KR100931913B1 - Systme for learning languages by training sentence - Google Patents

Abstract

PURPOSE: A language learning system through sentence training is provided to effectively improve linguistic ability by using a network or a portable electronic device. CONSTITUTION: A language learning system comprises a learner terminal(10) and a learning management server(20). The learning terminal is formed with a terminal with key input function and internet access function like a desktop PC, a note book PC and a mobile terminal. The learner terminal comprises a client(120) accessing to a learning management server and an application(130) providing interface for language learning. The learner terminal comprises a key input unit(110) inputting key data for sentence test and an output unit(140) outputting video and audio data for learning. The learning management server comprises a UI part(210) providing user interface, a database storing sentences according to the level, and a sentence training management part(40).

Description

Language Learning System through Sentence Training {Systme For Learning Languages By Training Sentence}

The present invention relates to a language learning system through sentence training, and more particularly, to construct a database in which sentences are classified for each level, to provide a test to a learner by sampling a sentence according to the learner's level from the database, and testing the test. The present invention relates to a language learning system that can improve language proficiency through reconstruction of a language by constructing a sentence to be erased from the screen faster than the typing speed of the learner.

The present invention has a expectation order for humans to process language in the brain in the acquisition of foreign languages, and provides an efficient cognitive psychological interface to the learners, so that learners learn the expected order of information processing for their native languages and other foreign languages. It can be applied to all educational systems related to linguistics.

In general, foreign language learning is difficult due to not only the characteristics of the language itself, such as pronunciation, vocabulary, and grammar, but also because the word order is different from Korean.

For example, Koreans can learn relatively easily in Japanese with the same word order, but have difficulty in acquiring English with different word orders.

However, the conventional language teaching method focuses only on the characteristics of the language itself, such as pronunciation, vocabulary, and grammar of foreign languages, and does not provide a learning method for reconstructing a native language into a foreign language sentence by naturally recognizing the order of foreign languages. It is true.

In the case of conventional learning, after understanding or memorizing grammar, the process of interpreting the native language is performed. However, the native language is recognized immediately without interpretation. Therefore, foreign languages go through a process called interpretation, unlike the native language, and the language is actually processed in the head twice. In this way, it is processed as 'knowledge', not as a natural language, as the native language. do.

Accordingly, in the process of interpreting a foreign language as a native language, it becomes a habit to construct a word order of a foreign language according to a native language.

Acquisition of language means that the order in which information processing is expected should be made in the same way as the learning language.In the case of Koreans, subjects expect the next subject, but in the case of English users, subjects always expect the descriptors. Because of its characteristics, training to recognize English word order as it is is an efficient learning method that does not place a burden on the brain's ability than changing the word order. This is a language acquisition divece by Dr. Norm Chomsky. It is widely known as a theory.

Therefore, it is necessary not only to grasp meaning but also to comprehend the language in the foreign language learner.

By interpreting, ie, fitting the language to the native language, the parsing action of the language cannot be acquired, and training of cognition, that is, the parsing order of the language as it is, is actually a task in learning a foreign language.

As an object of the present invention is to solve the problems described above, the objective of the present invention is to obtain information order for humans to process language in the brain in the acquisition of foreign languages, and to provide an efficient cognitive psychological interface to the human language, It is to provide a language learning system through sentence training that can improve foreign language proficiency by acquiring the expected order of information processing in and other foreign languages.

In order to achieve the above object, the language learning system according to the present invention provides a test problem by sampling sentences stored in the database and sentences stored in each level, and displays the sentences only for a set start time. And a sentence training manager to determine whether the sentence input from the learner is the same as the test sentence.

The database may further include a learner information storage unit that stores learner level information, and test data that is a result of a learner inputting a test problem provided through the sentence storage unit and the sentence training management unit in which sentences are divided into a plurality of levels. It characterized in that it comprises a test history storage for storing.

The sentence training manager may extract a test sentence from a sentence stored in the database and set a start time of the test sentence and a test sentence extraction module that provides a test problem and set the start time of the test sentence. It is characterized in that it comprises a learning control module for controlling to be displayed, and determines whether the result input from the learner matches the provided test sentence.

