KR102299001B1 - A customized translation method and apparatus for solving ambiguities in sentences and displaying omitted grammatical elements together - Google Patents

Abstract

The present invention provides a description of a method and an apparatus for visually displaying sentence elements and performing translation using them. In a sentence translation method according to an embodiment of the present invention, data of a first sentence of a first language is obtained, wherein the first sentence includes a first morpheme and a second morpheme, the step of expressing the first sentence A first graph is generated, wherein a node of the first graph corresponds to each first morpheme of the first sentence, and an edge of the first graph corresponds to each of the first morphemes of the first sentence. It corresponds to the second morpheme, and when each node and the expression corresponding to each edge are concatenated while traversing the first graph in the first order, the first sentence is completed. and connecting the trunk lines, replacing the expression of the first language corresponding to each node of the first graph and the expression of the first language corresponding to each edge with the expression of the second language, respectively, the first graph obtaining a second sentence of the second language by connecting the expressions of the second language corresponding to each node and each edge while traversing in a second order that is at least partially different from the first order; and outputting the second sentence as a result of translation into two languages.

Description

A user-customized translation method and device for resolving ambiguity in sentences and displaying omitted grammatical elements

The present invention relates to a method and apparatus for visually displaying sentence elements and performing translation using them. More particularly, it relates to a method and apparatus for visually displaying sentence elements constituting a sentence and a relationship between sentence elements and performing clear and accurate translation using the same.

A sentence is composed of various sentence elements such as words, word order, and structure. However, since sentence elements constituting sentences are different for each language, there is a limit to databaseizing various sentences.

In particular, machine translation performed by computers in recent years mainly consists of statistical-based machine translation and artificial neural network machine translation. Since the statistical-based machine translation and artificial neural network machine translation utilize deep learning, the computer learns by itself based on the input learning data. Therefore, a large amount of learning data is essential for high-accuracy machine translation, and a long learning time is required to process a large amount of learning data.

In addition, the computer only displays the translation result to the user, but does not display information about the translation process and the translation method. Since the translation process is not displayed visually, the user cannot confirm that the translation has been performed correctly, and the user does not know which sentence elements of the sentence need to be corrected even if the translation result is not satisfactory. Therefore, it is required to provide a technology for solving this problem.

Korean Patent Publication 2018-0101943 (2018.09.14)

The technical problem to be solved by the present invention is to provide a method and an apparatus for visually identifying the word order of a sentence and a relationship between the sentence elements by displaying the sentence elements as nodes and trunk lines.

Another technical problem to be solved by the present invention is to provide a method and an apparatus for visibly grasping the meaning of a sentence element that is not displayed in a sentence in a sentence.

Another technical problem to be solved by the present invention is to provide a method and an apparatus for providing an accurate translation result by receiving a user's feedback on a translation result displayed by visualizing a sentence element.

The technical problems of the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the following description.

In a sentence translation method according to an embodiment of the present invention for solving the above technical problem, data of a first sentence of a first language is obtained, wherein the first sentence includes a first morpheme and a second morpheme , generate a first graph expressing the first sentence, wherein a node of the first graph corresponds to each first morpheme of the first sentence, and an edge of the first graph is Corresponding to each second morpheme of the first sentence, the first sentence is completed by concatenating the expressions corresponding to each node and each edge while traversing the first graph in the first order connecting the nodes and the trunk lines of the first graph so as to be possible, replacing the expressions of the first language corresponding to each node of the first graph and the expression of the first language corresponding to each edge with the expression of the second language, respectively Step, obtaining a second sentence of the second language by traversing the first graph in a second order that is at least partially different from the first order and connecting each node and the expression of the second language corresponding to each edge and outputting the second sentence as a result of the translation of the first sentence into a second language.

The generating of the first graph according to an embodiment may include generating a first graph including a node corresponding to the first morpheme even when the first morpheme is not included in the first sentence. can

The generating of the first graph according to an embodiment may include selecting a node corresponding to the first morpheme when the first morpheme is any one of a null pronoun and a first morpheme deleted as the same noun phrase. It may include generating a first graph including the.

The generating of the first graph according to an embodiment may include, when the second morpheme is any one of an adjective and a quantifier, generating the first graph including self-edges corresponding to the second morpheme. may include

The generating of the first graph according to an embodiment may include generating the first graph including connecting edges corresponding to the second morpheme when the second morpheme is any one of a verb and a preposition. may include

The generating of the trunk line of the first graph according to the embodiment may include generating the first graph including the trunk line corresponding to the second morpheme even when the second morpheme is not included in the first sentence. may include

The generating of the first graph according to an embodiment may include generating a first graph including an edge corresponding to the second morpheme when the second morpheme is a zero copula. have.

The generating of the first graph according to an embodiment may include changing one or more of a source and a destination of the trunk in response to a predetermined user input for the trunk of the first graph, and the changed trunk It may include the step of generating a first graph including.

The generating of the first graph according to an embodiment may include merging the node with another node in response to a predetermined user input for the node of the first graph, and a first including the merged node. 1 may include generating a graph.

The generating of the first graph according to an embodiment may include dividing the node in response to a predetermined user input for the node of the first graph and a first including the divided one or more nodes. It may include the step of generating a graph.

The generating of the first graph according to an embodiment may include: in response to a predetermined user input for a node of the first graph, dividing the node and replacing the node with two or more new nodes; and generating a first graph including the two or more replaced nodes.

The generating of the first graph according to an embodiment may include: in response to a predetermined user input for a node of the first graph, dividing the node and replacing the node with one or more nodes and one or more edges; and generating a first graph including the replaced one or more nodes and one or more edges.

The generating of the first graph according to an embodiment may include, in response to a predetermined user input for any one of an edge and a node of the first graph, merging the edge and the node to be replaced with a new node ; and generating a first graph including the replaced one or more nodes.

According to an embodiment, the obtaining of the second sentence of the second language may include when at least one of an expression of the second language corresponding to each node or an expression of the second language corresponding to each trunk is an omissible morpheme. , obtaining a second sentence of the second language without concatenating the omissible expressions of the second language.

The step of obtaining the second sentence of the second language according to an embodiment may include any one of a case in which the expression of the second language corresponding to each node is deleted as an intangible pronoun or the same noun phrase, or a first sentence corresponding to the trunk line. When the expression of the two languages is a zero junction, the method may include obtaining a second sentence of the second language without connecting the expressions of the second language that can be omitted.

The step of obtaining the second sentence of the second language according to an embodiment may include traversing the first graph in a second order that is at least partially different from the first order and specified using a previous user's translation record. The method may include obtaining a second sentence of the second language by connecting nodes and expressions of the second language corresponding to each trunk.

The step of obtaining the second sentence of the second language according to an embodiment may include traversing the first graph in a third order that is at least partially different from the second order in response to receiving a predetermined user input. The method may include obtaining a second sentence of the second language by connecting each node and an expression of the second language corresponding to each trunk.

In a sentence translation method according to another embodiment of the present invention for solving the above technical problem, generating first tag information of the first word including the meaning of the first word expressed in a first language, the first In response to obtaining data of a fourth sentence that is a translation target sentence including a word, third tag information of the first word is generated using the data of the fourth sentence, and third tag information of the first word may include the step of including the first tag information of the first word, and the step of replacing the fourth sentence with an expression of a second language using the third tag information of the first word.

According to an embodiment, the generating of the third tag information of the first word from the data of the fourth sentence may include, in response to obtaining the data of the third sentence including the first word, the data of the third sentence generating second tag information of the first word using In response to obtaining the data of the fourth sentence that is the translation target sentence, third tag information of the first word is generated using the data of the fourth sentence, wherein the third tag information of the first word is It may include the step of including the second tag information of the word, and the step of replacing the fourth sentence with the expression of the second language using the third tag information of the first word.

The generating of the first tag information of the first word according to an embodiment may include generating the first tag information that further includes information related to an attribute of the first word that exists only in the first language. may include

The generating of the first tag information of the first word according to an embodiment may include generating the first tag information including gender data of the first word when the first word is not expressed differently according to gender. It may include the step of generating.

The generating of the first tag information of the first word according to an embodiment may include, when an honorific word exists in any one or more of the first language and the second language, class information including the rank of the first word. It may include generating the first tag information including.

