Foreign Sentence Structure Learning System Based on Basic Sentence Structure Unit Decomposition

Abstract

The foreign language sentence structure learning system based on basic sentence structure unit decomposition of the present invention is executed by a computing device, the computing device including: at least one processor; and at least one memory for storing multiple instructions executable by the processor. The present invention is characterized by including: a sentence decomposition unit, which, when a user inputs natural language composed of a foreign language, decomposes a compound sentence corresponding to the input natural language into multiple basic sentences; a sentence structure type determination unit, which, if the decomposition of the compound sentence is completed by the sentence decomposition unit, determines the sentence structure type of the multiple basic sentences by identifying one of the morphemes or words included in the decomposed multiple basic sentences; an additional information designation unit, which, when the sentence structure type determination unit completes the determination of the sentence structure type of the multiple basic sentences, designates a portion of the morphemes or words included in the multiple basic sentences as additional information; and an additional information storage unit, which, if the designation of the additional information is completed, matches the additional information with one of the morphemes or words not designated as additional information to store the additional information in a pre-stored basic morpheme dictionary table including basic morpheme categories. Furthermore, various embodiments as described in this specification are possible.

Description

Technical Field

[0001] This invention relates to a foreign language sentence structure learning system based on the decomposition of basic sentence units. Specifically, it relates to the following technique: when a user inputs natural language composed of foreign language, the system identifies whether the compound sentence corresponding to the input natural language is a morpheme-based sentence or a word-based sentence; decomposes the identified compound sentence into multiple basic sentences; determines the sentence type of the basic sentences based on the syllables of the basic sentences; extracts additional information from the basic sentences based on the determined sentence type; and updates the tree information related to the basic morphemes or basic words of the basic sentences based on the extracted additional information, thereby learning the foreign language sentence structure. Background Technology

[0002] In the development of artificial intelligence technology, "language understanding technology" is a representative field that actively utilizes and commercializes artificial intelligence technology. "Language understanding technology" is represented by Natural Language Processing (NLP), which includes the technology of understanding the meaning from the vast texts expressed by people, extracting and classifying the information contained in the text, and then directly generating text.

[0003] However, when using artificial intelligence to analyze the sentence structure of sentences composed of foreign languages, it is difficult to learn by identifying the smallest units of sentences composed of multiple sentence structures from each country and foreign languages. Therefore, companies developing technologies for processing natural language using artificial intelligence are developing various techniques for analyzing and learning the sentence structure of foreign language sentences.

[0004] As an example, Korean Patent Publication 10-2014-0019094 (Method for Reconstructing Lexical Meaning Patterns for Identifying Korean Sentence Structures) discloses the following technology: generating sentences composed of morphemes, syllables, and phrases, and grasping the meaning of sentences through defined lexical meaning patterns.

[0005] However, the aforementioned prior art only discloses techniques for understanding sentence meaning through lexical meaning patterns, but does not disclose techniques for: when inputting natural language composed of foreign language input by the user, identifying whether the compound sentence corresponding to the input natural language is a morpheme-based sentence or a word-based sentence, decomposing the identified compound sentence into multiple basic sentences, determining the sentence type of the basic sentences based on the syllables of the basic sentences, extracting additional information from the basic sentences based on the determined sentence type, and updating tree information related to the basic morphemes or basic words of the basic sentences based on the extracted additional information, thereby learning the foreign language sentence structure. Therefore, a technique that can solve this problem is needed. Summary of the Invention

[0006] Technical problems to be solved

[0007] The purpose of this invention is to provide a foreign language sentence structure learning system based on the decomposition of basic sentence units. When natural language composed of foreign language is input, the system identifies whether the complex sentence corresponding to the input natural language is a morpheme-based sentence or a word-based sentence. The system decomposes the identified complex sentence into multiple basic sentences, determines the sentence type of the basic sentences based on the syllables of the basic sentences, extracts additional information from the basic sentences based on the determined sentence type, and updates the tree information related to the basic morphemes or basic words of the basic sentences based on the extracted additional information. Even if the subject and object are omitted or homonyms are contained in the sentence composed of foreign language, the system collects additional information in a way that understands the meaning of the context appropriate to the previous dialogue.

[0008] Problem-solving methods

[0009] An embodiment of the present invention provides a foreign language sentence structure learning system based on basic sentence structure unit decomposition, executed by a computing device. The computing device includes: at least one processor; and at least one memory for storing multiple instructions executable by the processor. The system is characterized by comprising: a sentence decomposition unit, which, when a user inputs natural language composed of a foreign language, decomposes a compound sentence corresponding to the input natural language into multiple basic sentences; a sentence structure type determination unit, which, if the sentence decomposition unit completes the decomposition of the compound sentence, determines the sentence structure type of the multiple basic sentences by identifying one of the morphemes or words included in the decomposed multiple basic sentences; an additional information designation unit, which, when the sentence structure type determination unit completes the determination of the sentence structure type of the multiple basic sentences, designates a portion of the morphemes or words included in the multiple basic sentences as additional information; and an additional information storage unit, which, if the designation of the additional information is completed, matches the additional information with one of the morphemes or words not designated as additional information to store the additional information in a pre-stored basic morpheme dictionary table containing basic morpheme categories.

[0010] Preferably, the sentence decomposition unit includes: a language recognition unit, which, when the natural language is input, identifies the country of each foreign language corresponding to the natural language to determine whether the language of the identified country is a morpheme-based first language or a word-based second language; a morpheme decomposition unit, which, when the language recognition unit confirms that the input natural language is the first language, identifies multiple morphemes included in the compound sentence corresponding to the input natural language and decomposes the compound sentence according to morphemes; a word decomposition unit, which, when the language recognition unit confirms that the input natural language is the second language, identifies multiple words included in the compound sentence corresponding to the input natural language and decomposes the compound sentence according to words; a unit decomposition unit, which, when the compound sentence is decomposed according to morphemes or words by the morpheme decomposition unit or the word decomposition unit, analyzes the decomposed morphemes or words and decomposes the compound sentence into predicate units based on the analysis results; and a basic sentence generation unit, which, if the compound sentence is decomposed into predicate units by the unit decomposition unit, generates multiple basic sentences from the compound sentence decomposed into predicate units.

[0011] The aforementioned sentence type determination unit may include: a segment confirmation unit, which, when multiple basic sentences based on the first language or multiple basic sentences based on the second language are generated by the aforementioned basic sentence generation unit, confirms multiple segments that respectively constitute the generated multiple basic sentences based on the first language or multiple basic sentences based on the second language; a component classification unit, which, if the confirmation of multiple segments based on the multiple basic sentences based on the first language or multiple basic sentences based on the second language is completed by the aforementioned segment confirmation unit, confirms the part of speech of multiple morphemes or multiple words included in the multiple sentences based on pre-stored part-of-speech classification information, so as to classify the sentence components of each sentence based on the confirmed part of speech; and a sentence type determination unit, which, if the classification of the sentence components of the multiple sentences based on the aforementioned component classification unit is completed, confirms the combination relationship between the classified sentence components, thereby determining the sentence type of the multiple basic sentences based on the first language or multiple basic sentences based on the second language.

[0012] The above sentence types can be sentence forms of multiple basic sentences determined by the combination relationship between each sentence component of multiple syllables or the combination relationship between each sentence component of multiple words. The above sentence types include: Type 1, composed of subject and predicate; Type 2, composed of subject, adverbial and predicate; Type 3, composed of subject, object and predicate; Type 4, composed of subject, complement and predicate; and Type 5, composed of subject, object, complement and predicate.

