KR101954887B1 - Method, apparatus and computer program for recommending alternative word for muticultural members - Google Patents

Abstract

The present invention relates to an alternative word recommendation method, apparatus, and computer program for a multicultural member. The present invention relates to an alternative word recommendation method, apparatus, and computer program for a multicultural member, A method of recommending, comprising: receiving user characteristic information and a translation request of a first word from a user; extracting one or more candidate words similar in meaning to the first word from a word database of a second language; Calculating a degree of difficulty of the candidate word by using the attribute-based score and the difficulty-by-attribute weight, and providing the user with the degree of difficulty of the candidate word and the candidate word, Wherein the difficulty weight for each attribute includes at least one of a nationality of the user, a native language of the user, A second country residence period in which the second language of the user is an official language, or a second language learning period of the user. According to the present invention, according to the present invention, not only a word level set by an expert but also difficulty levels are evaluated according to attributes of words such as lexical type, number of syllables, complexity of words, and the like, Can provide more suitable alternative words to the user.

Description

[0001] METHOD, APPARATUS AND COMPUTER PROGRAM FOR RECOMMENDING ALTERNATIVE WORD FOR MUSICULTURAL MEMBERS [0002]

The present invention relates to an alternative word recommendation method, apparatus, and computer program, and more particularly, to an alternative word recommendation method, apparatus, and computer program for multicultural members.

Multicultural society is accelerating due to the development and internationalization of transportation and communication. Multicultural society has the advantage of contributing to the development of society by bringing out various resources and cultural communication. However, from the individual viewpoint of each member of the multicultural society, users living in other countries suffer from many inconveniences due to language barriers and cultural barriers.

In the case of the Republic of Korea, we provide immigrant interpreter / translation support services to minimize conflicts in multicultural societies and to support the language of marriage immigrants. However, there is a high frequency of interpreting Korean information, It is true. In the case of marriage immigrants, Koreans have many years of experience in learning Korean while living in Korea. In many cases, they have the necessary level of Korean language ability. However, communication with educational institutions and administrative agencies requires difficult Korean language such as foreign language, As a result, multicultural members are experiencing difficulties in communicating with public institutions.

On the other hand, the development of artificial intelligence technology is improving the translation technology, and the result of the translation is improving. However, the information translated into the languages of multicultural members is likely to accept the translated information as it is without verification, and as a result, the translation error is likely to lead to erroneous decision making.

On the other hand, the national language institute and Korean service providers provide information on easy Korean language in the multicultural society. Most services provide easy Korean language services based on the difficulty assessed based on national survey, frequency of use, and curriculum of Korean language. This makes it easier for Korean language learners to judge Korean language difficulties The level of difficulty of multicultural members may be somewhat different.

An object of the present invention is to provide an alternative word recommendation method, an apparatus, and a program that considers characteristics and levels of individual users such as the country of origin of a user (a multicultural member) and a residence period.

In addition, the present invention evaluates the degree of difficulty by not only the word level set by the expert but also the attribute of the word such as the lexical type, the number of syllables, the complexity of the word, and recommends the alternative word considering the property, To provide a more suitable alternative word to the user.

Another object of the present invention is to improve the accuracy of the alternative word recommendation by continuously learning the words defined by the user at a high degree of difficulty and adjusting the difficulty weight for each word that the user perceives.

In order to achieve the above object, the present invention provides a method for recommending an alternative word of a second language having a meaning similar to that of a first word of a second language to a user whose first language is a first language, Receiving a conversion request of a first word, extracting at least one candidate word whose meaning is similar to the first word in a word database of a second language, calculating a score for each attribute of the candidate word, Calculating a difficulty score of the candidate word by using a star score and a difficulty weight by attribute, and providing the user with a difficulty score of the candidate word and the candidate word, The national language of the user, the mother language of the user, the second language of the user, And a period, or it characterized in that the one corresponding to the user characteristic information includes at least one of the second language learning period of the user.

