KR102356376B1 - System for providing english learning service using part of speech from sentence elements - Google Patents

Abstract

Provided is a system for providing an English learning service using parts of speech from sentence elements. The system comprises: a user terminal which indicates article, possessive, and participle, which are at least one sentence element to be combined with the noun included in at least one sentence, as adjective (Adj), auxiliary verb, which adds a meaning to verb, as adverb (Adv), gerund, which functions as noun, as noun (N), and to-infinitive as at least one of noun (N), adjective (Adj), and adverb (Adv), and outputs the same; and an English learning service providing server including a storage unit which tags at least one sentence element in at least one sentence and stores the same in a database; an adjective mapping unit which maps article, possessive, and participle, combined with noun in the at least one sentence component, into an adjective, and stores the same; an adverb mapping unit which maps auxiliary verb, which adds a meaning to verb, into adverb, and stores the same; a noun mapping unit which maps gerund, which functions as noun, as noun (N), and stores the same; a to-infinitive mapping unit which maps to-infinitive into any one of noun (N), adjective (Adj), and adverb (Adv), and stores the same; and an output unit which displays a sentence to be learned as five parts of speech including nouns (N), verb (V), adjective (Adj), adverb (Adv), and preposition (Prep) when the sentence to be learned is displayed on the user terminal, and outputs the same. Therefore, the reading skills of learners can be increased.

Description

English learning service providing system using five parts of speech of sentence elements {SYSTEM FOR PROVIDING ENGLISH LEARNING SERVICE USING PART OF SPEECH FROM SENTENCE ELEMENTS}

The present invention relates to a system for providing an English learning service using 5-part-of-speech of sentence components, redefining 8 parts of speech into 5 parts of speech, dividing sentence components into 4 components and encoding them, and guides the translation order in English and Korean, respectively. provide a platform that can

Grammar is not just learning the rules of a language, but is the basis for acquiring a specific language and using it for communication and other purposes. Grammar is an essential element in learning a language, and especially, plays a very important role in learning the language in a non-native-speaking country without being a native speaker. In the field of foreign language education, the grammar of English grammar learning textbooks can be viewed as a kind of practical grammar concept that is set to be beneficial to grammar learning by simplifying the grammar system for learners. However, in the process of simplification of the system, if it is oversimplified to the point of being regarded as an inaccurate explanation in terms of grammar due to excessive generalization or simplification, learners may have to learn the grammar system from the beginning the moment they enter the intermediate/advanced learning stage. In English education, grammatical terms are also located within a large category of grammar. In order to simply describe or explain language rules, it is most efficient to use grammatical terms that represent core concepts.

At this time, a method for intuitively understanding the grammar system has been researched and developed. (Published on May 24, 2013), part-of-speech separators including nouns (N), verbs (V), adjectives (Adj), adverbs (Adv), and conjunctions (Conj) are written and printed, respectively, and the type of grammar Divide the pre-stored sentences into parts of speech to be classified according to In order to analyze English sentences by processing the remaining fields that are not filled in as blanks so that they are not printed, each part-of-speech of English is displayed as a figure, and the organic coupling relationship between related grammatical elements is diagrammed. In displaying the 8 parts of speech in figures, the verb is ○ (circle or oval), the noun is [ ], the pronoun is , the adjective is ( ), the adverb is < >, the conjunction is ◇, the preposition is ▽, and the interjection is <! Based on the method of displaying in a figure such as >, the to-infinitive, circular infinitive, present participle and gerund, and past participle are used as different shapes among quasi-verbs that have almost the same properties as verbs, but are used as parts of speech other than verbs in sentences. Each method for displaying is disclosed.

However, the above configuration is only a schematic diagram of a method for distinguishing verbs, nouns, pronouns, adjectives, adverbs, conjunctions, prepositions, and exclamations in English according to the 8 part-of-speech system. it is not Textbooks for English learning include grammar elements and are provided to learners. In particular, since the parts of speech and sentence elements that make up the sentences are important meta-languages used in language education, education that meets the educational purpose You are obliged to present the content. Part-of-speech is indispensable for any grammatical skill, and although these parts of speech and sentence elements form the basis of school grammar and general grammar, they are sometimes used ambiguously without accurate cutting and verification, and They are mixed in a state of not being clearly distinguished. The parts of speech and sentence elements in English refer to different things, and if they are used in a mixture, it causes a big problem in setting grammatical terms for English education. In conclusion, it is necessary to discuss and improve the definitions of parts of speech and sentence elements that are currently used in English textbooks and various examination books. It requires teaching methods and systems.

In one embodiment of the present invention, the 8 parts of speech in general English are concluded with 5 parts of speech, and the 5 parts of speech are made, and among the sentence components, the verb, the quasi-verb, the relative pronoun/relational adverb, and the adverb phrase/adverb clause 4 sentence components By symbolizing by star, 5 parts of speech are written as alphabet symbols, 4 sentence components are displayed as symbols, and the order of English direct translation and Korean literal translation order is indicated by the number of sentence components in units of sentence components, so that complex parts of speech and grammar are converted into semantic units. It is possible to improve reading comprehension skills by making it possible to classify the sentence components, to increase the sentence construction ability by indicating the sentence constituents with symbols, and to improve the reading ability of which part should be interpreted by breaking it into a phrase or clause. It is possible to provide an English learning service providing system using 5 parts of speech. However, the technical task to be achieved by the present embodiment is not limited to the technical task as described above, and other technical tasks may exist.