[0013] The aforementioned additional information designation unit may include: a morpheme extraction unit, which extracts a second morpheme from among the multiple morphemes included in the multiple basic sentences based on the aforementioned first language that determine the aforementioned sentence type, and identifies the second morpheme as additional information related to the aforementioned first morpheme; a first tree information update unit, which, if the extraction of the second morpheme is completed, designates the second morpheme as additional information related to the aforementioned first morpheme and updates the tree information of the aforementioned first morpheme; a word extraction unit, which extracts a second word from among the multiple words included in the multiple basic sentences based on the aforementioned second language that determine the aforementioned sentence type, and identifies the second word as additional information related to the aforementioned first word, and updates the tree information of the aforementioned first word; and a second tree information update unit, which, if the extraction of the second word is completed, designates the second word as additional information related to the aforementioned first word and updates the tree information of the aforementioned first word.

[0014] The tree information mentioned above can be tree structure information belonging to each basic morpheme or basic word according to the sentence type mentioned above. The second morpheme designated as the above additional information is updated to the lower level information of the above basic morpheme, or the second word designated as the additional information is updated to the lower level information of the above basic word.

[0015] When the tree information of the first morpheme is updated by the tree information update unit, the additional information storage unit may update one of the additional information in the multiple basic morpheme categories included in the pre-stored basic morpheme dictionary table based on the updated content, and synchronize the pre-stored basic morpheme dictionary table based on the tree information of the first morpheme. Alternatively, when the tree information of the first word is updated by the tree information update unit, the additional information storage unit may update one of the additional information in the multiple basic word categories included in the pre-stored basic word dictionary table based on the updated content, and synchronize the pre-stored basic word dictionary table based on the tree information of the first word.

[0016] The pre-stored basic morpheme dictionary may include data based on multiple basic morpheme categories, where the basic morphemes serve as the basis for extracting the second morpheme according to the sentence type. To provide the function of searching and extracting additional information included in multiple basic morpheme categories, each basic morpheme category is formed by a graphical structure. The pre-stored basic word dictionary may include data based on multiple basic word categories, where the basic words serve as the basis for extracting the second word according to the sentence type. To provide the function of searching and extracting additional information included in multiple basic word categories, each basic word category is formed by a graphical structure.

[0017] Invention Effects

[0018] The foreign language sentence structure learning system based on basic sentence unit decomposition of the present invention can learn foreign language sentence structures with multiple grammatical systems when inputting natural language composed of foreign language. It identifies whether the compound sentence corresponding to the input natural language is a morpheme-based sentence or a word-based sentence, decomposes the identified compound sentence into multiple basic sentences, determines the sentence type of the basic sentence based on the segments of the basic sentence, extracts additional information from the basic sentence based on the determined sentence type, and updates the tree information related to the basic morphemes or basic words of the basic sentence based on the extracted additional information.

[0019] Furthermore, complex sentences can be broken down into multiple basic sentences, thereby accurately grasping the meaning of each basic sentence through a pre-stored morpheme dictionary or a pre-stored word dictionary.

[0020] Furthermore, by storing additional information in morpheme categories included in a pre-stored morpheme dictionary or in word categories included in a pre-stored word dictionary, the additional information used as learning data can be gradually secured. Attached Figure Description

[0021] Figure 1 This is a block diagram illustrating an embodiment of the present invention of a foreign sentence structure learning system based on the decomposition of basic sentence units.

[0022] Figure 2 This is a block diagram illustrating the sentence decomposition unit of a foreign sentence structure learning system based on basic sentence structure unit decomposition according to an embodiment of the present invention.

[0023] Figure 3 This is a block diagram illustrating the sentence type determination unit of a foreign sentence structure learning system based on basic sentence type unit decomposition according to an embodiment of the present invention.

[0024] Figure 4 This is a block diagram illustrating the combination relationships of sentence components in an external sentence structure learning system based on basic sentence structure unit decomposition according to an embodiment of the present invention.

[0025] Figure 5 This is a block diagram illustrating the additional information specification section of a foreign sentence structure learning system based on basic sentence structure unit decomposition according to an embodiment of the present invention.

[0026] Figure 6 This is a block diagram illustrating the additional information storage unit of a foreign sentence structure learning system based on basic sentence structure unit decomposition according to an embodiment of the present invention.

[0027] Figure 7 This is a diagram illustrating an example of the internal structure of a computing device according to an embodiment of the present invention. Detailed Implementation

[0028] Hereinafter, various embodiments and / or implementations are described with reference to the accompanying drawings. In the following description, for illustrative purposes, several specific details are disclosed to help understand more than one embodiment. However, those skilled in the art will understand that such embodiments can be performed without the specific details. The following description and drawings detail specific exemplary embodiments of more than one embodiment. However, such embodiments are illustrative and may utilize a portion of the principles and methods of various embodiments; the descriptions described include such embodiments and their equivalents.

[0029] The terms "example", "implementation", "illustration", etc. used in this specification may not be construed as any of the described implementation methods, different implementation methods, or implementation methods that are superior to or have advantages over the design.

[0030] Furthermore, terms such as “including” and / or “including…” imply the presence of the corresponding feature and / or structural element, but do not exclude the presence or addition of more than one other feature, structural element, and / or combination thereof.

[0031] Furthermore, terms including ordinal numbers such as "first," "second," etc., can be used to describe multiple structural elements, and the aforementioned structural elements are not limited to the terms mentioned above. The terms are only used to distinguish between two structural elements. For example, without departing from the scope of this invention, a first structural element can be named a second structural element, and similarly, a second structural element can be named a first structural element. The terms "and / or," etc., include a combination of multiple related described items or one of multiple related described items.

[0032] Furthermore, in embodiments of the present invention, unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as understood by one of ordinary skill in the art to which this invention pertains. Terms as defined in commonly used dictionaries have the same meaning as in the relevant technical context and should not be interpreted as having an ideal or excessive meaning unless explicitly defined in embodiments of the present invention.

[0033] Figure 1 This is a block diagram illustrating an embodiment of the present invention of a foreign sentence structure learning system based on the decomposition of basic sentence units.

[0034] Reference Figure 1A foreign sentence structure learning system 100 (hereinafter referred to as the learning system), implemented in a computing device including at least one processor and at least one memory storing multiple instructions that can be executed by the processor, may include a sentence decomposition unit 101, a sentence type determination unit 103, an additional information specification unit 105, and an additional information storage unit 107.

[0035] Prior to the disclosure of the technical features of this invention, "foreign language sentence" was a general term encompassing the languages ​​of multiple countries. Consequently, in the case of foreign language sentences, ambiguity arises due to the different grammatical systems and the smallest units of sentences (e.g., morphemes or words) of each country, and necessary sentence components such as subjects and objects may be added or omitted depending on the context. Therefore, when a user inputs a sentence composed of foreign languages, the learning system 100 of this invention can identify the country of the input sentence, distinguish whether the language of the identified country is a morpheme-based first language or a word-based second language, extract additional information based on the language type of the input sentence, and generate additional tree information by classifying the extracted additional information into basic morphemes or basic words.

[0036] This invention relates to techniques including sentence structure analysis in Natural Language Processing (NLP). More precisely, the aforementioned learning system 100 is a system for learning foreign sentences by identifying patterns in foreign sentences based on tree information, in order to grasp the meaning of sentences and provide appropriate responses based on the identified patterns.