A sentence translation apparatus according to another embodiment of the present invention for solving the above technical problem includes a memory into which a sentence translation program is loaded and a processor for executing the sentence translation program loaded in the memory, wherein the sentence translation program comprises: An instruction to obtain data of a first sentence of a language, wherein the first sentence includes a first morpheme and a second morpheme, an instruction, and a first graph expressing the first sentence, wherein the first A node of the first graph corresponds to each first morpheme of the first sentence, and an edge of the first graph corresponds to each second morpheme of the first sentence, and 1 An instruction for connecting nodes and edges of the first graph so that the first sentence is completed when each node and an expression corresponding to each edge are concatenated while traversing the graph in a first order, the first An instruction for replacing the expression of the first language corresponding to each node of the graph and the expression of the first language corresponding to each edge with the expression of the second language, respectively, in which the first graph is in a second order that is at least partially different from the first order An instruction for obtaining a second sentence of the second language by connecting each node and an expression of the second language corresponding to each trunk while traversing in the second order, and the second as a result of the translation of the first sentence into the second language It may include an instruction to output a sentence.

A sentence translation apparatus according to another embodiment of the present invention for solving the above technical problem includes a memory into which a sentence translation program is loaded and a processor for executing the sentence translation program loaded in the memory, wherein the sentence translation program comprises: In response to an instruction for generating first tag information of the first word including the meaning of the first word expressed in one language, and data of a fourth sentence that is a translation target sentence including the first word to generate third tag information of the first word using the data of the fourth sentence, wherein the third tag information of the first word includes first tag information of the first word and an instruction for replacing the fourth sentence with an expression of a second language by using third tag information of the first word.

1 is a flowchart of a method of grasping the meaning of input data using a graph according to an embodiment of the present invention.
2 is a flowchart of a method of translating a sentence by visualizing it in a graph according to an embodiment of the present invention.
FIG. 3 is a flowchart for describing in detail some operations of FIG. 1 .
4 is a flowchart for describing in detail some operations of FIG. 3 .
FIG. 5 is a flowchart for describing in detail some operations of FIG. 4 .
6 to 7 are diagrams for explaining a method of interpreting a plurality of sentences including a specific word in common according to an embodiment of the present invention.
8 is a diagram for explaining a method of translating a sentence by visualizing it in the form of a matrix according to an embodiment of the present invention.
9 to 10 are diagrams for explaining in detail some operations of FIG. 2 .
11 is a diagram for explaining a method of translating a sentence by visualizing it in a graph form composed of nodes and trunk lines according to an embodiment of the present invention.
12 to 14 are diagrams for explaining a method of visualizing and displaying sentences including omitted morphemes according to an embodiment of the present invention.
15 is a diagram for explaining a method of visualizing and translating a sentence so as to display a morpheme corresponding to an order of interpreting a sentence and an edge in a graph according to an embodiment of the present invention.
16 is a diagram for explaining a method of translating a sentence by visualizing it as a figure according to an embodiment of the present invention.
17 is a diagram for explaining a method of visualizing and translating a sentence so that morphemes corresponding to trunk lines are not displayed in a graph according to an embodiment of the present invention.
18 to 19 are diagrams for explaining a method of visualizing and translating a sentence including an honorific word according to an embodiment of the present invention.
20 is a diagram for explaining a method of visualizing and translating a sentence including a word including a tag according to gender according to an embodiment of the present invention.
21 is a diagram for explaining a method of visualizing and translating a plurality of active voice sentences having the same meaning according to an embodiment of the present invention.
22 is a diagram for explaining a method of visualizing and translating a plurality of passive voice sentences having the same meaning according to an embodiment of the present invention.
23 is a diagram for explaining a method of translating each of a plurality of sentences into an image according to an embodiment of the present invention.
FIG. 24 is a flowchart for describing in detail some operations of FIG. 1 .
25 is a diagram for explaining a method of visualizing and translating a visualized passive voice sentence according to an embodiment of the present invention.
26 is a diagram for explaining a method of visualizing and translating a visualized passive voice sentence according to an embodiment of the present invention.
27 is a diagram for explaining a method of performing translation for each node and each trunk using a visualized sentence according to an embodiment of the present invention.
28 is a diagram for explaining a method of performing translation by visualizing a sentence including a word or compound word having an ambiguous meaning according to an embodiment of the present invention.
29 to 30 are diagrams for explaining a method of performing translation by visualizing a sentence using a GUI including a UI for changing a translation result according to an embodiment of the present invention.
31 is a diagram for explaining a method of performing translation by visualizing a sentence using a GUI including a UI for changing a direction of a trunk line according to an embodiment of the present invention.
32 is a diagram for explaining a method of performing translation by visualizing a sentence using a GUI including a UI for automatically correcting a specific morpheme according to an embodiment of the present invention.
33 is a diagram for explaining a method of performing translation by visualizing a sentence using a GUI including a UI for correcting a word order according to an embodiment of the present invention.
34 is a diagram for explaining a method of performing translation by visualizing a sentence by using a GUI including a UI for deleting an edge of a sentence visualized in a graph according to an embodiment of the present invention.
35 to 36 are diagrams for explaining a method of performing translation by visualizing a sentence using a GUI including a UI for changing the shape of a graph in which the sentence is visualized according to an embodiment of the present invention.
37 to 39 are diagrams for explaining a method of performing translation by visualizing a sentence using a GUI including a UI for changing a method of interpreting a word having an ambiguous meaning according to an embodiment of the present invention.
40 is a diagram for explaining a method of performing translation by visualizing a sentence using a GUI including a UI for correcting interpretation of omitted morphemes according to an embodiment of the present invention.
41 is a diagram for explaining a method of visualizing and interpreting a second sentence including a sentence element not specified in the first language according to an embodiment of the present invention.
42 is a block diagram of a text translation system visualized as a graph according to an embodiment of the present invention.
43 to 45 are diagrams for explaining a method of performing translation by visualizing a sentence using a previous translation record according to an embodiment of the present invention.
46 is a hardware block diagram of an apparatus for performing translation using a visualized sentence according to an embodiment of the present invention.
47 is a hardware configuration diagram of an apparatus for performing translation using a visualized sentence according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Advantages and features of the present invention, and a method for achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments published below, but may be implemented in various different forms, only these embodiments make the publication of the present invention complete, and common knowledge in the technical field to which the present invention pertains It is provided to fully inform the possessor of the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout.

Unless otherwise defined, all terms (including technical and scientific terms) used herein may be used with the meaning commonly understood by those of ordinary skill in the art to which the present invention belongs. In addition, terms defined in a commonly used dictionary are not to be interpreted ideally or excessively unless clearly defined in particular. The terminology used herein is for the purpose of describing the embodiments and is not intended to limit the present invention. As used herein, the singular also includes the plural unless specifically stated otherwise in the phrase.

Hereinafter, some embodiments of the present invention will be described with reference to the drawings.

Various types of data (eg, image, photo, text, sound, or raw data) may be obtained as text-type sentences through pre-processing for each. In particular, since a text-type sentence is expressed in a specific language, in order for the user to understand the sentence, it is necessary to translate the plurality of sentences into a language desired by the user.

However, when an unnatural or erroneous translation result is generated in the conventional machine translation in which no human is involved, there is a limitation in that the user has no choice but to directly correct the translation result or use the unnatural sentence as it is.

This is because the existing translation technology does not display any information on the method and process by which a sentence is translated, so that even if an unnatural or erroneous translation result is derived, the user cannot give feedback on the translation process.

When the sentence translation method according to some embodiments of the present invention is used, the user confirms the process from which the translation is derived, and corrects the cause of the mistranslation existing in the translation process, so that a more complete translation can be performed.

Hereinafter, a method of grasping the meaning of a sentence corresponding to the input data using various input data will be described in detail with reference to FIG. 1 .

In step S10, various types of input data may be received. The types of input data may be, for example, images, photos, texts, sounds, and various types of raw data.

In step S20, the inputted data in various forms may be pre-processed and converted into one or more sentences in text form. For example, the user's voice data may be converted into text form using speech to text (STT), and in the case of an image or video, the content of the content may be converted into text form using object recognition technology.

Thereafter, in step S30, the text-form sentence is divided into small sentence element units such as morphemes, and the meaning of each of the one or more sentences and words may be interpreted.

In step S40, a graph corresponding to the sentence may be generated.

In step S50, the content of the sentence may be grasped using the graph corresponding to the sentence. According to some embodiments of the present invention, a sentence corresponding to the graph may be translated into an expression in another language using the graph, and contents corresponding to the graph may be generated to correspond to the contents of sentences corresponding to a plurality of graphs. animations may be created. In addition, by expressing the sentence in the form of a graph, the meaning and configuration of the sentence can be systematically databased, and a machine, not a human, can systematically analyze the meaning of the sentence through the databaseization of the sentence.