As a technical means for achieving the above-described technical problem, an embodiment of the present invention provides at least one sentence component, which is an article, a possessive, and a participle, combined with a noun included in at least one sentence as an adj, An auxiliary verb that modifies a verb as an adverb (Adv), a gerund that functions like a noun as a noun (N), and a to-infinitive as a noun (N), an adjective (Adj), or an adverb (Adv). A storage unit that tags at least one sentence element in a user terminal and at least one sentence with a tag to form a database, and at least one sentence element adjective mapping that maps and stores articles that combine with nouns, possessives, and participles into adjectives An adverb mapping unit that maps and stores auxiliary verbs that decorate parts and verbs as adverbs, a noun mapping unit that maps and stores gerunds that function like nouns to nouns (N), and to infinitives that are nouns (N) and adjectives (Adj). ) and an infinitive mapping unit that maps and stores any one of adverbs (Adv), when a sentence to be learned in the user terminal is marked, noun (N), verb (V), adjective (Adj), adverb (Adv) and preposition ( Prep) including an English learning service providing server including an output unit for displaying and outputting the 5 parts of speech.

According to any one of the above-described problem solving means of the present invention, 8 parts of speech in general English are reduced to 5 parts of speech to become 5 parts of speech, and among the sentence components, verbs, quasi-verbs, relative pronouns, relative adverbs, adverb phrases, By encoding each of the 4 sentence components of the adverbial clause, 5 parts of speech are written as alphabet symbols, the 4 sentence components are displayed as symbols, and the English direct translation order and the Korean literal translation order are displayed numerically by the sentence component unit, making complex parts of speech It can improve reading comprehension ability by making it possible to classify and grammar into semantic units, improve sentence construction ability by displaying sentence constituents with symbols, and improve reading ability about which part should be interpreted by breaking it into phrases or clauses. there is.

1 is a diagram for explaining an English learning service providing system using 5 parts of speech of sentence elements according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating a learning service providing server included in the system of FIG. 1 .
3 and 4 are diagrams for explaining an embodiment in which an English learning service using 5 parts-of-speech of sentence components according to an embodiment of the present invention is implemented.
5 is an operation flowchart illustrating a method of providing an English learning service using 5-part-of-speech conversion of sentence elements according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art can easily implement them. However, the present invention may be embodied in several different forms and is not limited to the embodiments described herein. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a part is "connected" with another part, this includes not only the case of being "directly connected" but also the case of being "electrically connected" with another element interposed therebetween. . In addition, when a part "includes" a certain component, it means that other components may be further included, rather than excluding other components, unless otherwise stated, and one or more other features However, it is to be understood that the existence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded in advance.

The terms "about", "substantially", etc. to the extent used throughout the specification are used in or close to the numerical value when manufacturing and material tolerances inherent in the stated meaning are presented, and are intended to enhance the understanding of the present invention. To help, precise or absolute figures are used to prevent unfair use by unconscionable infringers of the stated disclosure. As used throughout the specification of the present invention, the term "step of (to)" or "step of" does not mean "step for".

In this specification, a "part" includes a unit realized by hardware, a unit realized by software, and a unit realized using both. In addition, one unit may be implemented using two or more hardware, and two or more units may be implemented by one hardware. Meanwhile, '~ unit' is not limited to software or hardware, and '~ unit' may be configured to be in an addressable storage medium or to reproduce one or more processors. Thus, as an example, '~' denotes components such as software components, object-oriented software components, class components, and task components, and processes, functions, properties, and procedures. , subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables. The functions provided in the components and '~ units' may be combined into a smaller number of components and '~ units' or further separated into additional components and '~ units'. In addition, components and '~ units' may be implemented to play one or more CPUs in a device or secure multimedia card.

Some of the operations or functions described as being performed by the terminal, apparatus, or device in the present specification may be performed instead of by a server connected to the terminal, apparatus, or device. Similarly, some of the operations or functions described as being performed by the server may also be performed in a terminal, apparatus, or device connected to the server.

In this specification, some of the operations or functions described as mapping or matching with the terminal means mapping or matching the terminal's unique number or personal identification information, which is the identification data of the terminal. can be interpreted as

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

1 is a diagram for explaining an English learning service providing system using 5 parts of speech of sentence components according to an embodiment of the present invention. Referring to FIG. 1 , an English learning service providing system 1 using five parts of speech of sentence elements may include at least one user terminal 100 and a learning service providing server 300 . However, since the system 1 for providing an English learning service using the 5 parts of speech conversion of the sentence elements of FIG. 1 is only an embodiment of the present invention, the present invention is not limitedly interpreted through FIG. 1 .

At this time, each component of FIG. 1 is generally connected through a network (Network, 200). For example, as shown in FIG. 1 , at least one user terminal 100 may be connected to the learning service providing server 300 through the network 200 . In addition, the learning service providing server 300 may be connected to at least one user terminal 100 through the network 200 .