[0037] According to one embodiment, when a user inputs natural language 101a composed of a foreign language, the sentence decomposition unit 101 can decompose the complex sentence corresponding to the input natural language 101a into multiple basic sentences. In this case, the natural language 101a input by the user is based on the structure of text input by the user's voice or by the user, and can be recognized by the sentence decomposition unit 101.

[0038] In this case, the sentence composed of the input natural language 101a can be a sentence composed of a complex sentence with complex meaning. Therefore, the sentence decomposition unit 101 can decompose the complex sentence into basic sentences (e.g., the smallest unit sentence). Thus, there can be multiple basic sentences decomposed and generated from the complex sentence.

[0039] In this case, when natural language is input from an external source, the sentence decomposition unit 101 can identify the country of the sentence corresponding to the input natural language. The sentence decomposition unit 101 can determine whether the language of the identified country is a morpheme-based first language or a word-based second language, thereby identifying the smallest unit of the sentence corresponding to the natural language.

[0040] According to one embodiment, the aforementioned morpheme is the smallest unit used for analyzing the aforementioned sentence, and can be the smallest meaningful linguistic unit. The aforementioned word can be a unit used for analyzing languages ​​of countries where sentences cannot be identified by morpheme units. A detailed explanation of how the sentence decomposition unit 101 decomposes the aforementioned complex sentence into basic sentences can be found in [reference needed]. Figure 2 .

[0041] According to one embodiment, when the sentence decomposition unit 101 completes the decomposition of a compound sentence, the sentence type determination unit 103 can identify the morphemes or words included in the decomposed plurality of basic sentences to determine the sentence type of each basic sentence.

[0042] According to one embodiment, the sentence type determination unit 103 can determine the sentence form included in each of the above-mentioned basic sentences. The sentence type can be the arrangement type of sentence components when forming a sentence, and can signify the form of the sentence. A detailed description of how the sentence type determination unit 103 determines the sentence type can be found in [reference needed]. Figure 3 and Figure 4 .

[0043] According to one embodiment, when the sentence type determination unit 103 completes the determination of the sentence type of each of the basic sentences, the additional information designation unit 105 can designate a portion of the morphemes included in each of the basic sentences as additional information. Furthermore, the additional information designation unit 105 can designate a portion of the words included in each basic sentence as additional information.

[0044] According to one embodiment, the additional information designation unit 105 can determine the sentence type of each of the above-mentioned basic sentences, and can designate segments, morphemes and words composed of the remaining morphemes or words other than the morphemes or words required as the basis for determining the sentence type as additional information.

[0045] More specifically, the aforementioned additional information designation unit 105 can identify the morphemes required to determine the sentence type of each basic sentence as basic morphemes (e.g., first morphemes), and identify morphemes within the sentence other than the aforementioned basic morphemes as second morphemes related to the aforementioned basic morphemes, thereby identifying the aforementioned second morphemes as the aforementioned additional information. Furthermore, the aforementioned additional information designation unit 105 can identify the words required to determine the sentence type of each of the aforementioned basic sentences as basic words (e.g., first words), and identify words within the sentence other than the aforementioned basic words as second words related to the aforementioned basic words, thereby identifying the aforementioned second words as the aforementioned additional information.

[0046] That is, the above-mentioned additional information designation unit 105 can determine that the above-mentioned second morpheme (or second word) is a morpheme that can be derived in relation to the first morpheme (or first word), and determine that it is not a necessary structure required to confirm the sentence type of the basic sentence, thereby designating the above-mentioned second morpheme (or second word) as additional information related to the above-mentioned first morpheme (or first word).

[0047] According to one embodiment, once the additional information is specified, the additional information storage unit 107 matches the additional information with morphemes not specified as additional information (e.g., the first morpheme) to store the additional information in a basic morpheme category included in a pre-stored basic morpheme dictionary table. The pre-stored basic morpheme dictionary table is a data table including basic morpheme categories where additional information is stored for each basic morpheme, and may include multiple basic morpheme categories. The basic morpheme category may be a category that stores additional information matching the basic morpheme. A detailed description related to the additional information storage unit 107 can be found in [reference needed]. Figure 5 and Figure 6 .

[0048] According to one embodiment, when the additional information is specified, the additional information storage unit 107 matches the additional information with words not specified as additional information (e.g., the first word) to store the additional information in a basic word category included in a pre-stored basic word dictionary table. The pre-stored basic word dictionary table is a data table including basic word categories where additional information is stored for each basic word, and may include multiple basic word categories. The basic word category may be a category that stores additional information matching the basic word. A detailed description related to the additional information storage unit 107 can be found in [reference needed]. Figure 5 and Figure 6 .

[0049] Figure 2 This is a block diagram illustrating the sentence decomposition unit of a foreign sentence structure learning system based on basic sentence structure unit decomposition according to an embodiment of the present invention.

[0050] Reference Figure 2 An extra-grammatical sentence structure learning system based on the decomposition of basic sentence units, implemented in a computing device including at least one processor and at least one memory storing a plurality of instructions executable by the processor (e.g., Figure 1 The foreign sentence structure learning system 100 (hereinafter referred to as the learning system) based on the decomposition of basic sentence units may include a sentence decomposition unit 200 (e.g., Figure 1 Sentence decomposition part 101).

[0051] According to one embodiment, when a user inputs a natural language 200a composed of a foreign language, the sentence decomposing unit 200 can decompose the compound sentence 200a corresponding to the input natural language into a plurality of basic sentences. For example, the sentence decomposing unit 200 can receive from a user "I ordered a cup of Americano while waiting for a meeting at a coffee shop", which is a natural language composed of a foreign language.

[0052] According to one embodiment, the sentence decomposing unit 200 may include a language recognition unit 201, a morpheme decomposing unit 203, a word decomposing unit 205, a unit decomposing unit 207, and a basic sentence generating unit 209.

[0053] According to one embodiment, when the natural language is input, the language recognition unit 201 can identify the country of each foreign language corresponding to the natural language, and confirm whether the recognized language of the country is a morpheme-based first language or a word-based second language. For example, Korean and Japanese can be the morpheme-based first language, but are not limited thereto. Also, English can be the word-based second language, but is not limited thereto.

[0054] According to one embodiment, when the input natural language is confirmed as the first language, the morpheme decomposing unit 203 can identify a plurality of morphemes included in the compound sentence corresponding to the input natural language, so as to decompose the compound sentence according to morphemes. For example, the morpheme decomposing unit 203 can decompose the compound sentence "I ordered a cup of Americano while waiting for a meeting at a coffee shop" into morphemes.

[0055] The morpheme decomposing unit 203 can decompose the compound sentence 200a "I ordered a cup of Americano while waiting for a meeting at a coffee shop" into a total of 15 morphemes: "at", "coffee", "shop", "wait", "ing", "meet", "ing", "when", "order", "ed", "a", "cup", "Ameri", "can", "coffee".

[0056] According to one embodiment, when it is confirmed by the language recognition unit 201 that the input natural language is the second language, the word decomposing unit 205 can identify a plurality of words included in the compound sentence corresponding to the input natural language, so as to decompose the compound sentence into words.

[0057] For example, the word decomposition unit 205 described above can decompose the compound sentence "I ordered a cup of Americanocoffee while waiting for the meeting at the coffee shop" according to word decomposition.

[0058] The word decomposition unit 205 above can decompose "I ordered a cup of Americano coffee while waiting for the meeting at the coffee shop" into 16 words: "I" v "ordered" v "a" v "cup" v "of" v "Americano" v "coffee" v "while" v "waiting" v "for" v "the" v "meeting" v "at" v "the" v "coffee" v "shop".