Hereinafter, a method of visualizing and translating a sentence according to an embodiment of the present invention will be described with reference to FIG. 2 .

According to step S100, the obtained first sentence may be divided into morpheme units.

The first sentence may be received from an external server or may be an already acquired sentence. Also, the first sentence may be divided into morpheme units through various techniques. A conventional natural language processing model may be used, or a deep learning-based morpheme analyzer may be used. The morpheme segmentation method is not limited thereto, and various morpheme analysis methods available to those of ordinary skill in the art may be used.

For example, when the sentence "I love you" is obtained, it may be divided into morpheme units such as "I/I/You/I love/I love" in step S100.

Also, according to some embodiments of the present invention, a sentence may be classified into a noun morpheme and other morphemes other than a noun. However, it should be noted that this is only an example of a criterion for discriminating the morphemes of a sentence, and the sentence elements according to some embodiments of the present invention are not necessarily divided into a noun morpheme and a non-noun morpheme.

The other morpheme may be further classified into a morpheme that modifies a noun and a morpheme that defines a relationship between nouns. The other morpheme that modifies the noun may be, for example, an adjective or a quantifier, and the morpheme that defines a relationship between nouns may be a verb or a preposition.

Accordingly, in "I love you", "I" and "you" are noun morphemes, "is" and "B" are morphemes that modify a noun among other morphemes, and "I love you" defines the relationship between nouns among other morphemes. It may be a morpheme that

In step S110, a morpheme that is not explicitly displayed in a sentence according to a specific language may also be analyzed. A morpheme that is not explicitly indicated in a sentence may be, for example, a null-pronoun, an equi-NP deletion, or a zero-copula. However, the present invention is not limited thereto, and morphemes that are not explicitly indicated in a sentence include all omissible words that may occur in various languages.

For example, when the sentence "John wants to go home" is obtained, in step S110, "John" omitted from the sentence as a subject in the meaning of "go" is analyzed so that the sentence "John wants (John ) to go home" is can be obtained.

A noun morpheme among the morphemes divided in step S120 may correspond to a node of the graph. For example, in the sentence "I love you", the pronouns "I" and "you" may correspond to nodes in the graph. However, it should be noted that, as described above, as a part-of-speech corresponding to a node, a noun is merely an example, and a part-of-speech other than a noun may correspond to a node.

Among the morphemes divided in step S130, other morphemes other than nouns may correspond to edges of the graph. The trunk line may include a self-loop edge and a connected edge. A description of each trunk line will be described in detail below with reference to FIG. 9 .

For example, in the sentence "John wants to go home", "wants" and "to go " may correspond to the edges of the graph.

In step S140, a graph composed of nodes and trunk lines corresponding to each morpheme of the sentence may be generated. Accordingly, one first graph may be generated corresponding to the obtained first sentence. A method of generating a graph will be described in detail below with reference to the corresponding drawings.

In step S150, the expression of the morpheme of the sentence corresponding to the node and the trunk of the first graph may be replaced with the expression of the language to be translated. According to some embodiments of the present invention, when an expression of a morpheme corresponding to each node and each edge of the graph is replaced from the first language to the second language, it can be translated according to each dictionary meaning, and a previous translation record or It can be translated using the meaning of the previously translated sentence. The translation method will be described in detail below with reference to FIGS. 4 to 7 .

In FIG. S160 , a sentence of the second language corresponding to the first sentence may be output by using the graph corresponding to the morpheme replaced with the expression of the second language in the expression of the first language. The sentence order of the second language may be determined according to the order of traversing the graph. The order of traversing the graph may be an order determined through past translation records, an order obtained through machine learning, or an order specified by a user. The detailed description will be described in detail in the corresponding drawings below.

Step S100 of FIG. 1 according to another embodiment of the present invention will be described in detail with reference to FIG. 3 .

In step S101, the meaning of the obtained morpheme of the first sentence may be grasped. The meaning of the morpheme may be a dictionary meaning, or a unique meaning of each morpheme in a corresponding sentence grasped using a plurality of sentences. A detailed description thereof will be described later with reference to FIGS. 4 to 7 .

In step S102, the first sentence may be visualized in a matrix form according to some embodiments of the present invention. For example, the first sentence may correspond to a product form of a matrix made of noun morphemes and a matrix made of other morphemes other than nouns. In addition, in the case of a matrix composed of other morphemes, a diagonal element corresponds to other morphemes for modifying a noun, and a non-diagonal element corresponds to other morphemes for defining a relationship with a noun. can be However, as described above, the part-of-speech corresponding to each matrix element is not limited thereto. A detailed description thereof will be described later with reference to FIG. 8 .

In step S103, morphemes that are not displayed in the first sentence may also be displayed as elements of the matrix. Through this, each sentence, matrix, and graph can all correspond.

Hereinafter, a method of interpreting the meaning of a sentence for translation will be described in detail with reference to FIG. 4 .

When translating a plurality of sentences, when the meaning of a specific sentence is interpreted according to some embodiments of the present invention, the meaning of the specific sentence may be interpreted using the meaning of a previous sentence.

For example, when the first sentence and the second sentence include morphemes having the same or related meaning, the first sentence and the second sentence are reinterpreted. For example, in interpreting the second sentence, the meaning of the first sentence may be used, and when the second sentence contains the same morpheme as the first sentence, the first sentence is interpreted using the meaning of the second sentence. can be reinterpreted.

In this way, by performing the translation of the final sentence using the meaning of the plurality of sentences, it is possible to clearly understand the meaning of a part where a translation error such as a homonym or an ambivalent expression may occur.

Specifically, it is possible to interpret the meaning of the first sentence in step S200, and store the analysis result of the first sentence in step S201. Thereafter, the meaning of the second sentence may be interpreted in step S202, and the analysis result of the second sentence may be stored in step S203.

The first sentence and the second sentence are analyzed again using the analysis result of the first sentence and the second sentence by integrating the analysis result of the first sentence and the second sentence stored in the steps S201 and S203 in the step S204.

By repeating these steps, the meaning of the final sentence may be interpreted in step S205 for a plurality of sentences.

Hereinafter, a method in which the meaning of a word included in each sentence is interpreted will be described in more detail with reference to FIG. 5 .

First, in order to grasp the meaning of the sentence obtained in step S210, the dictionary meaning of each word is grasped.

After determining the meaning of the first sentence including each word in step S211, the meaning of the corresponding word in the first sentence is stored in step S212. The meaning of the corresponding word in the first sentence may include information about other words that modify the word or related to the word.

For example, in the sentence "My name is Younghee", the dictionary meaning of "I" is "myself", but in the sentence above, the meaning of "My name is Younghee" may be added to the sentence. The meaning of each word added to each sentence may be stored through separate tagging.

After the meaning of each word in the second sentence is interpreted in step S213, tag information including the information obtained in the second sentence in step S214 may be added.

According to some embodiments of the present invention, in the case of a word commonly included in a plurality of sentences, tag information including information obtained from each sentence may be added to each word in the order of obtaining the plurality of sentences.

In step S215 , the meaning of each word may be grasped through all finally obtained sentences, and the meaning of each sentence including each word may also be grasped using each word whose meaning has been grasped.

Hereinafter, a process of grasping the meaning of each word and the meaning of a sentence will be described in detail with reference to FIGS. 6 to 7 .

In FIG. 6 , information of “person’s name” is tagged 220 in the word “withdrawal” before receiving the input of the sentence. After the first sentence 221, “withdrawal is a marine” 221 , is obtained, the word “withdrawal” is further tagged 222 with information “marine”. In addition, when the second sentence 223, "withdrawal boarded the ship" (223) is obtained, the word "withdrawal" is further tagged with the degree of "boarding the ship", and the word "ship" is In addition to the information of "a structure made to float on water", which is the enemy's meaning, more information is tagged "withdrawal boarding".

When the next third sentence 225, the sentence “Chul-su’s ship is broken” is obtained, the tagging of information “the ship is broken” is further added to the word “Chul-su”, and the word “ship” is “destroyed” Additional tagging of the information “Jim” is added. “Pear” included in the second sentence 223 and “ship” included in the third sentence 225 have the same dictionary meaning and refer to the same context, so even if the third sentence 225 is input, the existing Add tag information by inheriting the word "ship" that has added tagging information to . Since "ship" included in the second sentence 223 and the third sentence 225 refers to the same thing, all tagging for "ship" including the tagging information added in the third sentence after input of the third sentence Based on the information, the second sentence may be reinterpreted ( 227 ).

