KR101624421B1 - Method for Recommending Emoticon and User Device for Recommending Emoticon - Google Patents

Abstract

A user terminal recommending an emoticon recommendation method and an emoticon is disclosed. The disclosed emoticon recommendation method includes: generating a hash value of text input from a user using a hash function; Generating a text signature of a bit string of K (where K is a natural number) bit string using the hash value according to a predetermined rule; And displaying at least one emoticon selected from the emoticon sets according to the similarity of the text signatures and the emoticon signatures.

Description

[0001] The present invention relates to an emoticon recommendation method and a user terminal that recommends an emoticon,

The present invention relates to a method for recommending an emoticon and a user terminal for recommending an emoticon, and more particularly to a method for recommending an emoticon using text mining and a user terminal recommending an emoticon.

In a multi-user chat application in a smartphone environment, users use emoticons to communicate their emotions effectively. An emoticon is a combination of an emotion and an icon, which is composed of sentence symbols, special characters, or images. Emoticons were first reported to be used by Carnegie Mellon college student Scott Pallman in the early 1980s, and in the early days when PC communication and the Internet had not become commonplace, smiley symbols (smiley symbols) It is also called.

As users' demand for emoticons grows, apps like KakaoTalk and Line provide users with various types of emoticons. To use emoticons, you can either search for the emoticons you want to use, or select the emoticons that the app itself recommends. A related prior art document is Korean Patent Publication No. 2011-0026218.

To recommend emoticons in the app itself, we recommend emoticons based on how often they use your emoticons. Since it does not reflect the intention of the user, it is highly likely that the emoticon corresponding to the meaning of the text inputted by the user is not recommended, and the user must search for the emoticon to be used through direct search.

The present invention provides a method for recommending emoticons corresponding to contents of text input from a user and a user terminal recommending emoticons.

According to an aspect of the present invention, there is provided a method of generating a hash value of a text input from a user using a hash function, Generating a text signature of a bit string of K (where K is a natural number) bit string using the hash value according to a predetermined rule; And displaying at least one emoticon selected from the emoticon sets according to the similarity of the text signatures and the emoticon signatures.

According to another aspect of the present invention, there is provided a method of generating a hash value for at least one keyword assigned to an emoticon using a hash function, Generating an emoticon signature of K (where K is a natural number) bit string using the hash value according to a predetermined rule; And storing the emoticon signature is provided.

According to another aspect of the present invention, there is provided a computer-readable recording medium having embodied thereon a hash value generator for generating a hash value of a text input from a user using a hash function; A text signature generation unit for generating a text signature of K (where K is a natural number) bit string using the hash value according to a predetermined rule; And a display unit for displaying at least one emoticon selected from the emoticon set according to the similarity of the text signatures and the emoticon signatures.

According to the present invention, by comparing the text signature of the bit string with the emoticon signature, rather than comparing the similarity between the text itself and the keyword of the emoticon, it is possible to provide a fast operation speed and reduce the capacity required for storing the emoticon keyword.

Also, according to the present invention, by generating a text signature by dividing text by N-gram, it is possible to provide strong recommendation performance even for grammatical errors such as spacing and typhus.

Further, according to the present invention, it is possible to improve the satisfaction of the user by recommending various emoticons to the user in the order of high similarity.

1 is a view for explaining a user terminal recommending an emoticon according to an embodiment of the present invention.
FIG. 2 is a diagram for explaining a method of recommending emoticons according to an embodiment of the present invention. Referring to FIG.
3 is a diagram for explaining a method of recommending emoticons according to another embodiment of the present invention.
4 and 5 are diagrams showing recommendation emoticons according to an embodiment of the present invention.
FIG. 6 is a diagram for explaining an emoticon recommendation method according to a specific embodiment of the present invention.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention. Like reference numerals are used for like elements in describing each drawing.

The present invention relates to a method and a user terminal for recommending an emoticon corresponding to contents of text input from a user to a user. Specifically, the present invention converts an input text into a text signature of a bit string having a specific value, compares the text signature with a plurality of emoticon signatures, and displays emoticons according to the similarity. Here, the emoticon signature is generated from at least one keyword assigned to the emoticon as a signature of a bit string having a specific value, such as a text signature.

