each paragraph of the second sentence, and keywords (A) and (C!) (maintenance) are extracted from each paragraph of the third sentence, they are modeled and stored as a type in which the development conjunction "however" and the conclusion conjunction "therefore" are linked in sequence. Also, if keywords A and B are extracted from each paragraph of the first sentence, keywords A and -B (negative) are extracted from each paragraph of the second sentence, and keywords (A) and (-B!) (improvement) are extracted from each paragraph of the third sentence, they are modeled and stored as a type in which the opposition conjunction "however" and the conclusion conjunction "therefore" are linked in sequence.

[0087] FIG. 4 is a diagram showing an example of a text analysis model used when constructing a text structure model of the present invention.

[0088] As shown in FIG. 4, the sentence structure model is modeled and stored into 14 types according to one-, two-, and three-line sentences. In the case of a two- or three-line sentence, a sentence analysis model including a developmental conjunction "however," an adversative conjunction "however," a conclusion-type conjunction "therefore," or a combination thereof can be used.

[0089] For example, in the case of an example sentence consisting of "Mankind desires peace. However, in order to realize peace, the following efforts must be made," in the present invention, the first sentence is divided into two components, "Mankind desires peace," and "Peace," and the second sentence is divided into two components, "In order to realize peace," and "The following efforts must be made," and keywords of each component are extracted. The first sentence is a sentence of the A, B type, and the second sentence is a sentence of the B, C type. As a result, the two sentences are modeled and stored as a type in which the developed conjunction "however" is connected between them.

[0090] FIG. 5 is a diagram showing an example of an additive model of a program structure map applied to the present invention.

[0091] As shown in FIG. 5, the additional model has two types: a conjunction link not mentioned in the sentence structure model, and a content unrelated to the sentence structure model.

[0092] For example, there are two types of sentences: one in which the first sentence is developed in the form of A1 and B1, and then the second sentence is developed in the form of A2 and B2 following the conjunction ``and'' (in this case, A1 and A2, and B1 and B2 are mutually related sentences), and one in which the sentence is developed with ``addition'' and content unrelated to the above sentence structure model.

[0093] FIG. 6 is a block diagram showing in detail the configuration of a device for guiding improvement of reading and writing skills according to an embodiment of the present invention.

[0094] As shown in FIG. 6, the reading and writing improvement guide device 100 includes a text element identification unit 110, a type modeling unit 120, a text analysis map construction unit 130, a reading processing unit 140, a writing processing unit 150, an evaluation unit 160, a control unit 170, etc.

[0095] The text element identification unit 110 defines the elements of each text, which may be an explanatory text for conveying information or an argumentative text for persuasion. For example, the present invention defines the text as three types of elements, which are a premise, a topic superstructure, and a supporting substructure.

[0096] The type modeling unit 120 pre-models the structure of one sentence or the type of linking structure of at least two sentences into 14 types of sentence structure models based on the basic components of a sentence defined by the sentence element identification unit 110, and stores the models in the database 300.

[0097] At this time, the type modeling unit 120 may classify users who read or write into levels according to age or ability, and compress, model, and store the types for each classified level. For example, the type modeling unit 120 may compress, model, and store in the database 300 the sentence structure model, which is modeled into 14 types in the lowest level, into a total of three types, such as one type for the development type conjunction linkage of "however," one type for the opposition type conjunction linkage of "however," and one type for the conclusion type conjunction linkage of "therefore," as shown in FIG. 4.

[0098] The text analysis map constructing unit 130 classifies each type modeled by the type modeling unit 120 by referring to a text analysis model, constructs a map as shown in FIG. 3, and stores and manages the constructed map in the database 300.

[0099] The reading comprehension processing unit 140 reads a predetermined sentence in response to the reading selection of the user terminal 200, analyzes the sentence to be provided to the user terminal 200 according to each modeled type of the map constructed by the sentence analysis map construction unit 130, and provides the analyzed result together with the original sentence to the user terminal 200, so that the user can correct errors in reading and practice reading correctly.

[0100] In addition, the reading comprehension processing unit 140 stores sentences necessary for the reading comprehension as big data, and when a user who has selected the reading comprehension study inputs a specific topic, it derives at least one sentence related to the topic and provides it to the user. When the user selects one of the at least one sentence derived according to the topic he or she inputted, it analyzes the sentence selected by the user and provides the analysis result together with the original sentence to the user terminal 200 so that the user can practice reading comprehension.

[0101] The composition processing unit 150 transmits a predetermined topic to the user terminal 200 in response to a composition selection made by the user terminal 200, and when the content of a composition created in response to the predetermined topic is received from the user terminal 200, the composition processing unit 150 analyzes the content of the composition according to each modeled type of the map created by the writing analysis map creating unit 130, and provides the analysis results to the user terminal 200 so that the user can practice writing.