The test sentence extraction module extracts learner level information stored in the database, and extracts a test sentence by sampling among sentences included in the learner level.

The learning control module may provide a sample sentence from a learner, measure an average typing speed with respect to the sample sentence from a learner, and set a time earlier than a time for inputting a sentence according to the typing speed as a start time. It is done.

And the learning control module is characterized in that the time set by the selected time in the range of 10 ~ 95% less than the time to input the sentence according to the typing speed, characterized in that the start time.

In addition, the learning control module is characterized by adjusting the start time by increasing or decreasing the typing speed when the typing speed is improved or decreased by checking the typing speed of the learner.

The learning control module compares the sentence input by the learner with the test sentence, and if it does not match with the correct answer, determines that it is an incorrect answer.

In addition, the learning control module is characterized in that to upgrade the learner level to a higher level when the test result of the learner is more than the average typing speed of the learner and the correct answer rate is more than the set value.

In addition, the learning control module provides a test sentence when a learner first accesses to measure a learner's average typing speed, extracts a sentence at the lowest level, and then creates a test problem, and sets a correct answer rate at or above the learner's average typing speed. In the above case, it is characterized by setting a learner's level by repeating the process of asking a test question with a sentence of a higher level.

As described above, the language learning system through sentence training according to the present invention can be easily learned by using a network or embedded in a portable electronic device.

In addition, it is possible to obtain an excellent learning effect in the speaking and writing process by inducing the learner to experience the information processing expectation order, that is, parsing practice, which is not performed in a general typing game or reading learning check process.

The present invention is different in the functional order of the foreign language and the native language, generally have two methods when a user for the purpose of foreign language acquisition acquires the language, one of which is to arrange the foreign language order in accordance with the expected order of information processing of the native language And the second method is to parse the information processing expectation of foreign language as it is (parsing method), and supports the second methodology. The second way, parsing methodology, is to acquire a foreign language.

Accordingly, the present invention is directed to deriving an efficient method in the brain memory procedure for the mechanism by which humans use language, and to allowing a user to efficiently acquire a foreign language.

Human memory has two mechanisms: procedural memory and anecdotal memory. The brain controls the procedure, and in particular, the language is coordinated by the Carlosome-Ismus, Wernicke, Broca, and the cerebral limbic system. There is a characteristic.

Referring to the cerebral language procedure, which is the core concept of the present invention, first, the input and output processes of the language go through different procedures. That is, when inputting a language, the arrangement of sentences is not accurately stored, and only the order of important parts of the syntax structure is detected and processed.

However, when outputting a language, the arrangement of words constituting the sentence must be correctly planned so that the output process appears. If this process is well done, you will have Broka fluency for the language. You see.

At this time, there is a semantic grasp step and semantic reconstruction step by the input procedure. Generally, semantic grasp is possible without broca fluency, but broca fluency has a semantic reconstruction step.

Therefore, the input process has two levels of difficulty, and in the process of foreigners acquiring a specific language, reading and listening are possible with some degree of learning, but writing and speaking must go through a reconstruction of meaning. It is a common phenomenon to have a disparity in speaking, reading, listening and writing areas.

In order to solve the imbalance between the input and output of the language, the present invention utilizes a typing game among language learning tools, and adds a step for reconstructing the meaning in the language input process. At this time, the process of reconstructing the semantics of the language, that is, having the information processing order of the language processing, must be performed. Accordingly, the user is required to secure Broca fluency in the language input and output process for the sentence. You will be provided with the effect.

Accordingly, through the structural features of the present invention, a general user who does not have a special learning objective or learning technique related to the information processing expected order (hereinafter, referred to as parsing) for reconstructing a language in the language input process is a general learning method. It provides a parsing experience essential for language reconstruction that cannot be reached by reading process, which has a direct effect on language output process, namely speaking and writing.