[0059] According to one embodiment, if the unit decomposition unit 207 decomposes the compound sentence according to the morphemes or words by performing the function of the morpheme decomposition unit 203 or the word decomposition unit 205, the decomposed morphemes or words are analyzed and the compound sentence is decomposed into predicate units based on the analysis results.

[0060] According to one embodiment, the unit decomposition unit 207 can distinguish which type of morpheme the decomposed morpheme belongs to. The types of morphemes can be categorized as free morphemes (morphemes that can be used alone (e.g., coffee, meeting, etc., order)), bound morphemes (morphemes that depend on other words for use (e.g., ~hall, ~style, finished)), content morphemes (morphemes with substantial meaning (e.g., the same as free morphemes)), and function morphemes (morphemes that add grammatical relationships or formal meaning (e.g., particles, suffixes, affixes)). The unit decomposition unit 207 can analyze the type of each of the decomposed morphemes. In this case, the unit decomposition unit 207 can distinguish and confirm the type of each of the decomposed morphemes based on pre-stored morpheme information.

[0061] According to one embodiment, the unit decomposition unit 207 may decompose the compound sentence into predicate units based on the above analysis result. The unit decomposition unit 207 may distinguish types of the decomposed morphemes, correspond to the above function words, and identify morphemes having the property of stating a sentence to decompose the compound sentence into predicate units. For example, among the decomposed morphemes "zài"v"kā fēi"v"tīng"v"děng"v"dài"v"huì"v"yì"v"shí"v"diǎn"v"le"v"yī"v"bēi"v"měi"v"shì"v"kā fēi, the unit decomposition unit 207 distinguishes function words, identifies "waiting" and "ordered" which have predicate properties, and decomposes the compound sentence "I ordered a cup of Americano coffee while waiting for the meeting at the coffee shop" into "while waiting for the meeting at the coffee shop" and "ordered a cup of Americano coffee", which are decomposed into predicate units.

[0062] According to one embodiment, the unit decomposition unit 207 may distinguish words included in conjunctions from the decomposed words. The unit decomposition unit 207 may distinguish the type of each decomposed word based on pre-stored word information to identify conjunctions. The unit decomposition unit 207 may decompose the compound sentence into predicate units based on the identified conjunctions.

[0063] According to one embodiment, the unit decomposition unit 207 may decompose the compound sentence into predicate units based on the above analysis result. The unit decomposition unit 207 distinguishes types of the decomposed words to distinguish conjunctions in a sentence. For example, the unit decomposition unit 207 may respectively distinguish the decomposed words "I"v"ordered"v"a"v"cup"v"of"v"Americano"v"coffee"v"while"v"waiting"v"for"v"the"v"meeting"v"at"v"the"v"coffee"v"shop, and identify the conjunction "while". Accordingly, the unit decomposition unit 205 may decompose the compound sentence "I ordered a cup of Americano coffee while waiting for the meeting at the coffee shop" into "I ordered a cup of Americano coffee" and "waiting for the meeting at the coffee shop", which are decomposed into predicate units.

[0064] According to one embodiment, when the compound sentence is decomposed into predicate units by the unit decomposition unit 205, the basic sentence generation unit 209 can generate multiple basic sentences from the compound sentence decomposed into predicate units. The basic sentences can be sentences containing only one meaning, rather than sentences using a complex meaning. For example, the basic sentence generation unit 209 can generate the sentences "while waiting for a meeting at a coffee shop" and "ordered an Americano" from the predicate units decomposed by the unit decomposition unit 205 into two basic sentences: "while waiting for a meeting at a coffee shop" and "ordered an Americano".

[0065] Furthermore, the basic sentence generation unit 209 can generate two basic sentences from “I ordered a cup of Americano coffee” and “waiting for the meeting at the coffeeshop”, which have been decomposed into predicate units by the unit decomposition unit 207.

[0066] Figure 3 This is a block diagram illustrating the sentence type determination unit of a foreign sentence structure learning system based on basic sentence type unit decomposition according to an embodiment of the present invention.

[0067] Reference Figure 3 An extra-grammatical sentence structure learning system based on the decomposition of basic sentence units, implemented in a computing device including at least one processor and at least one memory storing a plurality of instructions executable by the processor (e.g., Figure 1 The foreign sentence structure learning system 100 (hereinafter referred to as the learning system) based on the decomposition of basic sentence structure units may include a sentence type judgment unit 300 (e.g., Figure 1 Sentence decomposition part 103).

[0068] According to one embodiment, if through the sentence decomposition unit (e.g., Figure 1 If the sentence decomposition unit 101) completes the decomposition of the compound sentence, then the sentence type determination unit 300 can identify the morphemes or words included in the multiple basic sentences generated as a result of the decomposition of the compound sentence. In this case, the sentence type determination unit 300 can determine the morphemes or words included in the multiple basic sentences generated as a result of the decomposition of the compound sentence. Figure 1 The sentence decomposition part 101) identifies whether the sentence corresponding to the natural language is a sentence composed of the first language or a sentence composed of the second language, and distinguishes whether it is an identifying morpheme or an identifying word among the multiple basic sentences mentioned above.

[0069] According to one embodiment, the sentence type determination unit 300 can determine the sentence type of each basic sentence based on the above confirmation result.

[0070] According to one embodiment, the sentence type determination unit 300 may include a syllabustrum confirmation unit 301, a component classification unit 303, and a sentence type determination unit 305.

[0071] According to one embodiment, when the basic sentence generation unit (e.g., Figure 2 When the basic sentence generation unit 205 generates multiple basic sentences based on the first language or multiple basic sentences based on the second language, the segment confirmation unit 301 can confirm multiple segments that respectively constitute the generated multiple basic sentences based on the first language or multiple basic sentences based on the second language. A segment is each section that makes up a sentence, is the smallest unit of sentence components, and is a structure that is written as a space-separated unit.

[0072] According to one embodiment, the segment identification unit 301 can perform tokenization processing to identify at least one segment within a plurality of basic sentences based on a first language. In this case, when performing the tokenization processing, the segment identification unit 301 can perform morpheme tokenization instead of word tokenization. The segment identification unit 301 can identify morphemes included in the plurality of basic sentences and distinguish the types of tokenized morphemes. A detailed explanation of distinguishing tokenized morphemes can be found in [reference needed]. Figure 2 .

[0073] According to one embodiment, the segment identification unit 301 can distinguish the types of the morphemes to identify a morpheme composed of a combination of free morphemes and bound morphemes as a tag and designate it as a segment.

[0074] According to one embodiment, the morpheme-marking method described above can be used to identify multiple syllables included in the basic sentences. For example, the syllable identification unit 301 can identify syllables in two basic sentences: "waiting for a meeting in a coffee shop" and "ordered a cup of Americano coffee". The syllable identification unit 301 can perform morpheme marking for "waiting for a meeting in a coffee shop" to identify three syllables 301a: "in a coffee shop" v "waiting" v "meeting". Furthermore, the syllable identification unit 301 can perform morpheme marking for "ordered a cup of Americano coffee" to identify three syllables 301a: "ordered" v "a cup" v "Americano coffee".

[0075] According to one embodiment, when multiple basic sentences based on the second language are generated by the basic sentence generation unit, the segment confirmation unit 301 can confirm at least one segment that constitutes the generated multiple basic sentences based on the second language.