[0102] In addition, the composition processing unit 150 stores the contents of the compositions created by each user in response to a specific topic and the analysis results thereof in the database 300 for each user, thereby enabling each user who writes the composition to easily understand the degree of improvement of his / her composition.

[0103] In other words, the reading comprehension processing unit 140 and the composition processing unit 150 act as a guide to help learn the types of logical development methods, such as deduction, induction, and dialectic, by analyzing all types of text related to explanatory and expository texts required for reading comprehension and the contents of compositions created by users according to specific topics, according to each type modeled in the map constructed by the text analysis map construction unit 130.

[0104] The evaluation unit 160 generates evaluation questions for verifying reading comprehension and writing ability according to a predetermined period, provides the generated evaluation questions to the user terminal 200, and evaluates the user's reading comprehension and writing ability by scoring based on the answers input from the user terminal 200. The evaluation results are stored in the database 300 for management.

[0105] At this time, answers submitted by each user in response to an evaluation carried out at a predetermined interval can be provided to and graded by experts who are contracted with the reading and writing improvement guide device 100.

[0106] In addition, the evaluation unit 160 can adjust the level of reading comprehension or writing for each user based on the evaluation result.

[0107] The control unit 170 is a processor that controls the overall operation of each component of the reading and writing improvement guide device 100 .

[0108] FIG. 7 is a block diagram showing in detail the configuration of a type modeling unit according to an embodiment of the present invention.

[0109] As shown in FIG. 7, the type modeling unit 120 includes a sentence structure modeling unit 121 and an additional model modeling unit 122 .

[0110] As described in FIG. 3, the sentence structure modeling unit 121 models and stores 14 types of structures for one sentence or linking structures for at least two sentences, thereby supporting reading comprehension or writing practice for each user.

[0111] In this case, the sentence structure modeling unit 121 can compress the 14 types of one-, two-, and three-line sentences into three types of two-line sentences (i.e., one type connected with the development conjunction “however”, one type connected with the adversative conjunction “however”, and one type connected with the conclusion conjunction “therefore”) as shown in FIG. 4, model and store them.

[0112] As described in FIG. 5, the additional model modeling unit 122 models and stores two types of types for conjunction connections not included in each type of the sentence structure model and contents unrelated to the sentence structure model.

[0113] At this time, the additional model modeled through the additional model modeling unit 122 is used for both reading comprehension and writing.

[0114] FIG. 8 is a detailed block diagram showing the configuration of a reading comprehension processing unit according to one embodiment of the present invention, and FIG. 9 is a diagram showing an example of an original sentence and a sentence analysis result provided to a user terminal performing reading comprehension applied to the present invention.

[0115] As shown in FIG. 8, the reading comprehension processing unit 140 includes a sentence loading unit 141, a sentence analysis unit 142, and a guide information providing unit 143.

[0116] When reading comprehension is selected from the user terminal 200 , the sentence loading unit 141 loads explanatory and editorial sentences to be provided to the user terminal 200 from the database 300 , and outputs the loaded sentences to the sentence analysis unit 142 .

[0117] The sentence analysis unit 142 analyzes the sentences loaded by the sentence loading unit 141 according to the modeled types and a sentence analysis map, and outputs the analysis result to the guide information providing unit 143 .

[0118] More specifically, when a reading comprehension is selected from the logged-in user terminal 200, the sentence analysis unit 142 reads a sentence to be used for the reading comprehension from the database 300. At this time, the sentence to be read is a sentence that has not been provided to the user before.

[0119] In addition, the sentence analysis unit 142 divides the read characters into sentences, divides each divided sentence into unit components, extracts keywords, and stores them. At this time, if the divided sentence is in a compound sentence form, even if it is a two-line sentence, if it is in the form of a one-line sentence, it is determined to be a one-line sentence (all compound sentences are in one line). In deriving the keywords, in the present invention, various methods can be applied, such as detection by natural language analysis or detection by excluding particles from each syllable.

[0120] In addition, the sentence analysis unit 142 reads the extracted and stored keywords for the entire sentence, and then compares the keywords between each of the read sentences with each type of the sentence structure model of the map constructed by the sentence analysis map construction unit 130 to check whether a matching type exists up to the last sentence. If the matching type exists as a result of the confirmation, it analyzes the superstructure, and if the matching type does not exist, it analyzes the additional model or substructure of the superstructure, so that the reading comprehension can be guided. More specifically, if the matching type does not exist, it generates an analysis result including a superstructure, a substructure, or a combination thereof, including each type of one-stage and additional model, and if the matching type exists as a result of the confirmation, it generates an analysis result for a superstructure including each type of two-stage or three-stage sentence structure model, thereby separating the inputted given sentence into a superstructure and a substructure and performing a reading comprehension analysis (see FIG. 15). a,15b reference).