The present invention also provides an apparatus for recommending an alternative word in a second language having a similar meaning to a first word in a second language and having a lower level of difficulty for a user whose first language is a first language, A communication module for receiving the alternative word information and transmitting the information of the alternative word to the user terminal; a word database for storing candidate words of at least one second language similar in meaning to the first word; And an analysis module for calculating a score for each attribute of the candidate word, calculating a difficulty score of the candidate word using the score for each attribute and the difficulty weight for each attribute, and recommending the candidate word as the alternative word, Wherein the difficulty weight for each attribute includes at least one of a nationality of the user, a native language of the user, And in a second state residency, or it characterized in that the one corresponding to the user characteristic information includes at least one of the second language learning period of the user to the air as an official language.

According to the present invention as described above, it is possible to provide an alternative word recommendation method, an apparatus, and a program that considers characteristics and levels of individual users such as the country of origin of a user (a multicultural member) and a residence period.

Further, according to the present invention, not only the word level set by the expert but also the difficulty level is evaluated according to the attribute of the word such as the lexical type, the number of syllables, the complexity of the word, and the attribute that the user feels particularly vulnerable. Alternative words can be provided.

In addition, according to the present invention, since the difficulty weight for each word is adjusted by continuously learning the words defined by the user at a high degree of difficulty, the accuracy of the alternative word recommendation is enhanced.

1 is a diagram for explaining an alternative word recommendation apparatus and its operating environment according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating an alternative word recommendation method according to an embodiment of the present invention.
FIG. 3 is a flowchart for explaining another embodiment of the alternative word recommendation method according to the present invention;
FIG. 4 is a flowchart for explaining a reference word collection method according to an embodiment of the present invention; FIG.
5 is a flowchart for explaining a reference word collection method according to another embodiment of the present invention;
FIG. 6 is a diagram for explaining word attributes and user characteristics used in calculating difficulty scores according to an exemplary embodiment of the present invention.

The above and other objects, features, and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings, which are not intended to limit the scope of the present invention. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail.

In the drawings, the same reference numerals are used to designate the same or similar components, and all combinations described in the specification and claims can be combined in any manner. It is to be understood that, unless the context requires otherwise, references to singular forms may include more than one, and references to singular forms may also include plural forms.

The terminology used herein is for the purpose of describing particular illustrative embodiments only and is not intended to be limiting. Singular representations as used herein may also be intended to include a plurality of meanings, unless the context clearly dictates otherwise. The terms "and / or" " and / or "include any and all combinations of the items listed therein. The terms "comprises," "comprising," "including," "having," "having," "having," and the like have the implicit significance, Steps, operations, elements, and / or components, and does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / Steps, processes, and operations of the methods described herein should not be construed as necessarily enforcing their performance in such specific order as discussed or illustrated unless specifically concluded the order of their performance . It should also be understood that additional or alternative steps may be used.

In addition, each of the components may be implemented as a hardware processor, the components may be integrated into one hardware processor, or a combination of the components may be implemented as a plurality of hardware processors.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the example of the present specification, 'user' means a multicultural member, and may be understood as a user who has a first language as a mother language and requests an alternative word for any word in the second language.

For example, according to one embodiment of the present invention, the 'user' may be a foreigner who uses or primarily uses a language other than Korean, for example, English, Thai, Vietnamese (first language) The second language may be Korean. In another embodiment, the 'user' may be a Korean who is native to Korean, in which case the second language may be a language other than Korean.

That is, according to the present invention, the 'user' can request an alternative word that can replace the arbitrary word in a language not familiar to him or her, which is not his / her native language, if the degree of difficulty of the word is high , The alternative word recommendation apparatus of the present invention can recommend a word having a similar meaning to the word but a lower degree of difficulty as an alternative word when receiving the request from the user.

In the present specification, the case where the first language is a foreign language and the second language is Korean will be described as an example.

1 is a diagram for explaining an alternative word recommendation apparatus and its operating environment according to an embodiment of the present invention. Referring to FIG. 1, an alternative word recommendation apparatus (or server) 100 according to an embodiment of the present invention includes a communication module 110, an analysis module 130, and a word database 150, And a user database 170 for storing information.

The user terminal 50 is a terminal used by a multicultural member according to an embodiment of the present invention and is capable of communicating with the server 100 such as a smart phone 50a, a tablet PC 50b, a personal PC or a notebook computer 50c A display capable of displaying information, and an electronic device capable of receiving an input signal from a user. Therefore, the user terminal 50 in the present invention is not limited by the type of the electronic device if it is an electronic device capable of installing and operating the alternative word recommendation application according to the present invention.