Here, the network refers to a connection structure in which information exchange is possible between each node, such as a plurality of terminals and servers, and an example of such a network includes a local area network (LAN), a wide area network (WAN: Wide Area Network), the Internet (WWW: World Wide Web), wired and wireless data communication networks, telephone networks, wired and wireless television networks, and the like. Examples of wireless data communication networks include 3G, 4G, 5G, 3rd Generation Partnership Project (3GPP), 5th Generation Partnership Project (5GPP), Long Term Evolution (LTE), World Interoperability for Microwave Access (WIMAX), Wi-Fi (Wi-Fi) , Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network), RF (Radio Frequency), Bluetooth (Bluetooth) network, NFC ( Near-Field Communication) networks, satellite broadcast networks, analog broadcast networks, Digital Multimedia Broadcasting (DMB) networks, and the like are included, but are not limited thereto.

In the following, the term at least one is defined as a term including the singular and the plural, and even if the at least one term does not exist, each element may exist in the singular or plural, and may mean the singular or plural. it will be self-evident In addition, that each component is provided in singular or plural may be changed according to embodiments.

The at least one user terminal 100 displays all words in a sentence as 5 parts of speech using an English learning service-related web page, app page, program or application using 5 parts of speech of sentence components, and constructs sentences in sentences It may be a terminal that displays an element by symbolizing it into four components. In this case, the user terminal 100 may be a terminal connected to a professor terminal (not shown).

Here, the at least one user terminal 100 may be implemented as a computer that can access a remote server or terminal through a network. Here, the computer may include, for example, navigation, a laptop equipped with a web browser, a desktop, and a laptop. In this case, the at least one user terminal 100 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one user terminal 100 is, for example, as a wireless communication device that guarantees portability and mobility, navigation, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS(Personal Handyphone System), PDA(Personal Digital Assistant), IMT(International Mobile Telecommunication)-2000, CDMA(Code Division Multiple Access)-2000, W-CDMA(W-Code Division Multiple Access), Wibro(Wireless Broadband Internet) ) terminal, a smart phone, a smart pad, a tablet PC, etc. may include all kinds of handheld-based wireless communication devices.

The learning service providing server 300 may be a server that provides an English learning service web page, an app page, a program, or an application using five parts-of-speech of sentence components. And, the learning service providing server 300, when outputting a sentence by selecting any one textbook from the user terminal 100, or when outputting a pre-stored sentence, converts English words included in each sentence into five parts-of-speech. It may be a server that divides and displays the sentence components of a sentence divided into 5 parts of speech with 4 symbols. It may be a server that maps and stores. In addition, the learning service providing server 300, when outputting a sentence from the user terminal 100, divides the English words included in the sentence into 5 parts of speech and displays them, displays the pre-stored sentence components with symbols, and translates them into Korean. It may be a server that displays the order and the English translation order, respectively.

Here, the learning service providing server 300 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, navigation, a laptop equipped with a web browser, a desktop, and a laptop.

2 is a block diagram for explaining a learning service providing server included in the system of FIG. 1, and FIGS. 3 and 4 are English learning services using 5 parts of speech of sentence components according to an embodiment of the present invention. It is a diagram for explaining an implemented embodiment.

Referring to FIG. 2 , the learning service providing server 300 includes a storage unit 310 , an adjective mapping unit 320 , an adverb mapping unit 330 , a noun mapping unit 340 , an infinitive mapping unit 350 , and an output. It may include a unit 360 and a translation order marking unit 370 .

An English learning service application using the learning service providing server 300 or another server (not shown) that operates in conjunction with at least one user terminal 100 according to an embodiment of the present invention, using 5-part speech of sentence components; When transmitting a program, an app page, a web page, etc., at least one user terminal 100 can install or open an English learning service application, a program, an app page, a web page, etc. using 5 parts of speech of sentence components. there is. In addition, a service program may be driven in at least one user terminal 100 using a script executed in a web browser. Here, the web browser is a program that enables the use of a web (WWW: World Wide Web) service, and refers to a program that receives and displays hypertext written in HTML (Hyper Text Mark-up Language), for example, Netscape. , Explorer, Chrome, and the like. In addition, the application means an application on the terminal, for example, includes an app (App) executed in a mobile terminal (smartphone).

Before describing FIG. 2, the basic concept of parts of speech and sentence components will be described. The basic concepts described below will not be repeatedly described with reference to FIG. 2 .

<Part of speech and sentence elements>

Generally speaking, words with similar characteristics are grouped and classified into one category. In the traditional grammar category, the term Parts of Speech is used, and in the modern English grammar, the term Word Classes is sometimes used. The representative part-of-speech of English is classified into 8 parts of speech. Nouns (Noun, N), pronouns (Pronoun, ProN) , verbs (Verb, V), adjectives (Adjective, Adj), adverbs (Adverb, Adv), prepositions (Preposition, Prep) ), Conjunction, Conj, and Interjection, Int. On the other hand, sentence elements classified as elements constituting a sentence are elements necessary for the language to form one sentence unit. Subject (Subject, S), verb (Verb, V), object (Object, O), and complement ( Complement, C), and categorization into modifiers can be seen as the categories of general school grammar.