[0076] According to one embodiment, the segment identification unit 301 may perform tokenization processing to identify at least one segment within a plurality of basic sentences based on a second language. In this case, when performing the tokenization processing, the segment identification unit 301 may perform word tokenization rather than morpheme tokenization.

[0077] According to one embodiment, when multiple basic sentences are based on a second language, the segment identification unit 301 can perform tokenization processing to identify at least one segment within the sentence. In this case, when performing the tokenization processing, the segment identification unit 301 can perform word tokenization instead of morpheme tokenization. Word tokenization can be a tokenization method that designates words as the basis for tokenization in the sentence, essentially using spaces as the basis for writing.

[0078] According to one embodiment, the above-mentioned word tokenization method may include a first word tokenization method. This first word tokenization method may be a method that separates words based on spaces within an outer sentence, and distinguishes commas (,) and periods (.) as separate markers. Furthermore, it may be a method that classifies apostrophes as separate markers when an apostrophe (') exists within an outer sentence.

[0079] According to one embodiment, the above-mentioned word tokenization method may include a second word tokenization method. This second word tokenization method may involve changing all letters in a foreign language sentence to lowercase and removing punctuation marks such as periods (.), commas (,), and exclamation marks (!). However, this second word tokenization method does not remove apostrophes ('), but rather preserves them and uses them as tags for classification.

[0080] According to one embodiment, the segment confirmation unit 301 can perform the word tokenization process described above to confirm multiple segments included in the basic sentence. For example, the segment confirmation unit 301 can confirm segments in two basic sentences: "I ordered a cup of Americano coffee" and "waiting for the meeting at the coffee shop". The segment confirmation unit 301 can perform word tokenization on "I ordered a cup of Americano coffee" to confirm seven segments 301a: "I" v "ordered" v "a" v "cup" v "of" v "Americano" v "coffee".

[0081] According to one embodiment, when confirmation of a plurality of language sections of the plurality of basic sentences based on the first language or the plurality of basic sentences based on the second language is completed, the component classification unit 303 may confirm the part of speech of a morpheme or word included in the plurality of language sections based on pre-stored part-of-speech classification information, and classify the sentence component of each language section through the confirmed part of speech. In this case, the pre-stored part-of-speech classification information may be part-of-speech tag information.

[0082] According to one embodiment, the component classification unit 303 may define the part of speech of a morpheme or word through the pre-stored part-of-speech classification information. The pre-stored part-of-speech classification information may include reference morpheme information (including part-of-speech information) for defining the part of speech of each morpheme.

[0083] In more detail, the component classification unit 303 may define the part of speech of a morpheme or word through the pre-stored part-of-speech classification information, and may determine the sentence component of a structure identified as one token among the morphemes or words whose parts of speech are defined. For example, the component classification unit 303 may classify the parts of speech of "会" and "议" based on the pre-stored part-of-speech classification information. The component classification unit 303 may classify "会" as a noun and "议" as a noun. The component classification unit 303 may determine the sentence component of "会议", which is identified as one token, as an object based on the classified parts of speech.

[0084] Conversely, the component classification unit 303 may classify the parts of speech of "a", "cup", "of", "Americano", and "coffee" based on the pre-stored part-of-speech classification information. The component classification unit 303 may classify "a" as an article, classify "cup", "Americano", and "coffee" as nouns, and classify "of" as a preposition. The component classification unit 303 may determine the sentence component of "a cup of Americano coffee" as an object based on the classified parts of speech.

[0085] According to one embodiment, when the classification of the sentence component of each language section is completed by the component classification unit 303, the sentence pattern determination unit 305 may confirm the combination relationship between the classified sentence components to determine the sentence pattern type of the plurality of basic sentences based on the first language or the plurality of basic sentences based on the second language.

[0086] Refer to Figure 4 , Figure 4 shows a table including five sentence pattern types.

[0087] According to one embodiment, the first type is composed of a subject and a predicate; the second type is composed of a subject, an adverbial, and a predicate; the third type is composed of a subject, an object, and a predicate; the fourth type is composed of a subject, a complement, and a predicate; and the fifth type is composed of a subject, an object, a complement, and a predicate. That is, the above sentence types can be sentence forms of multiple basic sentences determined based on the combinational relationships between sentence components of each syllable.

[0088] According to one embodiment, the sentence pattern determination unit 305 can confirm the combination relationship of sentence components of each segment based on the content disclosed in the table above, thereby determining the sentence pattern type of multiple basic sentences based on the first language or multiple basic sentences based on the second language.

[0089] Figure 5 This is a block diagram illustrating the additional information specification section of a foreign sentence structure learning system based on basic sentence structure unit decomposition according to an embodiment of the present invention.

[0090] Reference Figure 5 An external sentence structure learning system based on the decomposition of basic sentence units, embodied in a computing device including at least one processor and at least one memory storing a plurality of instructions executable by the processor (e.g., Figure 1 The foreign sentence structure learning system 100 (hereinafter referred to as the learning system) based on the decomposition of basic sentence structure units may include an additional information specification unit 500 (e.g., Figure 1 Additional information designated section 105).

[0091] According to one embodiment, when determined by the sentence structure (e.g., Figure 3 When the sentence type determination unit 305) completes the determination of the sentence type of each of the above basic sentences, the additional information designation unit 500 may designate a portion of the morphemes or words included in each basic sentence in which the sentence type determination is completed as additional information.

[0092] According to one embodiment, the additional information designation unit 500 may include the morpheme extraction unit 501, the first tree information update unit 503, the word extraction unit 505, and the second tree information update unit 507.

[0093] According to one embodiment, the morpheme extraction unit 501 can extract a second morpheme from multiple morphemes included in multiple basic sentences of the first language based on a determined sentence type. This second morpheme is the remaining morpheme, excluding the first morpheme that corresponds to at least one of the multiple basic morphemes included in a pre-stored basic morpheme dictionary. The basic morpheme is a morpheme stored in the pre-stored basic morpheme dictionary. In the multiple basic sentences, it can be reference information used to extract second morphemes related to the basic morpheme and to match the extracted second morpheme. The first morpheme is a morpheme included in the multiple basic sentences and can be a structure corresponding to the basic morpheme. That is, the basic morpheme can be reference information used to extract additional information.

[0094] For example, the morpheme extraction unit 501 can extract the second morpheme from the multiple basic sentences 501a, namely "waiting for a meeting at a coffee shop" and "ordered an Americano". In this case, "meeting", "waiting", "coffee", and "order" are stored as basic morphemes in the pre-stored basic morpheme dictionary table. That is, in the multiple basic sentences 501a, the morpheme extraction unit 501 identifies "meeting", "waiting", "coffee", and "order" corresponding to the basic morphemes as the first morpheme, and extracts "at a coffee shop", "Americano", and "one cup" as the remaining morphemes other than the first morpheme as the second morpheme.

[0095] According to one embodiment, when the extraction of a second morpheme is completed from each basic sentence, the first tree information update unit 503 can update the tree information 503a of the first morpheme by designating the second morpheme as additional information related to the first morpheme. The tree information 503a is information about the tree structure belonging to each basic morpheme according to sentence type, and can be information that is updated as the second morpheme designated as the additional information is updated to the lower-level information of the basic morpheme.

[0096] According to one embodiment, the learning system generates and updates additional tree information 503a for each basic morpheme according to sentence type. This subordinates the second morpheme, as new additional information, to the lower-level information of the basic morpheme, thereby accumulating the necessary data for sentence structure analysis. That is, the tree information 503a can be information generated or updated based on sentence type and basic morpheme, as new morphemes related to the basic morpheme are added.