An example of the tagging information added to the words "withdraw" and "ship" as each sentence is acquired in order is as follows.

Withdraw(0) = (person's name) , Withdraw(1) = {Withdraw(0), Marine}, Withdraw(2) = {Withdraw(1), Boarding ship[1](0)}, Withdraw(3) = {evacuation(2), ship[1](1) is destroyed}, ship(0) = (dictionary meaning: a structure made to float on water), ship[1](1) = {ship[1] ](0), evacuation(1) rides}, boat[1](2) = {ship[1](1), wrecked}, boat[1](1) = boat[2](0), Ship[1](2) = Ship[2](1).

Referring to FIG. 7 , unlike FIG. 6 , the word “bae”, which is a homonym, is included in the sentence. In this case, tag information is added for each word.

For example, when the first sentence 231 “Chul-su is a marine”, the second sentence 233 “Chul-su ate a ship” and the third sentence 235 “Chul-su’s stomach hurts” is obtained, “withdrawal Since the word "(230) is inherited with the same meaning in each sentence, the word "withdrawal" tagged with information about the first sentence has information such as "withdrawal(1) = {marine}" (232), The word "withdrawal" tagged with more information for sentence 2 has the same information as "withdraw(2) = {withdraw(1), ate pear[1](0)}" (234), and the third sentence The word "withdrawal" tagged with more information about may have information such as "withdrawal(3) = {withdrawal(2), stomach[2](0) hurts}" (236).

In this case, the tag information for "pear" in the second sentence is pear[1](0) = (dictionary meaning: fruit of a pear tree), pear[1](1) = {pear[1](0), Cheolsu (1) eats} and the tag information for "pear" in the third sentence is pear[2](0) = (dictionary meaning: In the body of a person or animal, the It can be the area between the chest and the buttocks.), the stomach[2](1) = {the stomach[2]{0}, the stomach where the withdrawal (2) hurts}. That is, tag information is added to the "ship" of the second sentence with respect to the ship [1], and the tag information is added to the "ship" of the third sentence with respect to the ship [2].

The translation method according to some embodiments of the present invention enables a clearer and more natural translation of a plurality of sentences through repeated reinterpretation of the plurality of sentences by managing the tags for each word.

A method of visualizing a sentence in the form of a matrix according to an embodiment of the present invention will be described with reference to FIG. 8 .

The sentence "Alice in a dark forest threw a hard rock into a rabbit hole" (250) can be morphologically divided into: "Alice/in/a dark/forest/threw/a hard/rock/into /a/rabbit/hole". When a sentence is converted into a matrix according to some embodiments of the present invention, it may be converted into a matrix composed only of noun morphemes and a matrix composed only of other morphemes other than nouns. However, it should be noted that, as described above, sentences according to some embodiments of the present invention are not necessarily divided into noun morphemes and other morphemes.

Therefore, the above sentence is converted into a matrix (252) having {Alice, forest, rock, rabbit hole} as an element and a matrix (251) having {in, a, dark, threw, a, hard, into, a} as elements can be In addition, the diagonal elements of the matrix composed only of other morphemes may be other morphemes that modify noun morphemes. During the matrix multiplication operation, the element of the morpheme for modifying each noun may be disposed on a diagonal of a position to be multiplied by the element of the corresponding noun morpheme. For example, other morphemes such as a and dark modify forest, and other morphemes a and hard modify rock, so they are placed on the diagonal of the matrix corresponding to positions (2,2) and (3,3), respectively, the second and third It is placed in the position where it is multiplied by the noun.

In addition, according to another embodiment of the present invention, the sentence expressed in the matrix may be converted into a graph. In this case, each element of a matrix composed only of noun morphemes may correspond to a node of the graph, and diagonal elements of a matrix composed only of other morphemes may correspond to self-edges of nodes corresponding to noun morphemes located in the same row. In addition, other morphemes can be connected edges of the graph. For example, if the position of the other morpheme "threw" in the matrix is (1,3), the connecting edge corresponding to "threw" is a node corresponding to the first noun morpheme and a node corresponding to the third noun morpheme of the noun morpheme matrix. It could be something in between.

Hereinafter, with reference to FIG. 9 , a method in which a self-edge and a connected edge are distinguished and generated in a graph will be described in detail.

When other morphemes other than nouns are extracted through the morpheme morphemes of the sentence, it is determined in step S131 whether the other morphemes define the relationship between the noun morphemes.

Other morphemes defining the relationship between noun morphemes may be, for example, any one of a verb, a preposition, a verb phrase, or a prepositional phrase. However, it should be noted that this may appear in different forms for each language.

When the other morpheme does not define the relationship between the nouns in step S132, the trunk corresponding to the other morpheme may be generated as an own trunk of the node corresponding to the noun morpheme. Other morphemes that do not define the relationship between nouns may be quantifiers or adjectives. However, it should be noted that this may appear in different forms for each language.

In the case of other morphemes defining the relationship between nouns in step S133, they may correspond to connecting trunks connecting each noun.

In step S134, the direction of the connecting trunk line may be designated. The tone of the corresponding sentence may be determined according to the direction of the connecting edge. For example, when a node corresponding to a noun morpheme corresponding to a subject is a source node of the trunk and a node corresponding to a noun morpheme corresponding to an object is a destination node of the trunk, the corresponding graph may be an active sentence. Conversely, when the node corresponding to the noun morpheme corresponding to the object is the source node of the trunk, the corresponding graph may be a passive voice sentence.

The type of the sentence may be determined according to the type of the sentence to be translated, a preferred form may be specified using a user's translation record, or a sentence of a specified type using machine learning.

Hereinafter, a method of not displaying morphemes that are not explicitly displayed in a sentence on a graph will be described in detail with reference to FIG. 10 .

If a specific other morpheme can be omitted from the language of the translated sentence in step S141, an edge corresponding to the other morpheme may be omitted in step S142.

Alternatively, when a specific noun morpheme can be omitted from the language of the translated sentence in step S143, a node corresponding to the noun morpheme may be omitted in step S141.

However, note that the precedence relationship between step S141 and step S143 is not limited to the illustrated order, and step S143 may precede step S141, and step S141 and step S143 may be simultaneously performed in a parallel order.

Finally, in step S145, a graph including only the necessary nodes and edges may be displayed.

A method of visualizing and translating a sentence in the form of a graph composed of nodes and edges will be described with reference to FIG. 11 .

When the English sentence "I gave a book to her" is converted into nodes and edges (300a), "I", "book", and "her" are nouns, so they are converted into nodes. And since "a" modifies "book", it is converted into the self-edge of the node corresponding to "book", and "gave" and "to" are between "I" and "book", respectively, and "book" and "Her" Since it defines the relationship between them, it can be transformed into a connecting edge.

{Node('I'), Edge<'I','book'>, Edge<'book','book' It can be converted to a set representation of >,Node('book'), Edge<'book', 'her'> ,Node('her')}.

In addition, according to some embodiments of the present invention, information 301 about the reading order of the graph may be further displayed. Referring to FIG. 11 , it can be seen that the reading order for each node is “I”, “book”, and “her”. Accordingly, in the translation method according to some embodiments of the present invention, the order of extracted sentences may be different according to the reading order.

A case in which a sentence including an omitted morpheme is converted into a graph will be described with reference to FIG. 12 .

When the sentence "I gave her a book" is converted into a graph, the morpheme defining the relationship between the noun morpheme "her" and the noun morpheme "book" is omitted and not explicitly indicated in the sentence. Accordingly, according to some embodiments of the present invention, omitted morphemes of a sentence may also be displayed 302 on the graph. In FIG. 12 , trunk lines corresponding to omitted morphemes of a sentence are indicated by dotted lines, but the method of displaying the omitted morpheme is not limited thereto.

By displaying the omitted morpheme in a sentence of a specific language on the graph, it is possible to accurately translate a case in which the omitted morpheme is explicitly displayed in a later translated language.

Also, referring to FIG. 12 , unlike FIG. 11 , sentences having the same meaning as in FIG. 11 but in a different word order may be extracted by analyzing the graph in the order of nodes corresponding to “I”, “her”, and “book”.

A method of converting a sentence including an omitted noun morpheme into a graph will be described with reference to FIG. 13 .

For example, if the sentence "John wants (John) to go home" is converted into a graph, the sentence is {Node('John'), Edge<'John','(John)'> , (Node('(John)')), Edge<'(John)','home'>, Node('home')}.