There are many claims that the terms used to refer to parts of speech and sentence elements are inappropriate and need to be corrected, and since it can be considered as a reasonable argument to some extent, various studies related to this are being conducted. However, it cannot be said that the consistency of grammar rules is hindered by the use of wrong terms. Even if you use the wrong terminology, maintaining the consistency of the rules is a separate issue. If we look at the phenomenon of using the term itself incorrectly, not the error of the term itself, the representative example of the misuse of the term is the part-of-speech 'noun'. The term ‘noun’ is one of the representative eight parts of speech, and it is generally classified into ordinary, set, material, abstract, and proper nouns according to characteristics. Obviously, the term ‘noun’ itself is classified as a part-of-speech, and it can be said that there is no problem. However, in general school grammar, ‘noun’ is explained as a sentence element rather than a part-of-speech, resulting in a lack of clarity and consistency in grammar, thereby hindering the effectiveness of grammar learning.

<Part of speech ‘noun’ and sentence component ‘noun phrase’>

Various English teaching methods and textbooks describe or explain that ‘nouns can be used as subjects, objects, and complements of sentences’. This is a clear error in defining the part-of-speech 'noun' as the unit of 'subject', 'object', and 'bore' of the sentence element. A noun cannot play the role of a subject, object, or complement by itself, and in order for a noun to act as a sentence element, additional elements to the noun must be combined and transformed into a sentence element unit in the form of a 'noun phrase' until the sentence element 'subject' , object, and 'bore'. As shown in Table 1 below, the structure of English noun phrases is basically as follows.

Determiner + Noun
Determiner: Article
Progressive
demonstrative
Qantifier

The example sentences in Table 2 below are examples of comparing the case where the simple part-of-speech 'noun' is used in the place of the sentence element 'object' and the case where the 'noun phrase' is used.

(1) *John bought computer.
(2) John bought a computer.
(3) John bought his computer last year.
(4) John bought several computers last year.

In the above examples, the noun computer is used in place of the object, but example (1) is an inscription, whereas (2,3,4) is a main sentence. The structural difference between Inscription (1) and the rest of the example sentences is whether or not an additional vocabulary is attached in front of the computer used as the object to form a single phrase. In order to have consistency in the grammatical rule that a noun spoken in general school grammar can act as an object, all of the examples (1 to 4) must be the main sentence, but as a result, it can be seen that there is no consistency. To make up for this lack of consistency, another rule is created and explained that the indefinite article a/an or the definite article the must be used before singular nouns. If the rule that an article must be used before a singular countable noun is applied like this, in the case of example (3), even though there is no article in front of the computer, it is treated as a main sentence. Therefore, in order to explain example sentence (3), another grammatical explanation is needed as an exceptional rule that the possessive case and the article cannot be used together.

As described above, the basic grammatical rules are clarity, simplicity, consistency, and gradation. However, the grammatical explanation that ‘nouns can be used as subject, object, and complement’ used in school grammar does not fit any of clarity, simplicity, and consistency.

(5) I need water.

Also, in the case of 'water', which is classified as a noun in the same way as in Example (5), even if it is used without an article, it is treated as a main sentence. However, in the case of example (5), on the surface, the noun 'water' appears to be used in the place of the sentence element object, but it can be seen that this also takes the form of a noun phrase. Although not dealt with in school grammar, the types of articles are classified into the definite article the, the indefinite article a/an, and the unrelated article Ø. Table 4 below lists the types of articles in English.

Article Type Definite The Indefinite A/An Zero Ø

The irrelevant article Ø is a part of speech necessary for constructing a noun phrase as an article that is not visible on the surface, but does exist. Therefore, if the structure of example sentence (5) is simplified, I is the subject, need is the verb, and Ø water is the object (Object, NP). In the place of the sentence element object, ‘Ø + uncountable noun’, which is a noun phrase, is placed. Therefore, the explanation that the noun explained in school grammar or the study book can act as a sentence element is not suitable for explaining the grammar rule, and it infringes on the clarity, simplicity, and consistency of the grammar rule, resulting in lowering the learning effect. The fact that a noun as a part-of-speech cannot be used as a sentence element and can act as a sentence element in the form of a noun phrase is dealt with in various ways in general modern English grammar, and grammatical education for noun phrases is being conducted in school grammar as well. Therefore, rather than educating the role of noun phrases in the current school grammar as a new type of grammar, the part-of-speech and sentences are more clearly defined through the re-establishment of grammatical terms that hinder the simplicity, clarity, and consistency of existing teaching methods. Efforts are needed to separate the elements.

Therefore, in one embodiment of the present invention, the part-of-speech system and sentence component system, which were generally defined as 8 parts of speech and sentence components, are converted into 5 parts of speech, and the sentence components also have 4 sentence components. Divide it by , and mark it with a symbol. The premise that everything results in part-of-speech by designating parts of speech and sentence elements that play the role of adjectives as adjectives, parts of speech and sentence elements that act as adverbs as adverbs, and parts of speech and sentence elements that play the role of nouns as nouns. The following configuration will be described below. Accordingly, it is specified that the parts of speech and sentence components to be described below are different from the definitions of general parts of speech and sentence components.

Referring to FIG. 2 , in an embodiment of the present invention, all English words (part-of-speech and sentence components) are prepared on the premise that they result in part-of-speech. Therefore, as in the general definition, the part-of-speech and sentence components are not distinguished, and if they meet each meaning of the parts of speech defined below, they will be referred to as parts of speech, and if they meet the definitions of each sentence component defined below, they will be referred to as sentence components. do. Accordingly, the terms of parts of speech and sentence components may be used interchangeably.