[0097] For example, when extracting a second morpheme from multiple basic sentences 501a, the first tree information update unit 503 can assign the extracted second morpheme as lower-level information to the basic morpheme corresponding to the first morpheme. More specifically, the first tree information update unit 503 can assign the second morpheme "at the coffee shop" as lower-level information to the basic morphemes "meeting" and "waiting" to update tree information 503a. Furthermore, the first tree information update unit 503 can assign the second morphemes "Americano" and "a cup" as lower-level information to "coffee" and "order" to update tree information 503a.

[0098] According to another embodiment, when the extraction of a second morpheme is completed from multiple basic sentences 501a, the first tree information update unit 503 can treat the extracted second morpheme as lower-level information subordinate to the basic morpheme corresponding to the first morpheme. In this case, the first tree information update unit 503 can confirm the attributes of the second morpheme.

[0099] According to another embodiment, the aforementioned attribute can distinguish whether the second morpheme is a morpheme indicating a location, an object (animal, person, item, etc.), or a quantity of objects. For example, when the second morpheme "coffee shop" is classified as a subordinate level of the basic morpheme, the first tree information update unit 503 can distinguish the second morpheme "coffee" as an object-representing morpheme and "hall" as a location-representing morpheme based on a pre-stored basic morpheme dictionary. As another example, in the case of Chinese, the suffix or morpheme at the terminal modifies the meaning of the entire word; therefore, the first tree information update unit 503 can determine the attribute of "coffee shop" as a location (where).

[0100] According to one embodiment, the word extraction unit 505 can extract a second word from the words included in a plurality of basic sentences based on a second language that determine the sentence type, as a remaining word other than a first word that corresponds to at least one of a plurality of basic words included in a pre-stored basic word dictionary.

[0101] The aforementioned basic words are words stored in a pre-stored basic word dictionary. They can be reference information used to match second words extracted from multiple basic sentences as morphemes related to the aforementioned basic words. The aforementioned first word, as a word included in multiple basic sentences, can be a structure corresponding to the aforementioned basic words. That is, the aforementioned basic words can serve as reference information for extracting additional information.

[0102] For example, the word extraction unit 505 can extract a second word from the multiple basic sentences 501b, namely "I ordered a cup of Americano coffee" and "waiting for the meeting at the coffee shop". In this case, "a cup of Americano coffee" and "meeting at the coffee shop" are stored as basic words in the pre-stored basic word dictionary. That is, in the multiple basic sentences 501b, the word extraction unit 505 can identify "a cup of Americano coffee" and "meeting at the coffee shop" as the first words corresponding to the basic words, and extract "I ordered", "Americano", and "waiting for the" as the second words, which are the remaining words other than the first words.

[0103] According to one embodiment, when the extraction of the second word is completed, the second tree information update unit 507 can update the tree information of the first word by designating the second word as additional information related to the first word. In this case, the tree information updated by the second tree information update unit 507 is information about the tree structure belonging to each basic word according to the sentence type, and can be information updated as the second word designated as the additional information is updated as the lower-level information of the basic word.

[0104] According to one embodiment, the learning system generates and updates additional tree information for each basic word according to sentence type. This causes the second word, as new supplementary information, to become subordinate hierarchical information of the basic word, thereby accumulating the data needed for sentence structure analysis. That is, the tree information can be generated or updated based on sentence type and basic words, as new words related to the basic words are added.

[0105] For example, when the second word is extracted from multiple basic sentences, the second tree information update unit 507 can use the extracted second word as lower-level information subordinate to the basic word corresponding to the first word. More specifically, the second tree information update unit 507 can use the second word "I ordered" as lower-level information subordinate to "acup of Americano coffee" as a basic word to update the tree information.

[0106] Figure 6 This is a block diagram illustrating the additional information storage unit of a foreign sentence structure learning system based on basic sentence structure unit decomposition according to an embodiment of the present invention.

[0107] Reference Figure 6 An extra-grammatical sentence structure learning system based on the decomposition of basic sentence units, implemented in a computing device including at least one processor and at least one memory storing a plurality of instructions executable by the processor (e.g., Figure 1 The foreign sentence structure learning system 100 based on the decomposition of basic sentence units (hereinafter referred to as the learning system) may include an additional information storage unit 600 (e.g., Figure 1 Additional information storage unit 107).

[0108] According to one embodiment, when the part is specified by additional information (e.g., Figure 5 When the additional information designation unit 500 completes the designation of additional information, the additional information storage unit 600 can match the additional information with morphemes that are not designated as additional information (e.g., the first morpheme) to store the additional information in the basic morpheme categories included in the pre-stored basic morpheme dictionary table.

[0109] According to one embodiment, the pre-stored basic morpheme dictionary table can be a data table including multiple basic morpheme categories that are based on the basic morphemes extracted according to sentence type. The pre-stored basic morpheme dictionary table is formed in a graphical structure for each of the multiple basic morpheme categories to provide search and retrieval functions for additional information included in each of the multiple basic morpheme categories.

[0110] According to one embodiment, when updated via the first tree information unit (e.g., Figure 5 When the first tree information update unit 503 updates the tree information 603 of the first morpheme, the additional information storage unit 600 can update additional information in one of the multiple basic morpheme categories included in the pre-stored basic morpheme dictionary table based on the updated content. In this case, when updating the additional information of the basic morpheme category 605 based on the updated content, the additional information storage unit 600 can update the additional information by distinguishing the hierarchical information of the basic morpheme category 605.

[0111] For example, when "meeting" or "waiting" is designated as a basic morpheme in the tree information 603, the supplementary information storage unit 600, as supplementary information for the basic morpheme, can subordinate "in a coffee shop" as a subordinate level of "waiting for a meeting," which is a second morpheme. This allows the tree information 603 to be updated. When updating the tree information 603, the supplementary information storage unit 600 can update the supplementary information based on the basic morpheme categories 605 of "waiting" and "meeting" based on the updated content. "Waiting" and "meeting" are included in multiple basic morpheme categories, which are included in a morpheme dictionary.

[0112] According to one embodiment, the aforementioned basic word categories may include higher-level items, middle-level items, and lower-level items. These higher-level, middle-level, and lower-level items can be included in the structure of each of the aforementioned basic word categories to facilitate the search and extraction of additional information.

[0113] Furthermore, when updating the additional information in the basic morpheme category 605, the additional information storage unit 600 can confirm the attributes of the additional information and store it in the superior-level item, intermediate-level item, and inferior-level item corresponding to the confirmed attributes. For detailed explanations related to the aforementioned attributes, please refer to... Figure 5 For example, when the attribute of the second morpheme, i.e., the additional information (coffee shop), is identified as "where," the additional information storage unit 600 can store the additional information (coffee shop) at the higher-level item of the higher-level item of the basic morpheme category 605 that has the "where" attribute. In this case, the second morpheme can be stored in the basic morpheme categories based on "waiting" and "meeting" as basic morphemes.

[0114] For example, the aforementioned additional information storage unit 600 can store the additional information (coffee shop) in a higher-level item with the attribute of "location," and can categorize the additional information in at least one intermediate-level item (business location) included in the higher-level item and in lower-level items included in the intermediate-level item (business location). In this case, the additional information storage unit 600 can determine the English word meaning "shop" in the additional information "coffee shop" which has the attribute of "hall" location, so that the additional information "coffee shop" can be stored in the lower-level item included in the intermediate level with the business location attribute.