At this time, even if the subject "John" is omitted in the meaning of "to go" in the sentence 310a of "John wants (John) to go home" and is not explicitly displayed in the sentence, the meaning of the omission from the graph A node 311 corresponding to the subject "John" may be displayed.

Also, according to another embodiment of the present invention, omitted morphemes may not be displayed in a graph. 14 , when the subject “John” is not explicitly indicated in the sentence in the meaning of “to go”, both the morpheme omitted in the language before translation and the language after translation are not explicitly included in the sentence. In this case, the node corresponding to the omitted “John” may not be generated for the sake of simplification of the graph.

Hereinafter, a method of analyzing a graph composed of a plurality of nodes and edges will be described with reference to FIG. 15 .

According to another embodiment of the present invention, when analyzing a graph composed of a plurality of nodes and edges, a separate indication may be added to the subject of a sentence. For example, in the case of the graph of FIG. 15 , a mark indicating that “Alice” is the entire subject may be added. In addition, indications 321 , 322 , and 323 for the analysis order of the graph may be further added. Therefore, when the user interprets the graph as a sentence, the graph can be interpreted as a one-dimensional sentence 320a by using the graph interpretation order display starting with the subject displayed on the graph.

In addition, when the sentence "Alice in a dark forest threw a hard rock into a rabbit hole." shown in FIG. 15 is translated into another language, the interpretation order (321, 322, 323) of the displayed graph may be changed. For example, if the above sentence is translated into Korean, it can be translated as "Alice in the dark forest threw a hard stone into the rabbit hole", and "Alice threw a hard stone into the rabbit hole in the dark forest" could be translated into Korean. have. That is, even when the same graph is translated, a plurality of sentences having different orders may be generated. In this case, separate marks 321 , 322 , and 323 corresponding to the interpretation order of each sentence may be added.

Also, referring to FIG. 16 , for a user who does not know a language after or before translation, all words corresponding to the graph may be displayed as images such as pictures or photos. Through this, even a user who does not know a specific language can intuitively grasp the meaning, structure, and relationship of each morphology of the sentence.

Referring to FIG. 17 , according to another embodiment of the present invention, a sentence may be visualized so that an edge of a graph is not displayed.

Since the trunk line of the graph means a relationship between nouns or a morpheme that modifies the noun, the direction of the trunk line may vary depending on the tone (eg, passive voice and active voice) of a sentence to be described later. Accordingly, in some embodiments, when understanding the meaning of a sentence is important, the edges of the graph may not be displayed. Through this, the user can easily obtain only the information of his or her interest in the interpretation and translation process of the sentence.

Hereinafter, with reference to FIG. 18 , a method of displaying a sentence in a graph in the case of a language in which an honorific word exists according to the class between the speaker and the subject of the sentence will be described.

For example, in the case of Korean, strict use of honorifics exists, but in the case of English, not all words corresponding to each of the honorifics of Korean exist. Therefore, when a sentence having an honorific word is translated into English and then the corresponding sentence is translated into Korean again, there is a case in which the honorific expression itself disappears or is translated awkwardly. To prevent this, according to the present embodiment, when a graph corresponding to a sentence is generated, information about a speaker of the sentence and a hierarchy between a subject and an object of the sentence may be further generated.

For example, in the case of the sentence "Deputy Director Lee delivered coffee to Deputy Director Kim", the speaker of the sentence has a higher rank than Deputy Director Lee and a lower rank than Deputy Director Kim. Accordingly, when the sentence is converted into the graph of FIG. 18, more information on the hierarchy diagram 340a on the relationship between the speaker, Deputy Director Kim, and the assistant manager is required.

In particular, in the case of the sentence "Deputy Director Lee delivered coffee to Deputy Director Kim" in Fig. 19, which has the same meaning as the sentence in Fig. 18, the speaker's rank is lower than that of Deputy Director Lee and Deputy Director Kim (340b), unlike the sentence in Fig. 18 ). In this case, through this embodiment, information about the sentence "Deputy Manager Lee delivered coffee to Deputy Manager Kim" in FIG. 19 and the sentence "Deputy Manager Lee delivered coffee to Deputy Manager Kim" in FIG. 18 are generated differently. More accurate text translation is possible.

Hereinafter, a case in which tag information according to gender is further generated with respect to a designated morpheme of a sentence according to another embodiment of the present invention will be described in detail with reference to FIG. 20 .

For example, in the case of Korean, the word "great" has the same form when it modifies a noun morpheme meaning female and a noun morpheme meaning male. However, in French, “Un grand” meaning “great” is used to modify a noun morpheme meaning male, and “Un grande” is used to modify a noun morpheme meaning female. In this case, according to the present embodiment, when French is translated into Korean, a tag including information on gender may be added to each word in addition to the translation result of “great”. For example, when the translation result is "Great Emperor", it may be displayed as "Great (Male) Emperor" by adding more tag information.

In this case, in a language in which the shape of each word is displayed differently according to gender, a tag including information on the gender may not be added to each word. For example, if the English word "Emperor" means a male emperor and "Empress" means a female emperor, it is not necessary to add a separate tag including gender information to the word.

Accordingly, when a word having a different shape according to gender is translated, a tag including information on gender is generated for each of the words, so that a more accurate result can be extracted in the subsequent translation process.

A case in which a plurality of sentences correspond to one graph will be described with reference to FIG. 21 . When a sentence is converted into a graph according to some embodiments of the present invention, as shown in FIG. 21 , a plurality of sentences having the same meaning or different word order may correspond to one graph. Therefore, in the case of performing the translation, it is possible to efficiently utilize the computing resources because the translation is performed on one corresponding graph without the need to perform the translation on each of the plurality of sentences.

Also, when the word order of the sentences extracted as a result of the translation is different from the word order intended by the user, a plurality of sentences of various word orders having the same meaning may be further extracted. As for the order of the plurality of extracted sentences, a statistical method using a user's previous translation record and multiple users' translation records may be used, and machine learning using a vast translation record may be used.

Also, even in sentences with the same meaning, the shape of the graph corresponding to each sentence may be different depending on whether the voice of the sentence is in the passive voice or the active voice. Referring to FIG. 22 , even in a sentence having the same meaning as in FIG. 21 , in the case of a graph corresponding to the sentence of the passive frame, the direction of the trunk line may be different from that of the graph of FIG. 21 . Therefore, the user can modify the tone of the extracted sentence by modifying the direction of the edge of the graph.

Through this, even when the user does not know any one of the languages before or after translation, the user can intuitively grasp the meaning and tone of the extracted sentences.

Referring to FIG. 23 , according to some embodiments of the present invention, graphs 351 , 352 , 353 corresponding to a plurality of sentences may be imaged ( 354 , 355 , 356 ).

An image corresponding to each graph may be generated, and an animation may be generated using the graph corresponding to a plurality of sentences including the same word. For example, when a plurality of graphs 351 , 352 , and 353 corresponding to a plurality of sentences including the word “John” are generated, using information related to a node corresponding to “John” in each graph, the A plurality of images or animations for "John" may be created. As described above, when the meaning of a sentence is interpreted according to some embodiments of the present invention, the meaning of a word included in a specific sentence uses not only the corresponding sentence but also the context of a plurality of sentences existing before and after the corresponding sentence, A plurality of sentences including a common word may be replaced with an image or animation having one content.

Hereinafter, a process in which a sentence of the first language converted into a graph is translated into a sentence of a second language according to another embodiment of the present invention will be described with reference to FIG. 24 .

In step S151, a morpheme of a sentence of the first language corresponding to each of the nodes and trunk lines of the graph may be replaced with an expression of the second language to be translated.

When the user's correction request feedback for the second sentence of the second language extracted from the graph in step S152 is present, a third sentence different from the second language may be further extracted in step S154. The third sentence may have different word order, sentence tone, word translation, morpheme modification relationship, and the like, depending on the type of user feedback.

The translation result for the second sentence extracted in step S153 or the third sentence extracted in step S155 may be stored. The information on the extracted sentence and information on the input of the user's feedback may be used to reflect it in the subsequent translation process. A detailed description will be given later in the corresponding drawings.

Hereinafter, a process in which the first sentence 400a, which is the expression of the first language, is translated into the second sentence 400b, which is the expression of the second language, will be discussed in detail with reference to FIGS. 25 to 26 .

When "A cat is touched by me", which is the first sentence 400a expressed in English, is to be translated into Korean, a graph 401a corresponding to the first sentence may be generated first. Each node and edge of the generated graph may correspond to each word of the first sentence.

Thereafter, words corresponding to nodes and trunk lines of the graph may be translated into Korean. In this case, as described above, information on gender and information on rank may be further stored according to the characteristics of each language.