<5 parts of speech>

The storage unit 310 may make a database by tagging at least one sentence component in at least one sentence with a tag. The morpheme part-of-speech tagger works to set the optimal morpheme part-of-speech for each word (morpheme) of the input sentence. Many language processing systems use a morpheme part-of-speech tagger for preprocessing, and the morpheme part-of-speech tagging performance plays a very important role in a language processing system using this. In order to improve the tagging performance, HMM (Hidden Markov Model), LHMM (Lexicalized Hidden Markov Model), CRF (Conditional Random Field) model, Transformation-based Learning, Memory-based Learning, Decision Tree, Maximum Entropy Model Statistical learning methods such as these may be used. In addition, in order to improve performance, a method of determining the result by combining other tagging models by voting may be used.

The morpheme part-of-speech tagging methods mainly use a statistical learning method using the corpus tagged with the morpheme part-of-speech. In English, the morpheme part-of-speech tagging methods based on the morpheme part-of-speech tagging corpus generally show an accuracy of 96% to 97% of word unit part-of-speech tagging. It has good performance as a preprocessor of a language processing system that processes word units, but is not good at all in a language processing system that processes sentence units such as machine translation. For example, when trying to automatically translate 100 sentences composed of an average of 20 words, if the morpheme part-of-speech tagging word accuracy rate is 97%, there is an error of 60 words (= 0.03 × 2,000 words). If it is assumed that 60 words have errors in different sentences, out of 100 sentences to be translated, 60 sentences each have one part-of-speech tagging error, and the process after parsing is performed. Therefore, there is still a lot of room and expectations for the performance improvement of the morpheme tagging model.

Accordingly, in an embodiment of the present invention, a method of adapting a morpheme part-of-speech tagging model learned from a part-of-speech tagged corpus of a general domain to a domain can be used without a corpus having a part-of-speech tagged corpus suitable for a specific domain. The adaptation method of the part-of-speech tagging model can use the LHMM-based English morpheme part-of-speech tagger learned with the PennTree Bank corpus, which is widely used in the English-speaking world, and can once again undergo the task of changing the part-of-speech according to the definition of the present invention.

<LHMM-based English morpheme part-of-speech tagger>

The LHMM method used in English morpheme part-of-speech tagger is almost identical to the well-known HMM method. The difference is that it not only transfers between parts of speech, but also includes vocabulary. The general HMM method has a disadvantage in that it does not consider transitions that vary between vocabularies by considering transitions only between morphemes and parts-of-speech. Due to these shortcomings, the performance tends to be lower than methodologies such as Maximum Entropy Model and CRF that can utilize vocabulary information. The LHMM method is a method of using these lexical features in the HMM method. The LHMM for morpheme part-of-speech tagging can be defined as a problem of finding a probability value at which the arrangement of morpheme parts-of-speech tags is maximized by a general HMM as shown in Equation 1 below.

Equation 1 is a list of input words of length n w1, ... , wn is a problem of finding the maximum probability of listing the parts-of-speech tags corresponding to each word. Equation 1 is the same as the general HMM equation. The only difference between LHMM and HMM is the addition of parts-of-speech combined with vocabulary to the part-of-speech set. In Equation 1, P(wi|ti) is called the output probability, and P(ti|ti-1,ti-2) is called the transition probability. The output probability and transition probability use statistical values for each probability from the part-of-speech tagged corpus as predicted values.

In the English morpheme part-of-speech tagger used in an embodiment of the present invention, N vocabulary can be selected by introducing a method using vocabulary qualities. Compared to the CRF or Maximum Entropy Model method, the N selected vocabulary words and the POS tagger can solve the problem of difficulty in using the vocabulary, suffix, and capitalization in the general HMM method. It has the advantage of being able to use various features along with the model simplicity, which is the advantage of

<LHMM-based English tagger domain adaptation method>

The method of adapting the corpus learning-based morpheme part-of-speech tagging model to fit the domain can be simply divided into two methods. The first method is a new learning method using a corpus tagged with a morpheme and part-of-speech suitable for a specific domain. This method has a disadvantage in that it takes a lot of time and money because it is necessary to manually attach a morpheme part-of-speech to a large-capacity corpus. The second method is to change the existing learned tagging model to fit the domain to which the value already learned from the tagging model is to be applied using the domain's raw corpus. According to this method, it has the advantage of not building a new domain learning corpus, but the problem of how to change the existing trained model using the domain primitive corpus remains. In an embodiment of the present invention, the method for solving the domain adaptation problem of the second method may be used.

In the LHMM-based morpheme part-of-speech tagging method, the values learned from the learning corpus are the transition probability P(ti|ti-1,ti-2) and the output probability P(wi|ti). Therefore, it must be reconfigured to fit the domain to which the transition probability and output probability are to be applied. The domain adaptation method builds an auto-tagged corpus by first automatically tagging the raw corpus of a specific domain with a mass pre-learned part-of-speech tagger. The transition probability P'(ti|ti-1, ti-2) and the output probability P'(wi|ti) are extracted from the constructed auto-tagged domain corpus. A vocabulary and a trigram with a large difference between the probability values extracted from the existing learning corpus and the probability values extracted from the auto-tagged domain corpus are extracted. An expert heuristically adjusts the transfer probability and output probability for the extracted vocabulary and Trigram with a large difference between the adaptive domain and the pre-learned domain to fit the domain with linguistic intuition. In general, it is known that it is very difficult to modify the learned statistical values by humans. Accordingly, a method capable of adjusting the probability values based on linguistic intuition can be used.