[0121] The guide information providing unit 143 generates guide information based on the analysis results of the sentence analysis unit 142, and provides the generated guide information together with the original text of the sentence to the user terminal 200, so that the user can learn the types of logical development methods of deduction, induction, and dialectic.

[0122] That is, the guide information providing unit 143 provides the original sentence and the sentence analysis result (i.e., guide information) to the user terminal 200 as shown in FIG. 9, so that the user can correct errors in reading and practice reading correctly.

[0123] At this time, the sentence analysis result provided to the user terminal 200 together with the original sentence may be provided in a form in which each sentence of the sentence and a type of link structure of the sentence are overlaid and displayed.

[0124] In addition, the original sentence and the sentence analysis result provided to the user terminal 200 can be displayed simultaneously on the screen of the user terminal 200. In consideration of the efficiency of reading practice, the original sentence is provided preferentially so that the user can read while grasping the sentence type. After a predetermined time has elapsed, the sentence analysis result is provided and a comparison is made as to whether the type grasped by the user from the original sentence matches the sentence analysis result, thereby enabling error correction of incorrect reading and accurate reading practice.

[0125] FIG. 10 is a block diagram showing in detail the configuration of an essay processing unit according to one embodiment of the present invention, and FIG. 11 is a diagram showing an example of a topic provided to a user terminal that writes essays applied to the present invention, and an example of the results of content analysis of an essay created by a user in response to the topic.

[0126] As shown in FIG. 10, the composition processing unit 150 includes a topic providing unit 151, an composition receiving unit 152, an composition analyzing unit 153, and a guide information providing unit 154.

[0127] When an essay is selected from the user terminal 200 , the topic providing unit 151 loads a specific topic to be used in the essay from the database 300 , and transmits the loaded specific topic to the user terminal 200 .

[0128] The essay receiving unit 152 receives the content of an essay created by a user in stages in accordance with a specific topic from the user terminal 200 that has been provided with a specific topic from the topic providing unit 151, and outputs the content of the received essay to the essay analysis unit 153.

[0129] The composition analysis unit 153 analyzes the content of the composition received by the composition receiving unit 152 according to the modeled types and a sentence analysis map, and outputs the analysis result to the guide information providing unit 154.

[0130] More specifically, when an essay is selected from the logged-in user terminal 200, the essay analysis unit 153 loads a specific topic from the database 300 and transmits it to the user terminal 200. At this time, the specific topic to be loaded must not have been provided to the user before.

[0131] In addition, when the composition analysis unit 153 receives the content of a composition created by a user in stages according to the specific topic from the user terminal 200, it divides the content of the received composition into individual sentences, divides each of the divided sentences into unit components, extracts keywords, and stores them.

[0132] Next, the composition analysis unit 153 reads the keywords for the entire sentence extracted and stored, and then compares the keywords between each of the read sentences with each type of the sentence structure model of the map constructed by the sentence analysis map construction unit 130 to check whether a matching type exists up to the last sentence. If the matching type exists as a result of the check, it analyzes the superstructure, and if the matching type does not exist, it analyzes the additional model or substructure of the superstructure, thereby guiding the composition. More specifically, if the matching type does not exist, it generates an analysis result including a superstructure, substructure, or a combination thereof including each type of one-stage and additional model, and if the matching type exists as a result of the check, it generates an analysis result for a superstructure including each type of two-stage or three-stage sentence structure model, thereby separating the inputted given sentence into a superstructure and a substructure to perform composition analysis (FIG. 15). a,15b reference).

[0133] Based on the analysis result by the composition analysis unit 153, the guide information providing unit 154 generates an analysis result (i.e., guide information) for the composition that the user directly created in response to the specific topic, as shown in Fig. 11. That is, the guide information generating unit 154 generates guide information indicating whether the content of the composition is suitable for each type of the sentence structure model modeled in advance, or whether the content of the composition is incorrectly created and does not correspond to each type modeled in advance.

[0134] In addition, the guide information providing unit 154 provides the generated guide information to the user terminal 200, so that the user can practice writing by correcting or rewriting the content of the composition that the user created.

[0135] At this time, the guide information for the composition provided from the guide information providing unit 154 to the user terminal 200 may be provided in a form in which each sentence of the composition and a type of link structure of the sentence are overlaid and displayed.

[0136] In addition, the guide information providing unit 154 can generate and provide information on which part has an error, information on corrections to errors in the content of the composition, etc., in addition to analysis information for each type of the composition (e.g., types A, B, C, -C, and C! on a drawing).