An application program for executing an embodiment of the present invention may be installed in the user terminal 50. The application program may be installed as an extension program of another application program such as a moving picture playback application program or an Internet browser. Accordingly, when a user watches a video lecture and then a word unknown to the subtitle is displayed, a conversion request is transmitted to the alternative word recommendation apparatus 100 according to an embodiment of the present invention when the word is clicked, The alternative word may be displayed on the user terminal 50 if the alternative word having a similar meaning to the unknown word is provided to the user terminal 50 by recommending an alternative word having a lower degree of difficulty. In another embodiment, alternate words may be displayed on the screen if the user is mouse-over a word that they do not know while surfing the web.

Hereinafter, functions of the alternative word recommendation apparatus 100 according to an embodiment of the present invention will be described.

First, the communication module 110 receives a conversion request of the first word of the second language from the user terminal 50, and transmits an alternative word corresponding to the conversion request to the user terminal 50. [ In addition, the communication module 110 can receive difficulty evaluation information on a certain word from the user terminal 50, and can receive the user characteristic information input by the user from the user terminal 50 to the user terminal 50, A second country residence period in which the second language of the user is an official language, or a second language learning period of the user.

In addition, the communication module 110 can transmit the score of the attribute of the reference word analyzed by the analysis module 130 to the user terminal 50 and the score of the property of the alternative word to the user terminal 50, Can provide information so that the score of each attribute and the difficulty score of each word can be confirmed.

The communication module 110 can be understood as a module 110 that transmits and receives data to and from a user terminal 50 using a wired or wireless network and supports data transmission and reception regardless of a communication method.

Therefore, the communication module 110 may include at least one of a mobile communication module, a wireless Internet module, and a short-range communication module 114. The communication module 110 may include at least one of a mobile communication module, And may be a wireless Internet module such as WLAN, WiFi Direct, DLNA Wibro, Wimax, HSDPA, and LTE. In addition, the communication module 110 may be any one of Bluetooth, RFID, IrDA, UWB, ZigBee, NFC, Wi-Fi, and Wi-Fi Direct technologies supporting short- Or the like.

The analysis module 130 is a module for recommending an alternative word having a similar meaning and a lower level of difficulty to the first word for which conversion is requested through the user terminal 50. The analysis module 130 extracts a candidate word from the word database 150, The score of each attribute is calculated, and the difficulty score of the candidate word is calculated using the attribute score and the attribute difficulty weight by property, and the candidate word can be recommended as an alternative word in the order of the lowest difficulty score. Wherein the difficulty weight for each attribute includes at least one of a nationality of the user, a mother tongue of the user, a second country residence period in which the second language of the user is an official language, or a second language learning period of the user May be a value corresponding to the information. The difficulty weight for each attribute may be a value that is updated by learning a reference word of a second language defined by a user at a high degree of difficulty.

More specifically, let's look at the functions of the analysis module 130. The analysis module 130 may extract one or more candidate words similar in meaning to the first word from the word database 150 using the first word received from the communication module 110. [

Let's take an example of a case in which a user A of Vietnamese national who has Vietnamese as a mother tongue and a ten year old resident in Korea requests conversion of the word 'farmer'. The user A can input a signal requesting an easy Korean translation of the word 'farmer' through an application program installed in the user terminal 50. In addition, user A can input information such as his / her nationality, Vietnamese, mother tongue: Vietnamese, and ten years of living in Korea using the above application program. Characteristic information of the user A and an easy Korean translation request are transmitted to the alternative word recommendation apparatus 100 through the user terminal 50 and the communication module 110 can receive the characteristic information and the translation request for the Korean word ' The request may be sent to the analysis module 130 to act on the user A to recommend the easier word with a similar meaning to the 'farmer'.

That is, the analysis module 130 can extract 'farmer', 'farmer', 'serf', and 'farmer', which are similar in meaning to 'farmer', from the word database 150, The star score can be calculated.