[0115] That is, the additional information storage unit 600 can update the basic morpheme category 605 included in the pre-stored basic morpheme dictionary table based on the updated content of the tree information 603 of the first morpheme, thereby enabling the pre-stored morpheme dictionary table to be synchronized based on the tree information of the first morpheme.

[0116] According to one embodiment, when the additional information is specified by the additional information specification unit, the additional information storage unit 600 can match the additional information with words (e.g., the first word) that are not specified as additional information to store the additional information in the basic word categories included in the pre-stored basic word dictionary table.

[0117] According to one embodiment, the pre-stored basic word dictionary table can be a data table including multiple basic word categories of basic words based on the basis of extracting the second word according to sentence type. The pre-stored basic word dictionary table is formed in a graphical structure for each of the basic word categories to provide search and extraction functions including additional information for each of the basic word categories.

[0118] According to one embodiment, when updated via the second tree information unit (e.g., Figure 5 When the tree information update unit 507 updates the tree information of the second word, the additional information storage unit 600 can update one of the additional information included in the multiple basic word categories in the pre-stored basic word dictionary table based on the updated content. In this case, when updating the additional information of the basic word category based on the updated content, the additional information storage unit 600 can update the additional information by distinguishing the hierarchical information of the basic word category.

[0119] For example, when the phrase "a cup of Americano coffee" is designated as a word included in the basic words in the aforementioned tree information, the additional information storage unit 600, as additional information for the basic words, can subordinate "I ordered" as a lower-level information of "a cup of Americano coffee." This allows the tree information to be updated. When updating the tree information, the additional information storage unit 600 can update the additional information based on the basic word categories of "Americano coffee" and "a cup of" based on the updated content. "Americano coffee" and "a cup of" are included in multiple basic word categories, which are included in a pre-stored lexicon.

[0120] According to one embodiment, the aforementioned basic word categories may include higher-level items, middle-level items, and lower-level items. These higher-level, middle-level, and lower-level items can be included in the structure of each of the aforementioned basic word categories to facilitate the search and extraction of additional information.

[0121] Furthermore, when updating the additional information in the aforementioned basic word category, the additional information storage unit 600 can confirm the attributes of the additional information and store it in the superior-level item, the middle-level item, and the subordinate-level item corresponding to the confirmed attributes. For detailed explanations related to the aforementioned attributes, please refer to... Figure 5For example, when the attribute of the second word, i.e., the additional information (I ordered), is identified as an action, the additional information storage unit 600 can store the additional information (I ordered) at the higher-level item with the "action" attribute in the higher-level item of the basic word category. In this case, the second word can be stored in the basic word category based on "a cup of Americanocoffee" as the basic word.

[0122] For example, the aforementioned supplementary information storage unit 600 can store the supplementary information (I ordered) in a higher-level item with an "action" attribute, and can categorize the supplementary information in at least one middle-level item (quantity) included in the higher-level item and in lower-level items included in the middle-level item (quantity). That is, the supplementary information storage unit 600 can update the basic word categories included in the pre-stored basic word dictionary table based on the updated content of the tree information of the first word, thereby enabling the pre-stored word dictionary table to be synchronized based on the tree information of the first word.

[0123] Figure 7 This is a diagram illustrating an example of the internal structure of a computing device according to an embodiment of the present invention.

[0124] Figure 7 An example of the internal structure of a computing device according to an embodiment of the present invention is shown. In the following description, references to the above-described [specific details] will be omitted. Figures 1 to 6 The description is redundant and unnecessary.

[0125] like Figure 7 As shown, the computing device 10000 may include at least one processor 11100, memory 11200, peripheral interface 11300, input / output subsystem 11400, circuitry 11500, and communication circuitry 11600. In this case, the computing device 10000 may correspond to user terminal A connected to the haptic interface device or the aforementioned computing device B.

[0126] As an example, memory 11200 may include high-speed random access memory, magnetic tape, static random access memory (SRAM), dynamic random access memory (DRAM), read-only memory (ROM), flash memory, or non-volatile memory. Memory 11200 may contain software modules, instruction sets, or various other data required for the operation of computing device 10000.

[0127] In this case, the processor 11100 controls access to the memory 11200 from the processor 11100 or other components such as the peripheral device interface 11300.

[0128] Peripheral interface 11300 can integrate input and / or output peripherals of computing device 10000 into processor 11100 and memory 11200. Processor 11100 can execute software modules or instruction sets stored in memory 11200 to perform various functions for computing device 10000 and process data.

[0129] The input / output subsystem 11400 can integrate multiple input / output peripheral devices into the peripheral device interface 11300. For example, the input / output subsystem 11400 may include a controller for integrating peripheral devices such as a display, keyboard, mouse, printer, or, if necessary, a touchscreen or sensor into the peripheral device interface 11300. According to another embodiment, input / output peripheral devices may also be integrated into the peripheral device interface 11300 without going through the input / output subsystem 11400.

[0130] Circuit 11500 may supply power to all or part of the components of the terminal. For example, circuit 11500 may include a power management system, a battery or more power sources such as AC, a charging system, a power failure detection circuit, a power converter or inverter, a power status display, or any other components for power generation, management, and distribution.

[0131] The communication circuit 11600 can use at least one external port to communicate with other computing devices.

[0132] Alternatively, as described above, the communication circuit 11600 may include RF circuitry to transmit and receive RF signals known as electromagnetic signals, thereby enabling communication with other computing devices.

[0133] This Figure 7 The embodiment described is merely one example of computing device 10000; computing device 11000 may be omitted. Figure 7 The components shown may also have Figure 7 Additional components, not shown, may have a structure or configuration combining two or more components. For example, a computing device for a communication terminal in a mobile environment, in addition to... Figure 7In addition to the components shown, the communication circuit 11600 may also include a touchscreen or sensors, and may further include circuitry for RF communication using various communication methods (WiFi, 3G, LTE, Bluetooth, NFC, Zigbee, etc.). Components that may be included in the computing device 10000 may be embodied as hardware, software, or a combination of both of the following: integrated circuits specialized for signal processing or applications.

[0134] The method of this invention is embodied in the form of program instructions executable by various computing devices and recorded on a computer-readable medium. In particular, the program of this embodiment can be composed of a PC-based program or a mobile terminal-specific application. Applications to which this invention is applicable can be set on a user terminal via file settings provided by a file allocation system. As an example, the file allocation system can transmit the aforementioned file via a file transfer unit (not shown) upon request from the user terminal.

[0135] The apparatus described above can be embodied as hardware structural elements, software structural elements, and / or combinations of hardware and software structural elements. For example, the apparatus and structural elements described in the embodiments can be embodied using one or more conventional or special-purpose computers, such as controllers, arithmetic logic units (ALUs), digital signal processors, microcomputers, field-programmable gate arrays (FPGAs), programmable logic units (PLUs), microprocessors, or any other device capable of executing or responding to instructions. The processing device can execute an operating system (OS) and one or more software applications running on the aforementioned operating system. Furthermore, in response to the execution of the software, the processing device can also access, store, manipulate, process, and generate data. For ease of understanding, a single processing device is used. Those skilled in the art will recognize that a processing device includes multiple processing elements and / or various types of processing elements. For example, a processing device may include multiple processors or a processor and a controller. It can also be other processing configurations, such as parallel processors.