The present invention compares the text signature of the bit string with the emoticon signature rather than comparing the text itself with the keyword of the emoticon, thereby providing a fast operation speed and reducing the capacity required for emoticone keyword storage. In addition, the present invention can efficiently recommend an emoticon corresponding to the contents of a text to a user even if there is an error in the text itself, for example, a space or an error.

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

1 is a view for explaining a user terminal recommending an emoticon according to an embodiment of the present invention.

1, a user terminal according to the present invention includes a hash value generation unit 101, a text signature generation unit 103, and a display unit 105.

The hash value generation unit 101 generates a hash value of text input from a user through an application such as a messenger using a hash function. The hash function allows you to have a unique hash value for each text.

More specifically, the hash value generation unit 101 may generate a hash value for each unit of text of the text segmented according to the N-gram. Here, N is a natural number, and can be variously set according to the embodiment. If the text entered by the user is "Perception is great" and N is 2, the unit characters are "perception", "angle", "is", "big", "big", " ". Thus, the hash value generation unit 101 generates a hash value for each of "perception", "angular", "is", "big", "big", "

The text signature generation unit 103 generates a text signature K (where K is a natural number) bit string, that is, a bit string in the form of K bits, using the hash value generated from the hash value generation unit 101, . K can be variously set according to the embodiment. The text signature generation unit 103 may generate a text signature by converting the hash value into binary numbers.

Alternatively, the text signature generation unit 103 may generate a text signature by dividing the hash value by K, determining the value corresponding to the remaining value as a digit of the K-bit string, and changing the value of the corresponding digit. For example, if the hash value of the "perception" is 1611021 and K is 4000, the remainder value mod of '1611021' divided by 4000 is '3021'. When the value of the K bit string is set to 0, the text signature generator 103 may generate a text signature by changing the 3021st bit value in the K bit string to '1'.

The text signature generator 103 converts the hash value of each unit character generated by the hash value generator 101 into a K bit string in accordance with a predetermined rule and ORs the converted K bit string to generate a text signature Lt; / RTI > Thus, a single text signature can be generated for the text "It's been a long time lapse ".

The display unit 105 displays at least one emoticon selected from the emoticon set according to the similarity between the text signature and the emoticon signature. The emoticon signature may be generated in the same manner as the text signature, and the emoticon signature may be generated using at least one keyword assigned to the emoticon. The emoticon signature may be generated at the user terminal or may be generated at the server and provided to the user terminal together with the emoticon. Although not shown, the user terminal according to the present invention may further include a storage unit for storing the emoticon signature.

Meanwhile, the user terminal according to the present invention may further include a similarity determination unit 107 for determining a similarity between the text signature and the emoticon signature stored in the user terminal. The similarity determination unit 107 compares the text signature with a plurality of emoticon signatures to determine whether the similarity is equal to or greater than a threshold value.

The similarity determination unit 107 may determine the degree of similarity based on the number of emoticon keywords corresponding to the text input from the user. For example, when there is an emoticon assigned with keywords such as " big day "and" tear " The determination unit 107 compares the text signature and the two emoticon signatures. The two digits where the "1" is located in the signature for the emoticon assigned with the keyword "big day" and "perception" are included in the place where "1" is located in the text signature for " However, only one of the two digits where "1" is located in the emoticon signature assigned with the keywords "big day" and "tear" corresponds to the place where "1 " Therefore, the similarity determination unit 107 can determine that the degree of similarity between the emoticon signature and the text signature for the emoticon assigned with the keywords "big day" and "late"

Alternatively, the similarity determination unit 107 may determine the similarity using the Jaccard coefficient, which is the most widely used association coefficient for comparing similarity and diversity. The Jaccard factor is defined as the size of the intersection of the sets divided by the union.

The similarity determination unit 107 can select emoticons for emoticon signatures that are equal to or larger than a threshold value in the emoticon set, and can select the emoticons in descending order of similarity in the emoticon set.

The display unit 105 displays the emoticons selected by the similarity determination unit 107, so that the emoticon corresponding to the text entered from the user can be recommended to the user. The display unit 105 displays an emoticon having a degree of similarity equal to or greater than a threshold value

That is, according to the present invention, one text signature is generated regardless of the type and length of the text input from the user, for example, a word or a sentence. Also, even though there are many keywords assigned to the emoticons, one emoticon signature per emoticon is generated. Therefore, according to the present invention, the similarity can be determined by comparing one text signature and one emoticon signature, thereby providing a fast computing speed. Also, even if a plurality of keywords are assigned to emoticons, only one emoticon signature is generated and a fast operation speed can be assured. Therefore, emoticons having similarity to the text content inputted from the user can be recommended by assigning a plurality of keywords to the emoticons .