According to an embodiment of the present invention, each word stored in the word database 150 may include a score for each attribute. The attributes of a word can be classified into a word level, a vocabulary, a syllable, a complexity, an emphasis, a tense, a gender, and can have a property-specific score according to predetermined criteria. For example, the difficulty score (hereinafter, referred to as 'attribute score') for each attribute can be calculated according to the criteria shown in Table 1 below.

division Word level Vocabulary Syllable Complexity High reputation tense By property
score
Calculation Criteria Awards: 1.4
Of: 1
Ha: 0.7 General language: 1
Loan word: 0.9
Chinese: 2
Foreign language: 0.8 1 to 3 syllables: 1
3 to 4 syllables: 1.2
5 to 7 syllables: 1.4
7 syllables or more: 2 Basic language: 1
Synonyms: 1.2
Derivative: 1.4 Basic language: 1
Highest word: 1.2 Current: 1
Past: 1.2
Future: 1.2

In the attribute classification of the above example, 'word level' is a value classified qualitatively using a database for evaluating the degree of difficulty of a word conventionally such as a national language source or a thesis. In the example above, the word level is classified as upper / middle / lower, and the score is set, but it can be further classified.

'Vocabulary' is a property defined in consideration of difficulty level depending on the kind of vocabulary. It can be classified into general language, foreign language, Chinese character, and foreign language. In the above example, foreigners who do not speak Korean as their mother tongue can deduce a foreign language or a foreign language, so a relatively low difficulty score is set. Since the Chinese characters often feel difficult for Koreans, relatively high difficulty scores can be set.

'Syllable' is an attribute that reflects the tendency to feel relatively difficult when the number of syllables constituting the word becomes longer. In the example shown in Table 1 above, the difficulty score is higher as the number of syllables increases.

'Complexity' is an attribute to reflect the difficulty evaluation whether the word is a basic word, a compound word, or a derivative word. In addition, it can be used as an attribute for evaluating the degree of difficulty, whether or not it is a high-level word, a tense (present / past / future), and whether a word exists in a gender.

Referring to Table 1 above, the score of each of the candidate words 'farmer', 'farmer', 'sergeant', and 'farmer' in the above embodiment is calculated as follows.

division Word level Vocabulary Syllable score Complexity score A high score Tentative score farmer Ha: 0.7 Chinese: 2 1 to 3 syllables: 1 Basic language: 1 Basic language: 1 irrelevant Farmer Ha: 0.7 General language: 1 1 to 3 syllables: 1 Synonyms: 1.2 Basic language: 1 irrelevant serf Awards: 1.4 Chinese: 2 1 to 3 syllables: 1 Basic language: 1 Basic language: 1 irrelevant farmer Awards: 1.4 Chinese: 2 1 to 3 syllables: 1 Basic language: 1 Basic language: 1 irrelevant

The computed attribute-specific score may be stored in the word database 150 and then used in alternative word suggestions based on other conversion requests. In addition, the attribute-based score can be stored in advance in each word in the word database 150 irrespective of the conversion request. In this case, the analysis module 130 can call the score for each attribute of the candidate word and the candidate word from the word database 150 Can be used.

The analysis module 130 that calculates the score for each attribute can calculate the difficulty score of each of the candidate words by using the score for each attribute and the difficulty weight for each attribute.

In this case, the difficulty weight for each attribute corresponds to the user characteristic information and may be stored in the user database 170 in advance.

The difficulty weight according to each attribute may have different values depending on the nationality of the user, the mother language of the user, the second country residence period in which the user's second language is the official language, or the second language learning period of the user, May be stored in the database 170. The difficulty weight for each attribute corresponding to the user characteristic information may be preset, and may be updated or changed by using (learning) information input from the user.

It is assumed that the initial values of the difficulty weights are set in advance according to the nationality in each of the following embodiments. Since the nationality of the user who transmitted the request is Vietnam, the analysis module 130 may call the difficulty weight for each attribute (Table 3) corresponding to Vietnam in the user database 170.

division Word level Vocabulary Syllable Complexity High reputation tense difficulty
weight 50% 20% 10% - 5% 15%

In the example of Table 3, the attributes of the words are classified into word level, vocabulary, syllable, complexity, high-order word, and tense, and the difficulty weights are 50%, 20%, 10%, 5%, 15% Is set. The difficulty weight corresponding to the nationality can be understood as having the following meaning. The effect of the vocabulary type (general language, foreign language, Chinese language, foreign language, etc.) is 20%, the influence of the number of syllables is 10%, the influence of the syllable is 5% Past, present, future, etc.) have a weight of 15%.