[0136] Software may include computer programs, code, instructions, or a combination of more than one of these, constituting a processing device in an on-demand manner or issuing instructions to the processing device independently or collectively. Software and / or data are permanently or temporarily embodied in a type of mechanical component, physical device, virtual equipment, computer storage medium, or device for interpretation by the processing device or for providing instructions or data to the processing device. Software may be distributed across computing devices connected via a network and stored or executed using a distributed method. Software and data may be stored on more than one computer-readable recording medium.

[0137] The method of the embodiment is embodied in the form of program instructions executable by various computer units and recorded on a computer-readable medium. The aforementioned computer-readable medium may include program instructions, data files, data structures, etc., individually or in combination. The program instructions recorded on the aforementioned medium may be specifically designed or configured for the embodiment, or may be program instructions known to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical media such as CD-ROMs and DVDs; magneto-optical media such as floppy disks; and hardware devices specifically configured to store and execute program instructions, such as read-only memory, random access memory, and flash memory. Examples of program instructions include machine code formed by a compiler and high-level language code executed by a computer using an interpreter, etc. The aforementioned hardware device operates as one or more software modules to perform the work of the embodiment, and vice versa.

[0138] As described above, the embodiments have been illustrated with specific examples and accompanying drawings. Those skilled in the art can make various modifications and variations based on the above description. For example, even if the described techniques can be performed in a different order than the described methods, and / or the described systems, structures, devices, circuits, and other structural elements can be combined or integrated in a different manner than the described methods, or replaced or substituted by other structural elements or equivalent technical solutions, appropriate effects can still be achieved. Therefore, other examples, other embodiments, and contents equivalent to the scope of protection of the invention also fall within the scope of protection of the invention described below.

Claims

1. A foreign language sentence structure learning system based on the decomposition of basic sentence units, executed by a computing device, the computing device comprising: At least one processor; as well as At least one memory is provided for storing multiple instructions that the processor can execute. The aforementioned foreign sentence structure learning system based on the decomposition of basic sentence units is characterized by the following features: The sentence decomposition unit decomposes a complex sentence corresponding to a foreign language into multiple basic sentences when the user inputs natural language. The sentence type determination unit, if the compound sentence is decomposed by the sentence decomposition unit, determines the sentence type of the multiple basic sentences by identifying one of the morphemes or words included in the multiple basic sentences that have been decomposed. The additional information designation unit, when the sentence type determination unit completes the determination of the sentence type of multiple basic sentences, designates a portion of the morphemes or words included in the multiple basic sentences as additional information; and If the additional information is specified, the additional information storage unit matches the additional information with one of the morphemes or words that are not specified as additional information and stores the additional information in the basic morpheme categories included in the pre-stored basic morpheme dictionary. The above sentence breakdown includes: The language recognition unit, when given the input of the natural language, identifies the country of each foreign language corresponding to the natural language to determine whether the language of the identified country is a morpheme-based first language or a word-based second language; The morpheme decomposition unit, when the language recognition unit confirms that the input natural language is the first language, identifies multiple morphemes included in the compound sentence corresponding to the input natural language, and decomposes the compound sentence according to morphemes. The word decomposition unit, when the language recognition unit confirms that the input natural language is the second language, identifies multiple words included in the compound sentence corresponding to the input natural language, and decomposes the compound sentence according to words. The unit decomposition unit, when the above-mentioned compound sentence is decomposed according to the above-mentioned morphemes or words due to the function of the above-mentioned morpheme decomposition unit or word decomposition unit, analyzes the decomposed morphemes or words and decomposes the above-mentioned compound sentence into predicate units based on the analysis results; and If the basic sentence generation unit decomposes the aforementioned complex sentence into predicate units through the aforementioned unit decomposition unit, then the complex sentence decomposed into the aforementioned predicate units will generate multiple of the aforementioned basic sentences. The above sentence type determination section includes: The segment confirmation unit, when generating multiple basic sentences based on the first language or multiple basic sentences based on the second language through the basic sentence generation unit, confirms the multiple segments that respectively constitute the generated multiple basic sentences based on the first language or multiple basic sentences based on the second language. The component classification unit, if the segment identification unit completes the identification of multiple segments based on multiple basic sentences of the first language or multiple basic sentences of the second language, then, based on pre-stored part-of-speech classification information, identifies the part-of-speech of multiple morphemes or words included in the multiple segments, so as to classify the sentence components of each segment based on the identified part-of-speech; and If the sentence pattern determination section completes the classification of sentence components in multiple syllables through the aforementioned component classification section, it confirms the combinational relationships between the classified sentence components, thereby determining the sentence pattern type based on multiple basic sentences in the first language or multiple basic sentences in the second language. The above-mentioned additional information designation includes: The morpheme extraction unit extracts a second morpheme from the multiple morphemes included in the multiple basic sentences based on the above-mentioned first language that determine the above-mentioned sentence type, and that is, in addition to the first morpheme, which corresponds to at least one of the multiple basic morphemes included in the pre-stored basic morpheme dictionary. If the extraction of the second morpheme is completed, the first tree information update unit will designate the second morpheme as additional information related to the first morpheme and update the tree information of the first morpheme. The word extraction unit extracts a second word from among the words included in multiple basic sentences based on a second language that determine the aforementioned sentence type, as the remaining word excluding the first word corresponding to at least one of the multiple basic words included in a pre-stored basic word dictionary; and If the extraction of the second word is completed, the second tree information update unit will designate the second word as additional information related to the first word and update the tree information of the first word. The tree information described above is the tree structure information belonging to each basic morpheme or basic word according to the above sentence type. The second morpheme designated as the above-mentioned supplementary information is updated to the subordinate level information of the above-mentioned basic morpheme, or the second word designated as supplementary information is updated to the subordinate level information of the above-mentioned basic word.

2. The foreign sentence structure learning system based on basic sentence structure unit decomposition according to claim 1, characterized in that, The above sentence types are sentence forms of multiple basic sentences determined based on the combinational relationships between sentence components of multiple syllables or sentence components of multiple words. The above sentence types include: The first type consists of a subject and a predicate; The second type consists of the subject, adverbial, and predicate mentioned above. The third type consists of the subject, object, and predicate mentioned above. The fourth type consists of the subject, complement, and predicate mentioned above; and The fifth type is composed of the above-mentioned subject, object, complement, and predicate.

3. The foreign sentence structure learning system based on basic sentence unit decomposition according to claim 1, characterized in that, When the tree information of the first morpheme is updated by the tree information update unit, the additional information storage unit updates one of the additional information categories of the pre-stored basic morpheme dictionary table based on the updated content, and synchronizes the pre-stored basic morpheme dictionary table based on the tree information of the first morpheme. Alternatively, when the tree information of the first word is updated by the tree information update unit, the additional information storage unit updates one of the additional information in the multiple basic word categories included in the pre-stored basic word dictionary table based on the updated content, and synchronizes the pre-stored basic word dictionary table based on the tree information of the first word.

4. The foreign sentence structure learning system based on basic sentence structure unit decomposition according to claim 3, characterized in that, The pre-stored basic morpheme dictionary includes data based on multiple basic morpheme categories. These basic morphemes serve as the basis for extracting the second morpheme according to the sentence type. To provide the function of searching and extracting additional information included in multiple basic morpheme categories, each basic morpheme category is formed by a graphical structure. The pre-stored basic word dictionary table includes data based on multiple basic word categories. The basic words are the basis for extracting the second word according to the sentence type. In order to provide the function of searching and extracting additional information included in multiple basic word categories, each basic word category is formed by a graphical structure.

Images