In the present invention, a language whose language reconstruction is not comprehensible is not completed when the sentence is presented because the data processing expected order action, that is, the parsing action is not complete and thus the word order is confused and the word order cannot be reconstructed as it is. It was derived in consideration of the phenomenon of not recognizing the words located in the part.When it is applied to the typing game, the sentence that is not interpreted is posted to require the speed faster than the typing speed of the user. If the end of the sentence disappears without hitting, the sentence that is not interpreted is based on the principle that you can no longer enter it.

That is, when language is input to human beings in the form of listening or reading, the structure of the sentence and the word order are not involved in understanding the language that is input, but the information necessary for reconstructing and outputting the language is based on the structure of the sentence. It is based on the characteristic that cerebral actions in language input and output, understanding and reconstruction are not equal.

In order to explain this again, in the case of a foreign language sentence having a sufficient understanding, the sentence can be memorized and reconstructed within a specific start time. The phenomenon that cannot be reconstructed is used for learning.

Therefore, in the case of a national language sentence, a sentence composed of more than 20 words is given and disappeared after being given voice or character information, but in the case of a foreign language, it can be processed only within 4 to 14 words on average. This phenomenon is demonstrated by famous statistics, such as the American Editors' Association's readability index according to age.

Therefore, in the conventional way of practicing interpretation or writing through grammar,

Because of the application of simple rules of grammar as a tool of interpretation, despite the fact that they are syntactically consistent, they do not actually constitute meanings or have opposite meanings. I have not been able to. The present invention enables parsing training that conventional learning methods cannot have in such a way that a lot of sentences are encountered, a syntax structure is learned, and longer sentences are handled, which is an important process inherent to the nature of the language.

That is, anyone can derive grammar from a language, but it is impossible to reconstruct a language using only grammar and words. In other words, a calculus can be derived by setting 1 + 1 = 2 as axiom, but it is impossible to prove axiom through calculus.

The parsing training and methodology according to the present invention is a more scientific training method in view of the impossibility of language reconstruction through the grammar, compared to the existing language acquisition through grammar and interpretation. Compared with the existing interpretation method to be acquired, the sentence is provided according to the parsing ability of the user, and the improvement is detected so that a higher level of experience can be experienced, and thus, it is superior in acquiring the ability to reconstruct a foreign language.

Hereinafter, specific embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram of a language learning system according to a preferred embodiment of the present invention.

The present invention can be provided not only in a network environment using an internet network, but also downloaded and installed in a terminal such as a computer or a notebook or embedded in an electronic device such as an electronic dictionary, a PMP, a PDA, a mobile terminal, a portable internet terminal, and the like. The present invention will be defined as a language learning system in a comprehensive term applicable to all the above embodiments.

1 is a system configuration diagram according to a first embodiment of the present invention.

Referring to FIG. 1, a language learning system according to a first embodiment of the present invention may be provided in a network environment using an internet network, and may include a learner terminal 10 and a learning management server 20.

Here, the learner terminal 10 is capable of inputting a key such as a desktop, a notebook, a mobile terminal, and any terminal capable of accessing the Internet.

The learner terminal 10 includes a client 120 for accessing a learning management server, an application 130 for providing an interface for language learning, an input unit 140 for inputting key data for a sentence test, and learning for learning. It may be configured to include an output unit 150 for outputting video and audio data.

In addition, the learning management server 20 samples the stored database 30 and the sentences stored in the database, in which the UI unit 210 and sentences that allow the access of the learner terminal and provide a user interface for a test are divided by levels. It may be configured to include a sentence training management unit 40 to provide a test problem, display the sentence only for a set start time, and determines whether the sentence input from the learner matches the test sentence.

2 is a system configuration diagram according to a second embodiment of the present invention.

Referring to FIG. 2, the language learning system according to the second exemplary embodiment of the present invention is provided when the language learning system is embedded in the electronic device 50, and receives an input unit 510 and a sentence for receiving key input from a learner and generating key data. A test problem is provided by sampling the stored database 30 and the sentences stored in the database by level, and displaying the sentence only during a set start time, and whether the sentence input from the learner matches the test sentence. It may include a sentence training management unit 40 for determining, an output unit 520 for outputting audio and video signals to provide an interface for learning, and a control unit 530 for controlling the entire device.