If there is an exaggeration in country A and there is no exaltation in country B, the users of country A will feel that it is not difficult to say that there is a higher level of presence in Korean than country B users. This is similar to how Koreans who do not distinguish the sex of a word feel more difficult when learning gender-sensitive languages such as French and Vietnamese.

The alternative word recommendation apparatus according to an embodiment of the present invention reflects the characteristics of the linguistic characteristic of the country in the attribute and recommends an alternative word in consideration of the user's language as a mother tongue, Can be recommended.

In another embodiment, the difficulty weight according to the property may have a different value depending on the user's residence period. For example, when a user who speaks a first language as a first language has a period of one year residing in a country that uses a second language, and more than ten years, the criteria for evaluating the degree of difficulty of each user may be different. For example, in the case of learning Korean, it is possible to acquire the accent in a short period of time. However, if it is judged that the vocabulary is difficult to acquire in the short term, the difficulty weight of the vocabulary attribute corresponding to the residence period of 1 year and the vocabulary corresponding to the residence period of 10 years The difficulty weights of the attributes may not differ greatly. In this case, however, the weight of the high end corresponding to the residence period of 10 years will be set to a value significantly lower than the difficulty weight of the high end corresponding to the residence period of one year.

In the above-described embodiment, it is assumed that only the nationality of the user characteristic information is considered. Since the user who requests the alternative word recommendation of the 'farmer' is a person of Vietnamese nationality, the difficulty score of each candidate word is calculated using the score of the attribute of the candidate word in Table 2 and the difficulty weight of each attribute in Table 3 as shown in Table 4 .

However, in the example of Table 4, the weighted score of each attribute is multiplied by the score of each attribute and the difficulty weight of each attribute, and the weighted score of each attribute is summed up to calculate the score of the candidate word The difficulty score was calculated.

division Difficulty score Vocabulary score Syllable score Complexity score A high score Tentative score Sum farmer Ha: 0.7 Chinese: 2 1 to 3 syllables: 1 Basic language: 1 Basic language: 1 irrelevant 0.95 Farmer Ha: 0.7 General language: 1 1 to 3 syllables: 1 Synonyms: 1.2 Basic language: 1 irrelevant 0.75 serf Awards: 1.4 Chinese: 2 1 to 3 syllables: 1 Basic language: 1 Basic language: 1 irrelevant 1.3 farmer Awards: 1.4 Chinese: 2 1 to 3 syllables: 1 Basic language: 1 Basic language: 1 irrelevant 1.3 difficulty
weight 50% 20% 10% - 5% 15%

The analysis module 130 delivers the difficulty score and the candidate word of each calculated candidate word to the communication module 110 and the alternative word recommendation apparatus 100 according to an embodiment of the present invention calculates the degree of difficulty of the candidate word and the candidate word And send the score to the user terminal 50.

In another embodiment, the alternative word recommendation apparatus 100 arranges the order of the difficulty of the candidate words in descending order, namely, farmer> farmer> farmer> farmer and farmer, and transmits them to the user terminal 50, You can recommend an alternative word.

The difficulty weights corresponding to the nationality, the mother tongue, the residence period, the learning period, and the like can be set differently by using the user characteristic information of the user using the alternative word recommendation apparatus, Can be adjusted to have a more personalized value. In this specification, the 'reference word' is a word used by the alternative word recommendation apparatus 100 according to an embodiment of the present invention to adjust the difficulty weight according to the attribute, It means a word in two languages.

That is, the alternative word recommendation apparatus 100 according to an exemplary embodiment of the present invention can update the difficulty weight for each attribute by a value that is specific to the user by learning the reference word received through the user terminal 50.

More specifically, the alternative word recommendation apparatus 100 collects reference words of a second language defined by the user at a high degree of difficulty from the user terminal 50, and adjusts the difficulty weights by attribute using the reference words have. The adjusted difficulty weight may be designated as a customized difficulty weight that is specific to a specific user. In calculating the difficulty score of the candidate word, the analysis module 130 may calculate the score of the candidate word using the attribute- The difficulty score can be calculated.