[0137] Meanwhile, the present invention can utilize artificial intelligence when analyzing the sentences used for reading or the contents of the compositions created directly by the user in the reading and writing improvement guide device 100 according to the types of sentence structure models and sentence analysis maps constructed in advance.

[0138] FIG. 12 is a diagram showing an example of estimating each type of sentence structure model using artificial intelligence when guiding the improvement of reading comprehension and writing ability applied to the present invention.

[0139] First, as shown in (a) of FIG. 12, the reading and writing improvement guide device 100 learns patterns for each type of sentence structure model for two-paragraph and three-paragraph sentences, generates an artificial intelligence learning model for each type of the sentence structure model shown in FIG. 3, and stores the generated artificial intelligence learning model in the database 300.

[0140] In this case, the learning data used for the learning is obtained by dividing each two or three paragraphs of text into two unit components, and visualizing the keywords extracted from each unit component using pixel values ​​according to whether they are the same, similar, or opposite. For example, the same pixel value can be assigned to the same keywords for each unit component, similar pixel values ​​within a predetermined range can be assigned to similar keywords, and white and black pixel values ​​can be assigned to opposite keywords.

[0141] In addition, when the reading and writing improvement guide device 100 receives a sentence to be analyzed, which is a sentence used for reading or a composition created directly by a user, as shown in FIG. 12(b), it performs preprocessing to divide a two- or three-line sentence into two unit components, and generates a pattern for the sentence to be analyzed according to whether the keywords extracted from each unit component are the same, similar, or opposite.

[0142] Next, the reading and composition improvement guide device 100 inputs the generated pattern into an AI learning model based on each type of sentence structure model previously constructed, estimates the most probable type among the results output from each AI learning model as the type for the analysis target sentence, analyzes the analysis target sentence according to the estimated type, and generates guide information for sentences used in reading or for the contents of a composition created directly by a user based on the analysis result, and provides it to the user terminal 200.

[0143] Next, an embodiment of the method for guiding the improvement of reading comprehension and writing ability according to the present invention will be described with reference to FIGS. 13 to 15. a,15b Here, the order of each step of the method of the present invention may be changed depending on the environment of use or the person skilled in the art.

[0144] FIG. 13 is a flow chart showing a process of constructing a program structure map used in the device for guiding the improvement of reading and writing skills according to an embodiment of the present invention.

[0145] As shown in FIG. 13, the reading and writing improvement guide device 100 defines basic sentence components of all sentences related to explanatory and expository texts (S100), and performs a type modeling step of pre-modeling and storing a type of structure for one sentence or a type of connection structure for at least two sentences based on the defined basic sentence components (S200).

[0146] At this time, as described above, the type modeling step can classify the types of sentence and sentence link structures into whole types, additional types, basic types, and additional types depending on the age and ability of the user who reads and writes, and model each type differently.

[0147] At this time, the type modeling step can classify users who read or write into levels based on age and ability, and compress and model each type of sentence and sentence connection structure for each classified level, as described above in FIG. 6.

[0148] In addition, the reading and writing improvement guide device 100 performs a sentence analysis map construction step of constructing a map for analyzing the predetermined sentence required for reading or the content of the composition created by the user according to each type modeled through the step S200, and stores the constructed map in the database 300 (S300).

[0149] As described above, after constructing and storing a map based on each type of structure for one sentence or each type of link structure for at least two sentences in steps S100 to S300, the reading and writing improvement guide device 100 applies the constructed map to reading and writing to practice reading and writing for each user. That is, the device analyzes a given sentence required for reading and the contents of a composition created by a user based on each type of the constructed map and a sentence analysis map, and guides reading and writing, thereby enabling each user to improve their reading and writing abilities.

[0150] Next, this is shown in Figs. 14 and 15. a,15b This will be described in detail with reference to.

[0151] FIG. 14 is a flowchart showing in detail the operation process of a method for guiding improvement of reading and writing skills according to an embodiment of the present invention, and FIG. a,15b 4 is a detailed flowchart illustrating an analysis process for reading comprehension or writing according to an embodiment of the present invention.

[0152] As shown in FIG. 14, the reading and writing improvement guide device 100 checks the content of a sentence used for reading comprehension or a composition created directly by the user according to the user's reading or composition selection (S400), and determines whether the content is a sentence used for reading comprehension or a composition created by the user as a result of the checking (S500).

[0153] If it is determined in step S500 that the sentence is used for reading comprehension, the reading and writing improvement guide device 100 analyzes the sentence loaded from the database 300 according to each type of the sentence structure model of the map constructed in step S300, and if it is the content of a composition created by a user, analyzes the content of the composition according to each type of the sentence structure model of the map constructed in step S300 (S600).