The language learning system according to the second embodiment may be implemented at a firmware, an operating system, and an application level embedded in an electronic device.

The first embodiment is the same as the language learning system except that the system manages learners on a network in a server, and the second embodiment is embedded in an electronic device and managed in units of learners. The method will be described in an integrated manner.

The language learning system according to the present invention includes a database 30 and a sentence training manager 40.

3 is a detailed block diagram of a database according to a preferred embodiment of the present invention.

Referring to FIG. 3, the database 30 is provided through a sentence storage unit 320 and a learner information storage unit 310 for storing the learner's level information, in which sentences are divided into a plurality of levels, and the sentence training manager. It may be configured to include a test history storage unit 330 for storing the test data that is the result of the learner input for the test problem.

The database 30 may be implemented as a single device, a network interface, or a mobile phone public network. In an embodiment of the single device, the database 30 may be mounted in a file system and a slot system, and a network adapter and a mobile phone. If it is made of a public network, unlike the terminal may be mounted on a separate server.

Sentences stored in the sentence storage unit 320 of the database 30 may be divided into levels in consideration of sentence length, vocabulary, and format.

For example, the word or spacing unit and the apostrophe (') are calculated for the sentence parsing unit, that is, "word count", "number of spaces + 1", "number of spaces + 1 + number of apostrophes", etc. The parsing number can be calculated and the level of the sentence can be classified according to the parsing number.

I got away. Number of parses: 3 (2 stripe rewriting + 1)

I wasn't threre. Number of parses: 3 (2 strips + 1) or

                           4 (2 strips + 1 apostrophe + 1)

If you were there, then they wouldn't get fired.

                   Number of parses: 9 (8 strips + 1) or

                            10 (9 Strips + 1 Apostrophe + 1)

Here, the apostrophe is formed by combining two words, and thus can be divided into one parsing unit or two parsing units. You can also consider the number of commas.

As described above, the level of the sentence may be distinguished only by the number of parsing, but the level may be classified in consideration of the word level, the grammar, and the sentence form together with the number of parsing.

The words can be divided into difficulty levels such as living English, middle school level, high school level, TOEFL level, TOEIC level, GRE level, and grammar is relatively easy grammar (for example, To infinitive) and difficult grammar (for example, Related sentences), and the sentence form can be divided into easy form (1 form) and difficult form (5 form), so that the level of sentences can be classified by considering the word, grammar, and sentence form together with the number of parsing. Can be.

For example, at the lowest level, the number of parsing is 4 or less, the word level is composed of living English level, the sentence form is one sentence, the upper level is between 4 and 7 parsing, and the word level is middle school. 3500 words and sentence types can be divided into sentence types, such as a sentence composed of 1 form, 2 forms.

The sentence training manager 40 extracts a test sentence through sampling based on a learner level among sentences stored in the database 30 and provides a test sentence extraction module 410 for providing a test problem and a start time of the test sentence. And a learning control module 420 configured to control the test sentence to be displayed only during a set start time, and determine whether a result input from a learner matches the provided test sentence.

Here, the start time means a time when the test sentence is displayed on the screen, and the start time is set to a time smaller than the time required for inputting all the sentences according to the average typing speed of the learner.

To this end, the learning control module provides a sample sentence from the learner, measures the average typing speed for the sample sentence from the learner, and inputs the time required for inputting all the sentences according to the measured average typing speed. It calculates and sets time shorter than the said input time as a sentence start time.

Here, the sentence start time may be set smaller than the input time, and may be set to 5 to 90% smaller than the input time in order to increase the learning effect. The sentence start time may be selected in consideration of the level of the sentence level and the degree of understanding of the learner. For example, a high level learner who has excellent language reconstruction and parsing ability can input the entire sentence even if the sentence start time is small. However, a low level learner can consider the learner level because it is difficult to enter a sentence even if the sentence start time is long. The sentence start time can be set.