The alternative word recommendation apparatus 100 receives the arbitrary word of the second language from the user terminal 50 and then receives the degree of difficulty evaluation information about the arbitrary word from the user terminal 50. When the degree of difficulty evaluation information is high, , Reference words can be collected by collecting arbitrary words as reference words. That is, it is possible to randomly provide a word in the form of a questionnaire, or to provide the user with a selected word according to the degree of difficulty, and then to evaluate the difficulty level of each word. And words with high difficulty levels can be used as reference words to adjust the difficulty weight.

In yet another embodiment, the alternative word recommendation device 100 may provide any word of the second language to the user terminal 50 and may receive a translation request or translation request for any word from the user terminal 50 , The frequency of translation requests or translation requests can be counted. If the counted frequency is more than the predetermined value, arbitrary words can be collected as reference words.

That is, if the user requests conversion of some words among contents provided through the user terminal 50 to an easy Korean (alternative word) or requests translation to the user's native language, It is possible to judge that the user recognizes the word and the translation requested word difficultly. In addition, the difficulty level of the corresponding word can be gradually increased as the request is repeated (as the frequency is increased), and the analysis module 130 learns the difficulty of the word and analyzes the attribute of the word, Can be utilized.

In adjusting the difficulty weight for each attribute using the reference word, the analysis module 130 calculates the score for each attribute of the reference word, analyzes a plurality of reference words, and if the average of the first attribute scores is equal to or greater than a preset reference value , The difficulty weight of each attribute may be adjusted by increasing the difficulty weight of the first attribute.

For example, let us assume that the user analyzes the words that are requested to be translated and, as a result, it is assumed that the words of common tongue / foreign language / Chinese language / foreign language are mixed, but all have syllable number of more than 5 syllables and are words representing future tense. In evaluating difficulty, the above users are not significantly affected by the type of vocabulary, but can be judged to be greatly affected by syllable and tense. Therefore, the analysis module 130 can adjust the customized difficulty weight value for each attribute in the direction of increasing the difficulty weight of the syllable and the difficulty weight of the tense without changing the vocabulary difficulty weight.

Also, the analysis module 130 may calculate the degree of difficulty weight according to the user characteristics and the attributes, using the user characteristic information received from the plurality of users and the customized difficulty weights according to the attributes specific to each user. The analysis module 130 may adjust the initial value of the difficulty weight for each attribute by using the degree of correlation.

For example, the alternative word recommendation apparatus 100 may be configured such that a plurality of users having different characteristics (nationality, mother tongue, residence period, learning period, etc.) collects the alternative word recommendation apparatus 100 of the present invention for a long period of time The difficulty weight value of each attribute corresponding to the user characteristic information stored in the user database 170 can be adjusted by using the information to be learned, i.e., the customized difficulty weight specific to each user and the characteristic information of each user.

In the embodiment of Table 3 described above, the difficulty weights according to the attributes of the Vietnamese nationality are classified into word level, vocabulary, syllable, complexity, high-order word and tense, and the difficulty weight is 50%, 20%, 10% , And 5% and 15%, respectively. As a result of analysis of user usage pattern and reference words according to translation request and translation request of 100 users of Vietnamese nationality, it was found that the correlation between Vietnamese nationality and high honorificance was high lets do it. In this case, the analysis module 130 may increase the difficulty weight corresponding to the attribute of the loyalty, so that the actual user characteristic may be reflected in the alternative word recommendation.

Hereinafter, an alternative word recommendation method according to an embodiment of the present invention will be described with reference to FIG. 2 to FIG.

2 is a flowchart illustrating an alternative word recommendation method according to an embodiment of the present invention. Referring to FIG. 2, the server may receive user characteristic information and a first word conversion request from a user terminal (S100). The server extracts one or more candidate words having similar meaning to the first word from the word database of the second language (S200), and calculates a score for each attribute of the candidate word (S300). Next, the server may calculate the difficulty score of the candidate word by using the attribute score and the attribute difficulty weight according to the property (S400). Here, the difficulty weight according to the attribute is determined by the nationality of the user, the mother language of the user, Or a second language learning period of the user, based on the user characteristic information.

In step 400, the server multiplies the attribute-based scores by the attribute-based scores and the attribute-based difficulty weights to calculate the attribute-weighted score, and calculates the difficulty score of the candidate word by summing all the attribute-weighted scores.