[0154] The step S600 is shown in FIG. a,15b Explaining in detail with reference to the above, the reading and writing improvement guide device 100 reads an explanatory or expository sentence used for reading comprehension, or the content of a composition created directly by a user according to a specific topic (S601), and divides the read sentence or the content of the composition into at least one sentence (S602).

[0155] In addition, the reading and writing improvement guide device 100 divides each sentence read in step S602 into unit components, extracts keywords for each unit component, and stores them in the database 300 (S603). For example, after dividing each sentence into two components, a main clause and a narrative clause, representative keywords for each component are extracted and stored.

[0156] After extracting and storing keywords for unit components of each sentence through step S603, the reading and writing improvement guide device 100 reads keywords for the entire sentence of the content of the read sentence or composition stored in the database 300 (S604).

[0157] In addition, the reading and writing improvement guide device 100 compares the structure of keywords for each unit component of the two sentences that have not been processed from the top of the keywords for the entire sentence read in step S604 with each type of the sentence structure model of the map constructed in step S300 (S605), and determines whether a matching type exists as a result of the comparison (S606).

[0158] If it is determined in step S606 that there is no type of sentence structure model that matches the structure of the keywords for each unit component of the two sentences, the reading and writing improvement guide device 100 generates and stores an analysis result of each type of one-level sentence structure model for the top sentence of the two sentences (S607), checks the relationship between the unmatched sentence and the next sentence (S608), and determines whether there is a match based on the results of the check (S609).

[0159] If the determination result of step S609 indicates a match, the reading and writing improvement guide device 100 continues to check the matching relationship with the next sentence and the sentence after that. If there are no more matching sentences, the reading and writing improvement guide device 100 compares the keywords of the last sentence or content and the previous sentence of the unmatched sentence with each type of the sentence structure model of the constructed map (S610), and determines whether or not there is a match as a result of the comparison (S611).

[0160] If a match is found in step S611, the reading and writing improvement guide device 100 determines whether the last sentence or content elaborates on a previous sentence of the unmatched sentence (S612). If the last sentence or content elaborates on a previous sentence of the unmatched sentence, the reading and writing improvement guide device 100 generates and stores a result of analyzing the sentences from the unmatched sentences to the last sentence that no longer has a matching sentence as the "support" of the substructure (S613), and then performs step S634 of determining whether it is the last sentence.

[0161] In addition, if no matching is found as a result of the determination in step S611, the reading and writing improvement guide device 100 generates and stores the results of adding and analyzing the sentences from the unmatched sentence to the last sentence that no longer has a matching sentence in the addition model (S614), and then performs step S634 to determine whether it is the last sentence.

[0162] In addition, if the result of the determination in step S612 is that the last sentence or the content of the sentence does not elaborate on the previous sentence of the unmatched sentence, the reading and writing improvement guide device 100 generates and stores the result of analyzing the sentence from the unmatched sentence to the last sentence that no longer has a matching sentence as a list type ``and'' in the additional model (S615), and then performs step S634 of determining whether it is the last sentence.

[0163] Meanwhile, if no match is found in step S609, the reading and writing improvement guide device 100 determines whether the unmatched sentence and the previous sentence are similar or identical (S616).

[0164] If it is determined in step S616 that the unmatched sentence and the previous sentence are similar or identical, the reading and writing improvement guide device 100 determines whether the unmatched sentence elaborates on the previous sentence (S617), and if the unmatched sentence elaborates on the previous sentence, generates and stores the result of analyzing the ``support'' in the substructure (S618), and then performs step S634 of determining whether it is the last sentence.

[0165] In addition, if the unmatched sentence and the previous sentence are not similar or identical as a result of the determination made in step S616, the reading and writing improvement guide device 100 generates and stores the results of addition and analysis in the addition model (S619), and then performs step S634 to determine whether the sentence is the last sentence.

[0166] In addition, if the unmatched sentence does not elaborate on a previous sentence as determined in step S617, the reading and writing improvement guide device 100 generates and stores the result of analyzing it as a list-type "and" in the additional model (S620), and then performs step S634 of determining whether it is the last sentence.

[0167] Meanwhile, if it is determined in step S606 that a sentence structure model type that matches the structure of the keywords for each unit component of the two sentences exists, the reading and writing improvement guide device 100 generates and stores an analysis result based on the type of the two-stage sentence structure model (S621), and determines whether it is the last sentence (S622).

[0168] If it is determined in step S622 that the sentence is the last sentence, the reading and writing improvement guide device 100 performs step S633 in which the analysis results are generated for the entire sentence by combining the analysis results according to each type of sentence structure model and each type of addition model.

[0169] On the other hand, if the result of the determination in step S622 is that it is not the last sentence, the reading and writing improvement guide device 100 compares the sentence next to the two sentences compared in step S605 with each type of the sentence structure model of the map constructed in step S300 by including the structure of keywords for each unit component (S623), and determines whether a matching type exists as a result of the comparison (S624).