For example, if a 20-character sentence is entered in 5 seconds, the learner's typing is 240 strokes per minute. When the average typing number of learners is measured at 240 strokes per minute, when the sentence is 16 characters, the input time required to input all the sentences is 4 seconds, and the sentence start time is set to be less than 4 seconds, for example, 3 seconds. Can be.

Therefore, the sentence start time can be calculated from the following equation.

Sentence start time = {(text length × 60 seconds) / average typing per minute} × ▽

Is a number greater than or equal to zero and less than one. For example, ▽ may be distributed in the range of 0.1 to 0.95 and may be selected by comprehensively considering the sentence level, word level, parsing number, grammar level, learner level, and the like.

The reason for setting the sentence start time smaller than the input time as described above is that since the arrangement of words in the human language processing procedure keeps pace with the voice memory and goes through the language reconstruction process, the auxiliary process and the language reconstruction process are performed. Only a rough sentence can be memorized in a short time to enter a sentence.

Therefore, in the case of a sentence understood to be able to reconstruct the language, even if the sentence start time is set smaller than the input time, input is possible, but in case of a sentence that is not understood, the input fails.

In other words, if a sentence is started for a shorter time than the input ability, the user may not be able to input successfully due to the phenomenon of not having a word arrangement procedure in the information processing expected order of a sentence that is not understood by the human language intelligence process.

Through the above process, the learner does not have special learning objectives related to parsing the language in the language input process, and parsing is essential for language reconstruction that is not reachable by the reading process of books, which is a general learning method for a general user. The experience provides a direct effect on the language output process, ie speaking and writing.

4 is a flowchart schematically showing a language learning method according to a preferred embodiment of the present invention, and FIG. 5 is an exemplary view according to FIG. 4.

4 and 5, a learner first loads user information for language learning.

Here, in the case of the first embodiment, a learner may access the learning management server through a learner terminal and use a language learning service through a login process for learner authentication. Here, the authentication key of the authorized authority may be used in the login process. In the second embodiment embedded in the electronic device, the embedded language learning system is activated by key input.

Next, the learner is identified (identification code 0110), and it is checked whether there is the learner level information (step 2) stored in the learner information storage unit of the database.

When the learner level information is present in the database, the test sentence extraction module analyzes the learner level information stored in the database, extracts test sentences by sampling among sentences included in the learner level, and provides a test problem to the learner.

More specifically, the test sentence extraction module generates a sentence list by extracting a test sentence by sampling among sentences included in the learner level, and when the sentence list is generated, the sentences included in the sentence list are sequentially displayed on the screen. To control.

When the test sentence is displayed on the screen as described above, the learner inputs the sentence in the same manner through the input means. Here, the input means includes all the input means that can configure a keyboard, such as a keyboard, a keypad, a mouse, a touch screen.

When the sentence start time set according to the learner's typing speed is finished, the sentence is deleted from the screen.

When the sentence is erased from the screen, the learner completes the remaining sentence that was not input after being erased through a parsing process for reconstruction of meaning.

Here, when a learner is given a test sentence, the learner may increase the parsing effect when learning while reading the sentence given with the typing input.

When the learner input is terminated through the above process, the learning control module compares whether the input sentence matches the test sentence and determines a correct answer or a wrong answer.

When the test for the sentences included in all sentence lists is finished, the learning control module stores the test results input from the learner in a database, and checks the correct answer rate of the test answers to determine whether the level is downgraded, maintained, or upgraded. You decide.

For example, if the batter speed and the correct answer rate exceeds the average batter speed and exceeds the set value, the learner sentence level may be upgraded. If the batter speed or the correct answer rate is less than the set value, the learner sentence level may be downgraded.

For example, if the learner's average typing speed is 300 typing and the correct answer rate for upgrading is 90%, the learner's test result is more than 300 typing, and if the correct answer rate is 90% or more, the sentence level is upgraded. Keep your level.

In addition, if the criterion for downgrading is set to an average typing speed of -20 and a correct answer rate of 70% or less, the test result downgrades the learner's sentence level if the typing speed is 280 or less, or 70% or less.