When the server calculates the difficulty score of the candidate word, the server may transmit the difficulty score of the candidate word and the candidate word to the user terminal (S500). In addition, candidate words may be recommended in descending order of difficulty score (S600), and the recommended words may be transmitted to the user terminal (S700).

The difficulty weight for each attribute may be updated by learning the reference word of the second language defined by the user at a high degree of difficulty. Referring to FIG. 3, the server collects reference words of a second language defined by a user at a high degree of difficulty (S10), and adjusts the difficulty weights by attribute using reference words (S30).

In step 30, the server calculates a score for each attribute of the reference word, analyzes a plurality of reference words, and increases the difficulty weight of the first attribute when the average of the first attribute scores is equal to or greater than a preset reference value, The weights can be adjusted.

The server sets the difficulty weight according to the adjusted attribute to the user-customized difficulty weight according to the attribute (S50), and when the conversion of the first word is requested from the user terminal (S100), after executing the steps 200 to 300, In operation S450, the degree of difficulty of the candidate word may be calculated using the attribute-based score and the customized difficulty weight for each attribute (S450).

4, the server provides the user with an arbitrary word of the second language (S12), receives the difficulty evaluation information on the arbitrary word from the user (S14) , It is determined whether or not the degree of difficulty evaluation information corresponds to a high degree of difficulty (S16). If so, any word can be collected as a reference word (S18).

5, the server provides the user with a random word of the second language (S11), and when receiving a translation request or translation request for any word from the user S13), the frequency of the conversion request or the translation request is counted (S15), and if the frequency is not less than the preset value (S17), any word can be collected as the reference word (S19).

Although not shown in the figure, the server calculates the degree of difficulty weight according to user characteristics and attributes by using user characteristic information received from a plurality of users and customized difficulty weights according to attributes specific to each user, The initial value of the difficulty weight for each attribute can be adjusted.

The embodiments of FIGS. 2 to 5 are examples in which the alternative word recommendation method of the present invention is implemented in a server, and the alternative word recommendation method of the present invention may be implemented through an application program installed in the user terminal. In this case, the 'user terminal' shown in FIGS. 2 to 5 may be replaced with 'user', and the 'server' shown in FIGS. 2 to 5 may be replaced with a 'user terminal'. Since only the subject who executes the method is changed in this embodiment, an example in which the alternative word recommendation method according to the present invention is executed through an application program installed in the user terminal will be omitted in this specification.

The methods described herein may be implemented by one or more computer programs executed by one or more processors. Computer programs include processor-executable instructions stored on a non-transitory type computer readable medium. The computer programs may also include stored data. Non-limiting examples of non-transitory tangible computer readable media are non-volatile memory systems, magnetic storage, and optical storage.

Certain embodiments of the techniques described above include processing steps and instructions described herein in an algorithmic form. The processing steps and instructions described above may be implemented in software, firmware, or hardware, and when implemented in software, may be downloaded to reside on other platforms used in real-time network operating systems And it can be operated from there.

The present invention also relates to an apparatus for performing the operations herein. Such a device may be specially constructed for a desired purpose or may comprise a general purpose computer selectively activated or reconfigured by a computer program stored on a computer readable medium that can be accessed by a computer. Such a computer program may be stored on a computer-readable storage medium of a type, such as, for example, floppy disks, optical disks, CD-ROMs, magneto-optical disks (ROMs), random access memories (RAMs), EPROMs, EEPROMs, magnetic or optical cards, application specific integrated circuits ASICs), or any type of medium suitable for storing electronic instructions and each coupled to a computer system bus, but are not so limited. Moreover, the computers referred to herein may include a single processor, or may be architectures that use multiple processor designs to enhance computing capabilities.

The algorithms and operations presented herein are not inherently related to any particular computer or other devices. Various general purpose systems may also be used with the programs according to the teachings herein or it may prove convenient to construct devices that are more specifically designed to perform the steps of the desired method. The structure required for a variety of these systems, along with their equivalent variants, will be apparent to those skilled in the art. Additionally, the present disclosure is not described in connection with any particular programming language. It should be understood that various programming languages may be used to implement the teachings of the present disclosure as described herein, and that any reference to a particular language is intended to be illustrative of the embodiments of the invention and the best mode.