[0170] If it is determined in step S624 that there is no type of sentence structure model that matches the structure of the keywords for each unit component of the three sentences, the reading and writing improvement guide device 100 generates and stores analysis results based on each type of two-level sentence structure model for the top two sentences out of the three sentences compared in step S623 (S625), and then performs step S608 to confirm the relationship between the unmatched sentence and the next sentence.

[0171] In contrast, if it is determined in step S624 that a sentence structure model type that matches the structure of the keywords for each unit component of the three sentences exists, the reading and writing improvement guide device 100 generates and stores the analysis results of each type of two-level or three-level sentence structure model for the three sentences compared in step S623 (S626), and determines whether it is the last sentence (S627).

[0172] If it is determined in step S627 that the sentence is the last sentence, the reading and writing improvement guide device 100 performs step S633 in which the analysis results of each type of sentence structure model and each type of addition model are combined to generate an analysis result for the entire sentence.

[0173] On the other hand, if the result of the determination in step S627 is that it is not the last sentence, the reading and writing improvement guide device 100 compares the next sentence of the three sentences compared in step S623 with each type of the sentence structure model of the map constructed in step S300 by including the structure of keywords for each unit component (S628), and determines whether a matching type exists as a result of the comparison (S629).

[0174] If it is determined in step S629 that there is no type of sentence structure model that matches the structure of the keywords for each unit component of the four sentences, the reading and writing improvement guide device 100 generates and stores analysis results for each type of two-level or three-level sentence structure model for the top three sentences out of the four sentences compared in step S628 (S630), and then performs step S608 to confirm the relationship between the sentence with no match and the next sentence.

[0175] In contrast, if it is determined in step S629 that a sentence structure model type that matches the structure of the keywords for each unit component of the four sentences exists, the reading and writing improvement guide device 100 generates and stores the analysis results of each type of two-level or three-level sentence structure model for the sentences compared in step S628 (S631), and determines whether it is the last sentence (S632).

[0176] If it is determined in step S632 that the sentence is not the last sentence, the reading and composition improvement guide device 100 repeats step S628. If it is the last sentence, the reading and composition improvement guide device 100 combines the results of analysis of the sentence used for reading or the content of the composition created by the user according to each type of sentence structure model and each type of additional model generated in steps S607, S613, S614, S615, S618, S619, S620, S621, S625, S626, S630, and S631 to generate an analysis result for the entire sentence (S633).

[0177] Meanwhile, after performing steps S613, S614, S615, S618, S619 and S620, the reading and writing improvement guide device 100 determines whether the sentence is the last sentence (S634), and if it is not the last sentence, compares the keywords of the previous sentence of the unmatched sentence and the next new sentence with each type of the sentence structure model (S635), and repeats steps S606 and after.

[0178] On the other hand, if it is determined in step S634 that the sentence is the last sentence, the reading and writing improvement guide device 100 performs step S633 in which the analysis results are combined according to each type of sentence structure model and each type of addition model to generate an analysis result for the entire sentence.

[0179] Referring again to FIG. 14, the reading and writing improvement guide device 100 generates guide information based on the results of analyzing the sentence used for reading or the entire sentence of the content of the composition created by the user according to each type of sentence structure model in step S600, and provides the generated guide information to the user terminal 200 so that the user can practice reading or writing (S700).

[0180] FIG. 16 is a diagram showing a hardware structure of a device for guiding improvement of reading and writing skills according to an embodiment of the present invention.

[0181] As shown in FIG. 16, the hardware structure of the reading and writing improvement guide device 100 includes a processor 1000, a memory 2000, a user interface 3000, a database interface 4000, a network interface 5000, a Web server 6000, and the like.

[0182] The processor 1000 is configured to analyze a structure for one sentence or a linking structure for at least two sentences by separating it into a superstructure and a substructure for an inputted given sentence or a composition created by a user, and to guide the reading or composition. That is, the processor 1000 is configured to compare keywords between each sentence for the inputted given sentence or a composition created by a user with each type, sequentially check whether a matching type exists up to the last sentence, and if the matching type exists, analyze it as the superstructure, and if the matching type does not exist, analyze it as an additional model of the superstructure or a substructure, thereby guiding the reading or composition.

[0183] The user interface 3000 provides an input and output interface to a user by using a GUI (Graphical User Interface).

[0184] The database interface 4000 provides an interface between a database and a hardware structure.

[0185] The network interface 5000 provides a network connection between devices owned by a user.

[0186] The Web server 6000 provides a means for users to access the hardware structure via a network. Most users can connect to the reading and writing improvement guide device 100 by remotely connecting to the Web server.