<A domain adaptation method for the output probability of LHMM>

In order to first extract the vocabularies whose output probability of LHMM is to be adjusted to fit the domain, vocabularies wi for which abs(P(wi|ti)- P'(wi|ti)) exceed an arbitrary threshold Wt are extracted. The output probability P(wi|ti) is the probability that an arbitrary word wi is output or occurs in an arbitrary part-of-speech ti. For example, the output probability P(wi=”write”|ti=VB) is the probability that “write” appears in the case of VB (verb form) part of speech. It is very difficult to predict the range of this probability value with language intuition. However, on the contrary, when the word write has NN (noun), VB (verb form), and VBP (present plural verb) as part-of-speech, you can express yourself with your own linguistic intuition about what kind of part-of-speech distribution “write” will appear in in a specific domain. can In other words, it is possible to predict how much “write” will appear as a verb and to what extent as a noun. Therefore, although it is difficult to express the output probability P(wi|ti) by human linguistic intuition, the probability P(ti|wi) of an arbitrary word becoming a specific part-of-speech can be expressed more accurately with linguistic intuition. In order for the English expert to correct the output probability with such linguistic intuition, in an embodiment of the present invention, the output probability P(wi|ti) is expressed as Equation 2 using the vocabulary part-of-speech probability P(ti|wi) .

In Equation 2, f(wi) denotes the frequency of occurrence of the word wi in the entire corpus, and f(ti) denotes the total frequency of the part-of-speech ti. In this system, the frequencies f(wi) and f(ti) for each word and each part-of-speech are stored, and the lexical part-of-speech probability P(ti|wi) is also stored, and the output probability P(wi|ti) is If necessary, it is calculated according to Equation (2).

To the expert who will adjust the output probability according to the domain, the frequency in the existing learning corpus for the extracted vocabulary wi and vocabulary wi, and the existing output probability P(ti|wi) for all parts of speech that the vocabulary can have are provided, In addition, the frequency and P'(ti|wi) of the vocabulary wi extracted from the auto-tagged domain corpus are also provided. In addition, example sentences tagged with each part-of-speech for the vocabulary wi are extracted from the domain corpus, and up to M sentences are provided. The expert indicates which part-of-speech the word should be tagged in the sentence provided as an example sentence. A random word wi has m parts of speech t1, ... , tm, an adjustable lexical part-of-speech probability P”(tp|wi) as shown in Equation 3 by the expert correctly tagged M sentences per part-of-speech is presented to the expert.

In Equation 3, f(tp|tj) is the number of sentences in which the actual correct answer is part-of-speech tp in the example sentences tagged with the part-of-speech tj, and Nj is the total number of example sentences of the part-of-speech tj. According to Equation 3, if all the example sentences automatically tagged in the domain corpus are correct, there is no need to adjust the previously learned vocabulary part-of-speech probability. For parts-of-speech with many errors, the lexical part-of-speech probability adjustment occurs significantly. Based on the lexical part-of-speech probability P”(tp|wi) presented by Equation 3, the expert calculates the existing lexical part-of-speech probability P(tp|wi) and the lexical part-of-speech probability P'(tp|wi) extracted from the domain corpus as a whole. , and change it to a vocabulary part-of-speech probability suitable for the final domain.

<A domain adaptation method for LHMM transition probability>

The transfer probability of LHMM indicates the grammatical characteristics of the target language. In an embodiment of the present invention, although the basic grammatical characteristics of the target language are largely maintained even when the domains are changed, it is thought that additional necessary grammatical characteristics will appear according to the domains. Therefore, if the existing learning corpus is a corpus with domain balance, it is assumed that the Trigram with high transition probability will be the same in all domains. In an embodiment of the present invention, in the method of reflecting the transition probabilities that vary according to the domain, a newly appeared Trigram or a Trigram that appeared small overall in the target language by showing the grammatical characteristics of the domain, but very high in the domain is extracted and the Trigram for the Trigram Adjust the transfer rate.

In an embodiment of the present invention, in order to extract a trigram for adapting the transition probability of LHMM to a domain, after arranging the trigrams in the order of the highest transition probability P'(ti|ti-1,ti-2) in the domain corpus, In the existing learning corpus, only P(ti|ti-1,ti-2) (including the probability value 0) with a probability value less than or equal to a certain threshold Tt is left. Using the sorted upper Trigrams, it is used in addition to the previously learned transition probabilities. The selected Trigram is used by an expert after checking whether it is grammatically correct through example sentences. For the selected Trigram ti-2,ti-1,ti, the method of adjusting the transition probability suitable for the domain is the transition probability P'(ti|ti-1,ti-2) among the previously learned Trigrams that are not selected first. Find the Trigram tq-2, tq-1,tq with the closest transition probability P'(tq|tq-1,tq-2) to . Then, the domain-adapted transition probability The transition probability P(ti|ti-1,ti-2) is calculated as in Equation 4 and adjusted to a transition probability suitable for the domain.