The criteria for the level upgrade and the downgrade may be set differently for each level, and may be arbitrarily determined in consideration of the sentence level and the number of sentences included in each level.

When the level is determined as described above, the test data, which is the test result, is stored in the test history storage unit of the database.

Here, the test data may include information such as a student identification code, a test date, a list of known sentences, a list of unknown sentences, a learner level, and the like.

Therefore, the learner can check the test history stored in the test history storage unit and can analyze the progress of improvement and achievement.

If there is no learner level information in the database, a learner test sentence is provided to the learner to measure the learner's typing speed. Here, the sentence start time is not applied when the typing test sentence is provided.

Through the input of the typing test sentence, the learning control module measures the number of typing per minute of the learner and calculates a sentence start time accordingly.

Subsequently, the test sentence extraction module sequentially extracts test sentences from the lowest level to the highest level and provides them to the learner.

More specifically, the test sentence is extracted by sampling among the sentences of the lowest level, and when the input from the learner does not exceed the set up criteria as a result of the test determination by the learning control module, the learner is determined as the lowest level, and the set level. When the up criterion is exceeded, the learner's level is determined by repeating the process of extracting a test sentence by sampling among the sentences of the next higher level.

On the other hand, the present invention is not limited to the above-described specific embodiments, anyone of ordinary skill in the art belonging to the present invention without departing from the gist of the invention claimed in the claims, the order of the processing procedure Various changes may be made, such as changing the type of hardware on which the component is to be located.

1 is a system configuration diagram according to a first embodiment of the present invention.

2 is a system configuration diagram according to a second embodiment of the present invention.

3 is a detailed block diagram of a database according to a preferred embodiment of the present invention.

4 is a flowchart schematically showing a language learning method according to a preferred embodiment of the present invention, and FIG. 5 is an exemplary view according to FIG. 4.

* Description of the symbols for the main parts of the drawings *

10: learner terminal 20: learning management server

30: database 310: learner information storage unit

320: sentence storage unit 330: test history storage unit

40: sentence training management unit 410: test sentence extraction module

420: learning control module

Claims (10)

A database in which sentences are classified and stored for each level; And A test sentence extraction module for providing a test problem by extracting a test sentence through sampling among sentences stored in the database; A sentence including a learning control module for setting a sentence start time of the test sentence, controlling the test sentence to be displayed only during the set sentence start time, and determining whether a result input from a learner matches the provided test sentence Including training management, The learning control module provides a sample sentence to a learner, measures an average typing speed of the sample sentence from the learner, and sets a time earlier than a time for inputting a sentence according to the typing speed as the sentence start time. Language learning system. The method of claim 1, The database is A learner information storage unit for storing learner level information; A sentence storage unit for storing sentences divided into a plurality of levels; And a test history storage unit configured to store test data which is a result of a learner inputting a test problem provided through the sentence training manager. delete The method of claim 1, The test sentence extraction module And extracting the learner level information stored in the database to sample test sentences from the sentences included in the learner level. delete The method of claim 1, The learning control module The language learning system, characterized in that for setting the start time to a time smaller than the selected ratio of 10 ~ 95% range than the time to enter the sentence according to the typing speed. The method of claim 1, The learning control module Checking the learner's typing speed, if the typing speed is improved or decreased, the language learning system, characterized in that for adjusting the start time as the typing speed is increased or decreased. The method of claim 1, The learning control module The language learning system, characterized in that the learner compares the input sentence and the test sentence is matched, if it is not the correct answer is determined as the wrong answer, and the correct answer and the wrong answer is stored in the database. The method of claim 8, The learning control module The language learning system, characterized in that for upgrading the learner level to a higher level when the typing speed of the learner is more than the learner's average typing speed and the correct answer rate is above the set value. The method of claim 8, The learning control module When the learner accesses for the first time, the test sentence is provided to measure the learner's average typing speed, the lowest level sentence is extracted, and the test question is asked. Language learning system, characterized in that to set the level of learners by repeating the process of questions to test questions.

Images