[0187] The steps of the above-described configuration or method can be realized as computer readable codes on a computer readable recording medium or transmitted via a transmission medium. The computer readable recording medium is a data storage device capable of storing data which can be read by a computer system.

[0188] Examples of computer readable recording media include, but are not limited to, databases, ROMs, RAMs, CD-ROMs, DVDs, magnetic tapes, floppy disks, and optical data storage devices. Transmission media can include carrier waves transmitted over the Internet or various types of communication channels.

[0189] The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

[0190] In addition, at least one of the components applied to the present invention may include or be implemented by a processor such as a central processing unit (CPU), microprocessor, etc., that performs its respective function, and two or more of the components may be combined into a single component that can perform all operations or functions for the two or more combined components.

[0191] In addition, a part of at least one of the components applied to the present invention may be performed by other components among these components, and communication between the components may be performed via a bus (not shown).

[0192] As described above, the present invention has been described with reference to the embodiments shown in the drawings, but these are merely illustrative, and a person skilled in the art will understand that various modifications and equivalent embodiments are possible. Therefore, the technical scope of the present invention should be determined by the following claims. [Industrial Applicability]

[0193] The present invention models the structure of one sentence or the type of connection structure of at least two or more sentences to construct a map, and applies the constructed map to reading comprehension and writing to learn types of logical development methods including deduction, induction, and dialectic. Therefore, a user can grasp the subject matter and complete a logical composition by himself / herself without the help of others, thereby greatly improving reading comprehension and writing skills.

[0194] In addition, the present invention supports learning reading and writing through maps that are modeled differently for each level set based on age and ability, so that each user can improve their ability through evaluation and checking of reading and writing that suits their own level, and can naturally achieve their learning goals through interactive writing checks.

Claims

1. 1. A processor-guided method for improving reading or writing skills, comprising: A pattern modeling step of modeling and storing a pattern of a structure for one sentence or a linking structure for at least two sentences in advance; A sentence analysis map construction step of constructing a map for analyzing a predetermined sentence required for reading comprehension or a composition created by a user according to each of the modeled types; Analyzing a sentence structure model type, which is a structure for one sentence or a linked structure for at least two sentences, for an inputted given sentence or a composition created by a user, by separating the type into a superstructure consisting of a premise or a topic and a substructure consisting of a support; guiding said reading or writing; having The analyzing step includes: A sentence analysis step of analyzing the inputted given sentence based on each modeled type and the sentence analysis map. The guiding step includes: providing the analyzed result together with the original text of the sentence to the user terminal, thereby enabling the user to correct errors in reading and practice reading correctly by learning a logical development method including deduction, induction, and dialectic for the inputted predetermined sentence; The sentence analysis step includes: reading the inputted predetermined sentence and dividing it into each sentence; Dividing each of the divided sentences into unit components and extracting keywords; comparing the extracted keywords between each of the sentences with each of the modeled patterns to check whether a matching pattern exists up to the last sentence; generating an analysis result including a superstructure, a substructure, or a combination thereof, including a one-line sentence structure model, an additional model, or a combination thereof, if there is no pattern matching the two-line pattern or the three-line pattern as a result of the checking; generating an analysis result for a superstructure including each type of the two-line sentence or three-line sentence structure model if a type matching the two-line sentence or three-line sentence structure model exists; a step of separating the inputted predetermined sentence into a superstructure and a substructure and reading the sentence; method.

2. 1. A processor-guided method for improving reading or writing skills, comprising: A pattern modeling step of modeling and storing a pattern of a structure for one sentence or a linking structure for at least two sentences in advance; A sentence analysis map construction step of constructing a map for analyzing a predetermined sentence required for reading comprehension or a composition created by a user according to each of the modeled types; Analyzing the inputted sentence or the composition created by the user by separating a sentence structure model type, which is a structure for one sentence or a linking structure for at least two sentences, into a superstructure consisting of a premise or a topic and a substructure consisting of a support; guiding said reading or writing; having The analyzing step includes: A composition processing step of analyzing the contents of the composition written by the user according to each modeled type and the composition analysis map, and providing the analyzed results to a user terminal; The composition processing step includes: A composition analysis step of analyzing the content of the composition received from the user terminal according to each modeled type and a composition analysis map; The guiding step includes: generating guide information related to the content of the composition based on the analysis result; providing the generated guide information to the user terminal so that the user can practice writing by correcting and rewriting the guide information; The composition analysis step includes: reading the content of the received composition and dividing it into individual sentences; Dividing each of the divided sentences into unit components and extracting keywords; comparing the extracted keywords between each of the sentences with each of the modeled patterns to check whether a matching pattern exists up to the last sentence; generating an analysis result including a superstructure, a substructure, or a combination thereof, including each type of one-line sentence, additional model, or a combination thereof, if the matching type is not present as a result of the confirmation; and generating an analysis result for a superstructure, including each type of two-line sentence or three-line sentence structure model, if the matching type is present as a result of the confirmation; and analyzing the received composition by separating the content into a superstructure and a substructure. method.