In order to extract the translated sentence 400b, the graph replaced with Korean may be interpreted in a specified word order. The designated word order may be a graph interpretation order corresponding to the word order of a sentence according to various criteria such as a word order preferred by a specific user or a word order preferred by most users using a previous translation record.

For example, the English sentence "A cat is touched by me" is a passive voice, so it can be translated as "a cat is touched by me" instead of "I touch the cat" when translated into Korean.

On the other hand, in the case of the active English sentence "I touch a cat" as shown in FIG. 25, unlike in FIG. 24, it may be translated as "I touch the cat".

Hereinafter, a method for translating a word in a sentence corresponding to each node and an edge of the graph will be described in detail with reference to 27 .

In the case of the graph 420 composed of two nodes and one trunk connecting the nodes, the expression of the first language corresponding to each node and the trunk is replaced with the second language. For example, if the Korean sentence "I love you" is translated into an English sentence, "I/-/Love/ It can be translated as (-active)/You/-/".

Through this, unlike the case where the entire sentence is translated, the translation is performed for each morpheme, and information on the relationship between each morpheme is separately stored, thereby enabling more systematic translation performance and management of the translation record.

Hereinafter, a method of translating a sentence including a phrase having an ambiguous meaning and a compound word will be described with reference to FIG. 28 .

A phrase with an ambiguous meaning occurs when the mathematical relationship between each word is unclear or when the word to define the relationship is unclear. According to the present embodiment, since the mathematical relationship between each word is expressed as a graph, the meaning of the sentence is not interpreted ambiguously.

For example, if you have the sentence "old man and woman", it can be ambiguous whether the word 'old' modifies 'male' or 'male and female'. Therefore, when matching the sentence to a graph, if “old” modifies “male”, a self-edge corresponding to “old” is generated to point to a node corresponding to “male”, and “old” is “male” and female", the self-edge corresponding to "old" is created to point to the node corresponding to "male and female".

In addition, compound words can be created and removed by merging and separating edges of graphs corresponding to sentences with nodes. For example, if it is natural for "old man" to be interpreted as "grandfather" instead of "old man", the own edge corresponding to "old" and the node corresponding to "man" are merged to create a node called "grandfather" can be replaced with Likewise, it goes without saying that "grandfather" can be divided into edges and nodes corresponding to "old" and "man". Such a compound word can use information stored in an existing dictionary DB.

Hereinafter, a method of performing translation using a GUI including a UI for changing a translation result according to another embodiment of the present invention will be described with reference to FIGS. 29 to 41 .

Referring to FIG. 29 , when a user wants to extract sentences having different word orders from the same graph, a new translation result may be obtained through a predetermined user input 501 .

For example, if the sentence "John can hit the ball with a bat." is translated into Korean and the sentence "John can hit the ball with a bat" is output (500a), the user has a different word order. A predetermined UI 501 may be selected to obtain a sentence. Through this, the user can obtain the sentence "John can hit the ball with the bat" (500b), and all of the information about the translation record and the user's feedback are recorded. Also, when there is a translation request from the same user for a similar sentence thereafter, the sentence in the word order selected by the user may be displayed preferentially.

In addition, when it is desired to translate the first sentence of the first language into the second sentence of the second language, the graph may display morphemes corresponding to the first language and morphemes corresponding to the second language, Both the first language and the second language may be displayed. As described above, by displaying the user's familiar language and the unfamiliar language in correspondence with the graph, the user can check whether the translation of each morpheme is correct for the corresponding sentence. In addition, by modifying the components of the graph corresponding to each morpheme, the wrong part of the translation can be easily identified and corrected.

Referring to FIG. 30 , when the user wants to correct the interpretation of a specific word in the translated sentence. By making a predetermined input for the specific word, the translation for the corresponding word may be corrected.

For example, when the Korean sentence “I touch the cat” is translated into English, the sentence “I touch the cat” may be output (510a). If the user wants to correct or delete the word "the", the interpretation result of the corresponding word is changed by inputting a predetermined input 511 to "the". When the sentence "I touch a cat" different from the sentence is output (510b) and the user selects it as the final translation result (512), in the case where it is ambiguous to express the singular for a specific word in the English sentence in the next translation, "a "'s priority will be higher than that of "the".

Referring to FIG. 31 , the corresponding component and the translation result corresponding to the corresponding component may be modified in response to an input of a user selecting 521 a component of the graph. For example, when there is a user input for selecting an edge of the graph, the direction of the corresponding edge may be modified.

When the direction of the trunk line is modified, the relation between the noun morphemes or the mathematical relation in which other morphemes modify the noun morpheme may be modified. When the direction of the trunk line is changed, the direction of the trunk line corresponding to the other morpheme defining the relationship between the noun morphemes is changed, and the tone of the sentence may change, and the formula relation may change. Also, it is possible to change the structure of a sentence by modifying the origin or destination of the trunk line.

For example, if you want to interpret the sentence "I touch the cat" in English (520a), you can change the direction of the trunk by making a designated input 521 for the trunk corresponding to the morpheme "touch" to obtain a passive voice sentence. There is (524). As the direction of the trunk line is changed, the English sentence obtained as a result of the translation is also reacquired from "I touch the cat" to "The cat is touched by me." (520b), and the Korean sentence may also be changed to correspond thereto. By changing the Korean sentences, the user can check whether the desired translation result is correct. In addition, the expressions 522 and 523 of other morphemes to be corrected by changing the tone of the sentence may be automatically modified.

Also, according to another embodiment, other morphemes corresponding to the own trunk lines may be displayed 522 and 523 without directionality. In the case of a self-edge, since both the origin and destination are the same, even if the direction is not indicated on the corresponding edge, different results are not derived in translation. Accordingly, in order to improve the user's readability of the graph, the dotted line or a specific figure may be matched with other morphemes corresponding to its own trunk lines.

Referring to FIG. 32 , the omitted morpheme may be explicitly displayed in response to a user input 531 for the omitted morpheme.

For example, if you want to translate the English sentence "I touch the cat" into Korean, if a sentence interpreted as "I touch the cat" is obtained (530a), the Korean word corresponding to "the" can be omitted and thus automatically omitted (531) may have been done. In this case, according to some embodiments of the present invention, even the omitted morpheme is displayed in the graph ( 532 ), so that the user can input a designated user input to explicitly display the omitted morpheme in the sentence after translation. As a result, the Korean word "he" corresponding to "the" is explicitly displayed in the translation result to obtain the sentence "I touch the cat" (530b).

Referring to FIG. 33 , in response to a user input 541 for changing the word order 542 of the sentence according to another embodiment of the present invention, the word order of the translated sentence may be changed 543 .

For example, if a Korean translation result of "I touch the cat" is output for the English sentence "I touch the cat", the user wants to interpret to obtain a sentence of word order such as "I touch the cat" Nodes can be selected as However, it should be noted that the user input for the method of changing the word order is not limited thereto.

Also, referring to FIG. 34 , when it is not desired to display an edge in the graph, the edge may not be displayed according to an input 551 of a designated user capable of deleting the corresponding edge. The deleted edges may be edges selected by a user or all edges existing in the graph.

The user may modify the translation result of the sentence more specifically by deleting some or all of the edges.

Also, referring to FIGS. 35 to 36 , a graph display method may be changed according to a user's selection.

In FIG. 35 , when the user wants to display the edges of the graph ( 561 ), or when the revision of the edges of the graph is frequent in consideration of the user's past translation revision record, the graph is in the form of displaying all the edges and nodes ( 560a ) can be displayed.

On the other hand, in FIG. 36 , when the user does not want to display the edges of the graph ( 562 ), or when there is no correction for the edges of the graph in consideration of the user's past translation revision record, the graph is a form in which the edges are not displayed ( 560b ) can

However, the present invention is not limited thereto, and depending on the user's translation record and user's settings, some or all of the graph may be displayed, such as in the form of a graph without the display of morphemes corresponding to the main line, or as a phrasing graph with an ambivalent meaning. may be

Hereinafter, a method of performing translation using a GUI including a UI for changing an interpretation method of a word having an ambiguous meaning will be described with reference to FIGS. 37 to 39 .

In the case of a word or phrase having an ambiguous meaning, the user's correction feedback may be input so that the translation is performed in a direction intended by the user.

For example, when the translation of the sentence "old man and woman" is performed when referring to FIG. 37 , a self-edge corresponding to "old" may point to a node corresponding to "man" (570a). In this case, the translated Korean sentence is interpreted as “an old man and an (not old, ordinary) woman”. If it is desired to obtain a translation result as “old man and old woman” for the analysis result, the user may modify the graph to obtain the desired translation result.