In order to test the domain adaptation method for the LHMM-based English morpheme part-of-speech tagger, an experiment for adapting the LHMM-based English morpheme part-of-speech tagger to the English textbook domain may be performed. The LHMM tagger is an LHMM-based English morpheme part-of-speech tagger learned at PennTree Bank, and the LHMM domain tagger adjusts the lexical part-of-speech probability for M vocabulary using the raw corpus of each commercially available English teaching material, It adapts the domain with the adjusted transition probability and builds an LHMM-based English morpheme part-of-speech tagger (one month's work by 3 experts). In the LHMM domain tagger, in addition to adjusting the lexical part-of-speech probability for M vocabulary, for example, almost all words in the -ed form in English textbooks are not used as VBD (past tense verb) but VBN (past participle). By using the characteristic, the VBD POS probability value can be reduced to X% and the reduced VBD probability value can be added to the VBN Vocabulary Probability value.

The adjective mapping unit 320 may map and store articles, possessives, and participles combined with a noun among at least one sentence component into an adjective. The adverb mapping unit 330 may store auxiliary verbs that decorate verbs by mapping them into adverbs. The noun mapping unit 340 may store gerunds having the same function as the nouns by mapping them into the nouns (N). The infinitive mapping unit 350 may store the to-infinitive by mapping it to any one of a noun (N), an adjective (Adj), and an adverb (Adv).

The output unit 360, when a sentence to be learned in the user terminal 100 is displayed, 5 parts of speech including a noun (N), a verb (V), an adjective (Adj), an adverb (Adv), and a preposition (Prep) It may include an output unit for outputting displayed as . The user terminal 100, at least one sentence component, which is an article, possessive, and participle combined with a noun included in at least one sentence, as an adjective (Adj), and an auxiliary verb that decorates a verb as an adverb (Adv), a noun A gerund that has the same function as a noun (N) can be output as a noun (N).

<Sentence elements>

, 준동사는

, 관계대명사 및 관계부사는 [ ], 부사구 및 부사절은 ( )의 문장성분기호로 표기하고, 관계대명사 및 관계부사는 [ ]의 [ 또는 ]에

표기를 하여 꾸며주는 문장구성요소로 이어주며, 부사구 및 부사절은 ( )의 ( 또는 )에

표기를 하여 꾸며주는 문장구성요소로 이어주도록 출력되도록 할 수 있다.Verb is one of the constituent elements of a sentence.

, the quasi-verb

, Relative pronouns and adverbs are marked with [ ], adverb phrases and adverb clauses are marked with ( ), and relative pronouns and adverbs are placed in [ or ] in [ ].

It is connected to sentence elements that are marked and decorated, and adverb phrases and adverb clauses are in ( or ) of ( ).

It can be output so that it can be marked and connected to a decorative sentence element.

The translation order notation unit 370, when the sentence output to the user terminal 100 is marked with 5 parts of speech, and is marked with four sentence component symbols that are verbs, quasi-verbs, relative pronouns and relative adverbs, adverb phrases and adverb clauses. , the English direct translation order can be written in numbers at the top of the sentence, and the Korean translation order can be written in numbers at the bottom of the sentence. In this case, the quasi-verb may include a to-infinitive (to+V), a gerund (Ving), and a participle (Ving). In addition, if a preposition (Prep), an adjective (Adj), and a noun (N) are combined in order in a sentence, they are treated as an adverb (Adv) phrase, a modifier phrase, and a zero element, and may not be treated as a sentence component. there is.

Hereinafter, an operation process according to the configuration of the learning service providing server of FIG. 2 will be described in detail with reference to FIGS. 3 and 4 as an example. However, it will be apparent that the embodiment is only one of various embodiments of the present invention and is not limited thereto.

Referring to FIG. 3 , it is a diagram illustrating the basic concept of converting 7 parts of speech into 5 parts of speech in addition to 8 parts of speech. 4D and 4E also provide specific reasons for this.

All grammars result in parts of speech, and usages are those that are helpful for writing (composition) and speaking. The key is to replace all English words with 5 parts-of-speech, and set the translation order by tying the 5 substituted parts-of-speech by unit. The word “unit by unit” is very important here because there are several grammatical terms or other parts of speech in addition to the five parts of speech. Accordingly, a teaching method called 5 parts of 8 speech was developed. For example, the articles a(an) and the are defined as adjectives because they must be combined with nouns, and adjectives and auxiliary verbs are defined as adverbs because they help the verb in the possessive case as well because they must be combined with a noun.

4A to 4C show that <5 parts of speech> and <symbolization of sentence elements> are both applied. According to an embodiment of the present invention, the sentences may be provided in the order [5 parts of speech->Sentence component symbol display->Korean/English translation order]. In this case, the abbreviations used are nouns (n), prepositions (p), verbs (v), adjectives (a), adverbs (ad), relative pronouns (r.p), and conjunctions (c), and lowercase or uppercase letters are not distinguished. 4D and 4E are outlines of 5 parts of speech conversion of 7 parts of speech for simplification of sentence elements, and FIG. 4f is a table summarizing 5 symbols and applied parts of speech and grammatical terms.

As for the method of providing an English learning service using the 5-part of speech of the sentence elements of FIGS. 2 to 4 as described above, the method of providing an English learning service using the 5-part of speech of the sentence elements through FIG. 1 is described above. Since it is the same as the described content or can be easily inferred from the described content, the following description will be omitted.