3. 1. A processor-guided method for improving reading or writing skills, comprising: A pattern modeling step of modeling and storing a pattern of a structure for one sentence or a linking structure for at least two sentences in advance; A sentence analysis map construction step of constructing a map for analyzing a predetermined sentence required for reading comprehension or a composition created by a user according to each of the modeled types; Analyzing the inputted sentence or the composition created by the user by separating a sentence structure model type, which is a structure for one sentence or a linking structure for at least two sentences, into a superstructure consisting of a premise or a topic and a substructure consisting of a support; guiding said reading or writing; having The typology modeling step includes: The method includes the step of modeling a structure of one sentence or a link structure type of at least two sentences into a sentence structure model and an additional model and storing the model; The sentence structure model is modeled into a total of 14 types, including one type of one-line sentence, four types of two-line sentences in which a development type conjunction "however" is connected between each sentence, three types of two-line sentences in which an opposition type conjunction "however" is connected between each sentence, two types of two-line sentences in which a conclusion type conjunction "therefore" is connected between each sentence, two types of three-line sentences in which a development type conjunction "however" is connected between the first sentence and the second sentence and a conclusion type conjunction "therefore" is connected between the second sentence and the third sentence, and two types of three-line sentences in which an opposition type conjunction "however" is connected between the first sentence and the second sentence and a conclusion type conjunction "therefore" is connected between the second sentence and the third sentence, and stored with reference to a sentence analysis model including a topic in a one-line sentence, a development type conjunction "however" is connected between each sentence in a two-line sentence, four types of two-line sentences in which a development type conjunction "however" is connected between each sentence in a two-line sentence, three types of three-line sentences in which an opposition type conjunction "however" is connected between the first sentence and the second sentence and a conclusion type conjunction "therefore" is connected between the second sentence and the third sentence, The additional model is stored in two types including a type for conjunctions not mentioned in the sentence structure model and a type for contents not related to the sentence structure model. method.

4. The analyzing step includes: a pre-processing step for generating sentence patterns for the inputted given sentences or user-generated compositions; an estimation step of estimating a type of sentence structure model, which is a structure for one sentence to be analyzed or a concatenation structure for at least two sentences, by inputting the generated pattern into a plurality of artificial intelligence learning models; The method of claim 1 , further comprising: generating an artificial intelligence analysis result according to the estimated type.

5. The artificial intelligence learning model is A pattern of each type of two-stage and three-stage sentence structure model is learned, and the sentence structure model is generated for each type. The learning data used for learning the artificial intelligence learning model is A two- or three-line sentence is divided into two unit components, and predetermined data is assigned according to whether the keywords extracted from each unit component are the same, similar, or opposite. The method according to claim 4.

6. The analyzing step includes: a pre-processing step for generating sentence patterns for the inputted given sentences or user-generated compositions; an estimation step of estimating a type of sentence structure model, which is a structure for one sentence to be analyzed or a concatenation structure for at least two sentences, by inputting the generated pattern into a plurality of artificial intelligence learning models; The method of claim 2 , further comprising: generating an artificial intelligence analysis result according to the estimated type.

7. The artificial intelligence learning model is A pattern of each type of two-stage and three-stage sentence structure model is learned, and the sentence structure model is generated for each type. The learning data used for learning the artificial intelligence learning model is A two- or three-line sentence is divided into two unit components, and predetermined data is assigned according to whether the keywords extracted from each unit component are the same, similar, or opposite. The method according to claim 6.

8. The analyzing step includes: a pre-processing step for generating sentence patterns for the inputted given sentences or user-generated compositions; an estimation step of estimating a type of sentence structure model, which is a structure for one sentence to be analyzed or a concatenation structure for at least two sentences, by inputting the generated pattern into a plurality of artificial intelligence learning models; The method of claim 3 , further comprising: generating an artificial intelligence analysis result according to the estimated type.

9. The artificial intelligence learning model is A pattern of each type of two-stage and three-stage sentence structure model is learned, and the sentence structure model is generated for each type. The learning data used for learning the artificial intelligence learning model is A two- or three-line sentence is divided into two unit components, and predetermined data is assigned according to whether the keywords extracted from each unit component are the same, similar, or opposite. The method according to claim 8.

10. An apparatus for guiding the improvement of reading or writing skills, comprising a processor configured to execute the method for guiding the improvement of reading or writing skills according to any one of claims 1 to 9.

Images