Even if the user does not know Korean, "old" in the graph does not indicate "woman", so the user can know that the translated Korean sentence is incorrect.

Accordingly, as shown in FIG. 38 , a drag-and-drop input 571 may be performed so that the node corresponding to “woman” is connected to the trunk line corresponding to “Old”. However, drag and drop is just an example of user input, and the user input method is not limited thereto, and may be various methods.

Also, as shown in FIG. 38 , the translation result may be corrected ( 570b ) through an input 572 for an edge corresponding to “Old”.

When the graph shown in FIG. 39 is generated through this (570c), since the edge corresponding to "Old" points to the node 573 corresponding to "Man and Woman", the translation result of the Korean sentence is "Old man and old woman" " can be modified to

Also, according to another embodiment of the present invention, the edges of the graph may be merged or divided. For example, when there are a plurality of trunks starting from a designated node and arriving at another designated node, the plurality of trunks may be merged into one trunk. When trunk lines corresponding to a plurality of morphemes are merged into one, the merged trunk line may be divided into a plurality of trunk lines again in response to a user input to divide the trunk line. Also, it goes without saying that the origin and destination of each trunk may be changed in response to a user input for each of the plurality of trunks.

Also, with reference to FIG. 40 , according to another embodiment of the present invention, a user can easily correct an erroneous interpretation of a word with an ambiguous meaning.

For example, in the case of the English word "sheep", since the singular and the plural are the same, a translation result different from the user's intention may be generated ( 580 ) in the translation process. According to some embodiments of the present invention, since information on whether a noun morpheme is singular or plural may be further displayed on the graph, the translation result may be modified according to a user's preset input 581 for this.

When the English sentence "I touch sheep" is translated into a Korean sentence, it can be translated as "I touch the sheep". In contrast, if the user's intended sheep was singular, the user uses an input 581 for "(plural)" meaning a plurality of sheep or a corresponding Korean "s" 583 as shown in FIG. A corrected translation result can be obtained. Accordingly, in FIG. 41 , “(plural)” is modified to “a”, so that “sheep” is a singular form, and a Korean translation sentence such as “I touch a sheep” can be obtained.

Hereinafter, a sentence interpretation system according to an embodiment of the present invention will be described in detail with reference to FIG. 42 .

Therefore, according to some embodiments of the present invention, the sentence analysis apparatuses 100a, 100b, 100c may interpret a text in a text form obtained by preprocessing 60 various data 10, 20, 30, 40, 50. . In addition, according to some embodiments of the present invention, the obtained sentence may be translated into an expression in a language desired by the user. In addition, when the user wants to correct the translated sentence, the user can easily and clearly edit the sentence by inputting the graph.

In addition, each of the sentence interpretation apparatuses 100a , 100b , and 100c may separately manage each translation record to further store information such as a translation pattern for each user and a context before and after the translation target sentence. Through this, the user may obtain a customized sentence interpretation result or a translation result.

Specifically, a method of performing translation using local storage for each translation task will be described with reference to FIGS. 43 to 45 .

Referring to FIG. 43 , according to an embodiment, different local storages may be used to perform translation for each user. For example, a local storage 431 that stores information about a translation record previously used by the user 1 430 or a sentence previously translated in the same translation process, and a local storage 433 that stores the translation record of the user 2 433 . Storage 434 may be different.

Also, according to another embodiment of the present invention, the local storages 431 and 434 may include only information about previous translation records in the same task. For example, when translating a large amount of text consisting of a plurality of paragraphs, the translation may be performed using information stored in the local storage only in the corresponding task with reference to the translation record of the previous sentence.

The information stored in the local storage includes word order information, preference information for merging or splitting nodes or edges, the number of sentence elements or words or morphemes, the number of preferred syllables, the similarity of the international phonetic symbols, Among information about information, gender information, preference for written or spoken language, whether foreign words are used, preference for sentence tone (eg, passive and active voice), position of punctuation marks or information on sentence tone such as closing sentences One or more may be included.

By managing the translation records differently for each user, it is possible to create a translation record desired by the user according to the preferred translation method and the preferred word order of the sentence. Also, according to another embodiment, a main storage 432 shared by all users may exist. The main storage 432 may store, for example, all information about dictionary meanings of a plurality of languages, characteristics of each language, a commonly preferred translation method, and common feedback on a translation result.

With such a distinction between local storage and main storage, technical terms in specialized fields can be translated naturally, and in the case of book translation, more natural translation is possible by using information about different styles for each author. It is not limited thereto, and by storing and managing individualized information for each translation, natural and professional translation results can be obtained compared to simple statistical-based translation that does not divide fields and tasks.

For example, when translating a Korean patent specification into an English patent specification, the word 'drawing' may be briefly displayed in the conventional translation technology as 'FIG', but when the Korean translation of 'FIG' of the English patent specification is performed According to some embodiments of the present invention, more information on the patent field may be added to be translated as a 'drawing' rather than a 'fig'.

Also, in the field of book translation, the sentence "In my younger and more vulnerable years", depending on the style, is "when I was younger and more fragile than I am now", "When I was young and ignorant of the world" or "When I was young, I was much It was softer and weaker.” It can be interpreted in various sentences. According to some embodiments of the present invention, a natural style similar to the preceding style may be translated using the translation record of the previous makeup.

For example, referring to FIG. 44 , even in the same English sentence "old man and woman", when "old" is interpreted as meaning modifying both "man and woman" when considering the user's previous translation record, "old man and woman" in Spanish It can be translated as "Viejo y Vieja" (442), meaning "man and old woman".

On the other hand, given the user's previous translation history, if a certain adjective is often interpreted as modifying only the noun that follows it, "Viejo" means "old male and (not old) female" in Spanish. y mujer" (444). As described above, according to the present embodiment, even for the same sentence, the user's intended translation result can be extracted using the user's previous translation record.

Even the same sentence may be translated by using a record in which the user has previously performed a translation or a record requested as feedback for the extracted translation result. The feedback on the translation result may be, for example, information on merged or separated nodes and edges.

For example, if there are many records of merging a graph element corresponding to a subject and a graph element corresponding to a preposition, a sentence that can be interpreted as "The woman saw a man with binoculars" is a graph element corresponding to the subject depending on the user. If there are many records that merge graph elements corresponding to and object, it can be interpreted as "The woman saw the man with binoculars". In a similar embodiment, in the case of a user who does not prefer the own edge of the graph, a graph in which the own edge and the corresponding node are merged may be generated.

Hereinafter, when a plurality of sentences are translated with reference to FIG. 45 , a method of performing translation in consideration of contexts of the plurality of sentences will be described. In the case of homonym words having various meanings, accurate translation is possible only when the sentence containing the word and the context before and after the sentence are understood. For example, the word "ship" could mean 'a part of a person's or animal's body', 'a kind of fruit', or 'a vehicle that floats on water'. According to the present embodiment, all dictionary meanings of the "ship" may be extracted from the main storage.

However, when the translation of the sentence "I have a good stomach" is performed, a problem arises that the meaning of the "pear" is completely unknown. In order to solve the problem of the present invention, in order to understand the meaning of a specific word, the meaning of the "ship" is grasped using a sentence including the specific word and a plurality of sentences existing before and after the sentence.

For example, if a plurality of sentences "I like sea activities. I like boats" are inputted, the most relevant " A vehicle floating on water" can be understood as meaning "ship". Accordingly, the plurality of sentences may be interpreted as "I like maritime activities. I like ship."

On the other hand, if a plurality of sentences "I like fruit. I like pears" are input, "One kind of fruit" most related to the sentence "I like fruits" among the meanings of the plurality of "pears" " can be understood as the meaning of the "ship". Accordingly, the plurality of sentences may be interpreted as "I like fruits. I like a pear".

The determination of the relation between the specific sentence and the specific word may be calculated using a vast translation record, and machine learning learned using a set of various words and example sentences may be used.

Hereinafter, an operation of hardware of the apparatus 100 for performing translation using a visualized sentence according to another embodiment of the present invention will be described in detail with reference to FIG. 46 .

The translation apparatus 100 includes a graph transformation unit 623 , a translation unit 624 , and a translation record DB 614 , and in some embodiments, a user interface 601 , a sentence element extraction unit 621 , and a matrix transformation unit ( 622), a sentence element DB 611 for each language, a grammar DB 612 for each language, and a dictionary DB 613 for each language may be further included.