Some embodiments omitted in this specification are equally applicable if their implementation subject is the same. It is to be understood that both the foregoing general description and the following detailed description of the present invention are exemplary and explanatory and are intended to be exemplary and explanatory only and are not restrictive of the invention, The present invention is not limited to the drawings.

Claims (11)

A method for recommending an alternative word of a second language having a meaning similar to that of a first word of a second language to a user whose first language is a first language,
Receiving user characteristic information and a first word conversion request from a user;
Extracting from the word database of the second language one or more candidate words having similar meaning to the first word;
Calculating a score for each attribute of the candidate word;
Calculating a difficulty score of the candidate word by calculating a weighted reflection score of each attribute using the attribute-based score and the difficulty-by-attribute weighting value, and summing the weighted reflection score of each attribute;
And providing the user with a difficulty score of the candidate word and the candidate word,
The difficulty weight by attribute
An alternative word corresponding to user characteristic information including at least one of the nationality of the user, the mother tongue of the user, the second country residence period in which the second language of the user is an official language, or the second language learning period of the user Recommended method.
The method according to claim 1,
The difficulty weight by attribute
Wherein the reference word is updated by learning a reference word of a second language defined by the user at a high degree of difficulty.
The method according to claim 1,
Collecting a reference word of a second language defined by the user at a high degree of difficulty;
Adjusting the difficulty weight for each attribute by using the reference word;
Further comprising the step of setting the adjusted difficulty weight for each property as a customized difficulty weight for each attribute specific to the user,
The step of calculating the degree of difficulty of the candidate word
And calculating a degree of difficulty of the candidate word using the attribute-specific score and the attribute-based difficulty weight for each attribute.
The method of claim 3,
Wherein the collecting of the reference words comprises:
Providing the user with an arbitrary word of a second language;
Receiving difficulty evaluation information on the arbitrary word from the user;
And collecting the arbitrary word as the reference word if the degree of difficulty evaluation information corresponds to the high degree of difficulty.
The method of claim 3,
Wherein the collecting of the reference words comprises:
Providing the user with an arbitrary word of a second language;
Receiving a translation request or translation request for the arbitrary word from the user;
Counting the frequency of the conversion request or the translation request, and collecting the arbitrary word as the reference word if the frequency is not less than a predetermined value.
The method of claim 3,
The step of adjusting the difficulty weight by attribute using the reference word
Calculating a score for each attribute of the reference word;
Analyzing a plurality of reference words and increasing the difficulty weight of the first attribute if the average of the first attribute scores is greater than a predetermined reference value.
The method according to claim 1,
Calculating a degree of correspondence between the user characteristics and the difficulty weights according to the attributes using the user characteristic information received from the plurality of users and the customized difficulty weights according to the attributes specific to each user;
And adjusting an initial value of the difficulty weight by attribute using the correlation.
The method according to claim 1,
Wherein the attribute includes at least one of a word level, a lexical type, a syllable number, a complexity, an emphasis or a tense.
The method according to claim 1,
The step of calculating the degree of difficulty of the candidate word
Multiplying the attribution score by the attribute difficulty weight for each attribute and calculating a weight reflection score for each attribute;
And summing up the weighted score of each attribute to calculate a difficulty score of the candidate word.
An apparatus for recommending an alternative word of a second language having a similar meaning to a first word of a second language and having a lower level of difficulty to a user whose first language is a first language,
A communication module for receiving the conversion request of the first word from the user terminal and transmitting the information of the alternative word to the user terminal;
A word database for storing candidate words of one or more second languages similar in meaning to the first words;
Calculating a score for each attribute of the candidate word by extracting the candidate word from the word database, calculating a weighted score for each attribute by using the score for each attribute and the difficulty weight for each attribute, And an analysis module for calculating a difficulty score of the candidate word and recommending the candidate word as the alternative word,
The difficulty weight by attribute
An alternative word corresponding to user characteristic information including at least one of the nationality of the user, the mother tongue of the user, the second country residence period in which the second language of the user is an official language, or the second language learning period of the user Recommended device.
An alternative word recommendation application stored in a computer-readable medium for executing in a computer any one of the methods of claims 1 to 9.

Images