FIG. 5 is a diagram illustrating a process in which data is transmitted/received between components included in the system for providing an English learning service using the 5-part of speech of the sentence components of FIG. 1 according to an embodiment of the present invention. Hereinafter, an example of a process in which data is transmitted and received between each component will be described with reference to FIG. 5, but the present application is not limited to such an embodiment, and the example shown in FIG. 5 according to various embodiments described above will be described. It is apparent to those skilled in the art that the data transmission/reception process may be changed.

Referring to FIG. 5 , the learning service providing server tags at least one sentence element in at least one sentence with a tag to form a database ( S5100 ).

Then, the learning service providing server maps and stores articles, possessives, and participles combined with a noun among at least one sentence component into an adjective (S5200), and maps and stores an auxiliary verb that modifies a verb into an adverb (S5300) .

In addition, the learning service providing server maps and stores a gerund that functions like a noun to a noun (N) (S5400), and the to-infinitive is mapped to any one of a noun (N), an adjective (Adj) and an adverb (Adv) and save it (S5500).

The learning service providing server, when a sentence to be learned in the user terminal is marked, displays and outputs it as 5 parts of speech including a noun (N), a verb (V), an adjective (Adj), an adverb (Adv), and a preposition (Prep) do (S5600).

The order between the above-described steps ( S5100 to S5600 ) is merely an example and is not limited thereto. That is, the order between the above-described steps ( S5100 to S5600 ) may be mutually changed, and some of these steps may be simultaneously executed or deleted.

As for the matters that have not been described with respect to the method of providing an English learning service using the 5-part-of-speech expression of the sentence elements of FIG. Since it is the same as the described content or can be easily inferred from the described content, the following description will be omitted.

The method of providing an English learning service using five parts-of-speech of sentence elements according to an embodiment described with reference to FIG. 5 is a recording medium including instructions executable by a computer, such as an application or program module executed by a computer. It can also be implemented in a form. Computer-readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, removable and non-removable media. Also, computer-readable media may include all computer storage media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.

The method for providing an English learning service using the five-part speech of sentence elements according to the embodiment of the present invention described above includes an application basically installed in a terminal (which includes a program included in a platform or an operating system basically installed in the terminal) may be executed), and may be executed by an application (ie, a program) installed directly on the master terminal by a user through an application providing server such as an application store server, an application, or a web server related to the corresponding service. In this sense, the method for providing an English learning service using five parts-of-speech of sentence elements according to an embodiment of the present invention described above is implemented as an application (that is, a program) installed by default in a terminal or directly installed by a user, and is implemented in a terminal It may be recorded on a computer-readable recording medium, such as

The above description of the present invention is for illustration, and those of ordinary skill in the art to which the present invention pertains can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. For example, each component described as a single type may be implemented in a dispersed form, and likewise components described as distributed may be implemented in a combined form.

The scope of the present invention is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

Claims (5)

delete delete At least one sentence component, which is an article, possessive, and participle that combines with a noun included in at least one sentence, is used as an adj, an auxiliary verb that modifies a verb into an adverb (Adv), and a gerund that functions like a noun. As a noun (N), the to-infinitive is a noun (N), an adjective (Adj), and a user terminal for outputting by displaying any one of the adverb (Adv); and
A storage unit for tagging at least one sentence element in at least one sentence with a tag to form a database, an adjective mapping unit for mapping articles, possessives, and participles that combine with a noun among the at least one sentence elements into adjectives and storing them; An adverb mapping unit that maps and stores an auxiliary verb that modifies a verb into an adverb, a noun mapping unit that maps and stores a gerund that functions like a noun to a noun (N), and the to-infinitive is a noun (N), an adjective (Adj) and Infinitive mapping unit that maps and stores any one of adverbs (Adv), when a sentence to be learned in the user terminal is marked, noun (N), verb (V), adjective (Adj), adverb (Adv) and preposition (Prep) ), an English learning service providing server including an output unit for displaying and outputting 5 parts of speech including;
including,
Among the sentence elements, the verb is

, the quasi-verb

, Relative pronouns and adverbs are marked with [ ], and adverb phrases and adverb clauses are marked with ( ).
The above relative pronouns and relative adverbs are in [ or ] of [ ].

It is connected with sentence elements that are marked and decorated,
The above adverb phrases and adverb clauses are in ( or ) of ( ).

It is output to connect with the sentence elements that are marked and decorated,
The English learning service providing server,
When the sentence output to the user terminal is marked with 5 parts of speech, and it is marked with four sentence component symbols, which are the verb, quasi-verb, relative pronoun and relative adverb, adverb phrase, and adverb clause, the order of English direct translation is at the top of the sentence. a translation order notation unit for writing in numbers and writing the Korean translation order in numbers at the bottom of the sentence;
further comprising,
The auxiliary verb includes to infinitive (to + V), gerunds (Ving) and participles (Ving),
When a preposition (Prep), an adjective (Adj), and a noun (N) are combined in order in the sentence, the adverb (Adv) phrase and the modifier phrase are treated as zero elements and are not treated as sentence constituents. English learning service providing system using 5 parts of speech of sentence components.
delete delete

Images