The user interface 601 may obtain a sentence to be translated from a user or an external server, and may output a matrix in which the sentence is converted, a graph, and a translated sentence. The acquired sentence may be transmitted to the sentence element extraction unit 621 . Also, the translated sentence may be received from the translation unit 624 . Also, feedback on the translation result input from the user may be transmitted to the translation unit 624 .

The sentence element extractor 621 may divide the sentence to be translated into morpheme units. Accordingly, the sentence can be divided by collecting information on morphemes for each language from the sentence element DB 611 for each language and the grammar DB 612 for each language. The sentence divided into morpheme units may be transmitted to the matrix transformation unit 622 or the graph transformation unit 623 .

The matrix transformation unit 622 may convert the sentence divided into morpheme units into a matrix form. In addition, the sentence converted into a matrix form may be transmitted to the graph transform unit 623 .

The graph conversion unit 623 may convert a sentence divided into morpheme units or a sentence in a matrix form into a graph. Also, the graph and the sentence corresponding to the graph may be transmitted to the translation unit 624 .

The translator 624 may interpret a sentence corresponding to each node and an edge of the graph. A word order of the interpreted sentence may be determined according to a traversal order of a graph corresponding to the sentence. The graph traversal order may be an order determined through past translation records, an order obtained through machine learning, or an order specified by a user.

Specifically, a translation result optimized for a user or context may be generated using the dictionary DB 613 for each language and the translation record DB 614 . Also, the translation unit 624 may generate a new translation result by using the user's feedback information on the translation result received from the user interface 601 , and transmit the information to the translation record DB 614 for storage.

As shown in FIG. 47 , the computing device 100 for performing translation using a visualized sentence according to another embodiment of the present invention includes a processor 710 , a memory 720 , a display 750 and a storage ( 760) and may further include a network interface 740 and a system bus in some embodiments.

One or more instructions 731 , 732 , and 733 loaded and stored in the memory 720 are executed through the processor 710 . Note that the computing device 100 performing the translation method according to the present embodiment may perform the translation method described with reference to FIGS. 1 to 47 even if there is no separate description.

The network interface 740 may receive the morpheme DB 761 and the dictionary DB 762 from an external server. The received information may be stored in the storage 760 .

The storage 760 may store the morpheme DB 761 , the dictionary DB 762 , and the translation record DB 701 .

The one or more instructions include a morpheme extraction instruction 731 for dividing a sentence into morpheme units, a graph transformation instruction 732 for generating a graph corresponding to the sentence, and a morpheme translation instruction for performing translation on each morpheme of the sentence ( 733) may be included.

In an embodiment, the morpheme extraction instruction 731 may divide the sentence into morpheme units and classify the sentence into a noun morpheme and other morphemes other than the noun morpheme.

In an embodiment, the graph transformation instruction 732 may generate nodes and edges of the graph to correspond to the noun morpheme and other morphemes.

In an embodiment, the morpheme translation instruction 733 may translate and display each of the elements of the graph corresponding to the noun morpheme and other morphemes into a language to be translated.

The methods according to the embodiments of the present invention described so far may be performed by executing a computer program implemented as computer readable code. The computer program may be transmitted from the first computing device to the second computing device through a network such as the Internet and installed in the second computing device, thereby being used in the second computing device. The first computing device and the second computing device include all of a server device, a physical server belonging to a server pool for cloud services, and a stationary computing device such as a desktop PC.

The computer program may be stored in a recording medium such as a DVD-ROM or a flash memory device.

Although the embodiments of the present invention have been described above with reference to the accompanying drawings, those of ordinary skill in the art to which the present invention pertains can realize that the present invention can be embodied in other specific forms without changing the technical spirit or essential features. can understand Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive.

Claims (24)

A method performed by a computing device, comprising:
obtaining data of a first sentence of a first language, wherein the first sentence includes a first morpheme of a first type and a second morpheme of a second type different from the first type;
A first graph representing the first sentence is generated, wherein a first node of the first graph corresponds to the first morpheme of the first type of the first sentence, and A first edge corresponds to the second morpheme of the second type of the first sentence, and each node and each edge of the first graph are traversed in a first order while traversing the first graph. connecting the first node and the first trunk of the first graph so that the first sentence is completed when expressions corresponding to are concatenated;
replacing expressions of the first language corresponding to each node and each edge of the first graph with expressions of a second language, respectively;
By traversing the first graph in a second order that is at least partially different from the first order, and connecting the expressions of the second language corresponding to each node and each edge of the first graph, the second language of the second language obtaining a sentence; and
outputting the second sentence as a result of the translation of the first sentence into the second language;
The step of generating the first graph comprises:
determining whether the second morpheme is an adjective, a determinant, a verb or a preposition, and generating the first graph including self-edges or connecting edges corresponding to the second morpheme according to the determination result;
Wherein the own trunk is an trunk in which the node of the destination and the node of the origin are the same, and the connecting trunk is an trunk connecting different nodes,
How to translate a sentence. The method of claim 1,
the first type is a noun type,
wherein the second type is a non-noun type,
How to translate a sentence delete The method of claim 1,
The step of generating the first graph comprises:
When the second morpheme is any one of an adjective and a determiner, generating the first graph including the self-edge corresponding to the second morpheme;
How to translate a sentence. The method of claim 1,
The step of generating the first graph comprises:
generating the first graph including the connecting edge corresponding to the second morpheme when the second morpheme is any one of a verb and a preposition;
How to translate a sentence. delete delete The method of claim 1,
The step of generating the first graph comprises:
In response to a predetermined user input for the first trunk of the first graph, changing any one or more of a departure point and a destination of the first trunk;
How to translate a sentence. The method of claim 1,
The step of generating the first graph comprises:
In response to a predetermined user input for the first node of the first graph, comprising the step of merging the first node with another node,
How to translate a sentence. The method of claim 1,
The step of generating the first graph comprises:
In response to a predetermined user input for the first node of the first graph, comprising the step of dividing the first node,
How to translate a sentence. 11. The method of claim 10,
The step of generating the first graph comprises:
In response to a predetermined user input for the first node of the first graph, dividing the first node and replacing it with two or more new nodes,
How to translate a sentence. 11. The method of claim 10,
The step of generating the first graph comprises:
In response to a predetermined user input for the first node of the first graph, dividing the first node and replacing it with one or more nodes and one or more trunk lines,
How to translate a sentence. The method of claim 1,
The step of generating the first graph comprises:
In response to a predetermined user input for any one of the first trunk and the first node of the first graph, the first trunk and the first node are merged and replaced with a new node,
How to translate a sentence. delete delete The method of claim 1,
Obtaining a second sentence of the second language comprises:
The second language corresponding to each node and each edge of the first graph while traversing the first graph in the second order that is at least partially different from the first order and designated using the translation record of a previous user obtaining the second sentence in the second language by concatenating the expressions;
How to translate a sentence. The method of claim 1,
Obtaining a second sentence of the second language comprises:
In response to receiving a predetermined user input, the second language corresponding to each node and each edge of the first graph while traversing the first graph in a third order that is at least partially different from the second order obtaining the second sentence in the second language by concatenating the expressions;
How to translate a sentence. delete delete delete delete delete memory into which the sentence translation program is loaded; and
A processor for executing a sentence translation program loaded into the memory,
Sentence translation program,
an instruction to obtain data of a first sentence of a first language, wherein the first sentence includes a first morpheme of a first type and a second morpheme of a second type different from the first type;
A first graph representing the first sentence is generated, wherein a first node of the first graph corresponds to the first morpheme of the first type of the first sentence, and A first edge corresponds to the second morpheme of the second type of the first sentence, and each node and each edge of the first graph are traversed in a first order while traversing the first graph. an instruction for connecting the first node of the first graph and the first trunk line so that the first sentence is completed when expressions corresponding to are concatenated;
instructions for replacing expressions of a first language corresponding to each node and each edge of the first graph with expressions of a second language, respectively;
By traversing the first graph in a second order that is at least partially different from the first order, and connecting the expressions of the second language corresponding to each node and each edge of the first graph, the second language of the second language instructions to get a sentence; and
and instructions for outputting the second sentence as a result of the translation of the first sentence into the second language,
The instruction for generating the first graph is,
and instructions for determining whether the second morpheme is an adjective, a determiner, a verb or a preposition, and generating the first graph including an own trunk line or a connecting edge corresponding to the second morpheme according to the determination result;
Wherein the own trunk is an trunk in which the node of the destination and the node of the origin are the same, and the connecting trunk is an trunk connecting different nodes,
computing device. delete

Images