Also, according to the present invention, by generating a text signature by dividing text by N-gram, it is possible to provide strong recommendation performance even for grammatical errors such as spacing and typhus.

Further, according to the present invention, it is possible to improve the satisfaction of the user by recommending various emoticons to the user in the order of high similarity.

FIG. 2 is a diagram for explaining a method of recommending emoticons according to an embodiment of the present invention. Referring to FIG.

In FIG. 2, an emoticon recommendation method performed in the user terminal as shown in FIG. 1 will be described as an embodiment. A character application provided basically or a separately downloaded messenger application may be stored in the user terminal. The above-mentioned application provides emoticons.

The user terminal uses the hash function to generate a hash value of the text input from the user (S201). At this time, the user terminal can generate a hash value for each unit of text of the divided text according to the N-gram, and N can be variously set according to the embodiment.

The user terminal generates a text signature of K (where K is a natural number) bit string using the hash value according to a predetermined rule (S203). The user terminal can generate a K-bit string for each of the hash values generated in step S201 and OR the generated K-bit string to generate a text signature.

The user terminal divides the hash value by K and generates a K-bit string for each hash value by dividing the hash value by K and setting the value of the position corresponding to the remaining value according to the division operation in the K- To generate a text signature.

The user terminal displays at least one emoticon selected from the emoticon set according to the similarity between the text signature and the emoticon signature (S205). That is, the user terminal displays an emoticon according to the degree of similarity, thereby recommending the emoticon corresponding to the text content input by the user to the user.

The user terminal can determine whether the similarity of the text signature and the emoticon signature is equal to or greater than the threshold and display only emoticons above the threshold. Alternatively, the user terminal can determine the similarity of the text signatures and emoticon signatures, and display the emoticons in descending order of similarity.

Meanwhile, the emoticon signature may be downloaded together from the server when the messenger application is downloaded, or separately from the server to the user terminal. Also, the emoticon recommendation function can be activated / deactivated according to the user's request, and the user terminal can display emoticons according to the similarity of the text signatures and the emoticon signatures according to the user's request.

3 is a diagram for explaining a method of recommending emoticons according to another embodiment of the present invention. More specifically, in Fig. 3, a method of generating an emoticon signature for emoticon recommendation is described as an embodiment. Emoticon signatures can be created in the same way as text signatures, but there may be some differences because a plurality of keywords are set in the emoticons.

Emoticon signature generation may be performed at the user terminal or at the server. The user terminal can generate the emoticon signature by receiving the emoticon data and the keyword assigned to the emoticon from the server. Alternatively, the keyword of the emoticon may be assigned by the user, and the user terminal may generate the emoticon signature using the emoticon keyword assigned by the user.

The server generates a hash value for at least one or more keywords assigned to the emoticon using the hash function (S301). A plurality of keywords may be assigned to the emoticon. The server can generate a hash value for each of the unit characters of the keyword segmented according to the N-gram.

The server generates an emoticon signature of K (where K is a natural number) bits using the hash value according to a predetermined rule (S303). The server may generate a K bit string for each of the hash values of the keyword and OR the generated K bit string to generate the emoticon signature. Therefore, one emoticon signature can be generated for a plurality of keywords.

The server stores the generated emoticon signature (S305). The server may transmit the emoticon data and the emoticon signature to the user terminal together with the request of the user terminal.

According to the present invention, since the emoticon signature of the bit string can be stored without storing the keyword of the emoticon for the emoticone recommendation service, the capacity for storing the emoticons keyword can be reduced. Also, even if the number of keywords assigned to the emoticon increases, the size of the entire bit string of the emoticon signature is not changed by the OR operation, so that the memory can be efficiently used.

4 and 5 are diagrams showing recommendation emoticons according to an embodiment of the present invention.

FIG. 4 shows a recommendation emoticons for the "perception is big" input from a user, and FIG. 5 shows a recommendation emoticons for "a complete smile". As shown, the recommended emoticons faithfully reflect the contents of the text input from the user.

In particular, in the case of FIG. 5, although the text is out of the grammar, an emoticon is recommended to the user. As described above, the present invention divides text into unit characters and compares the similarity with the emoticon signature to recommend the emoticon. Therefore, even if some grammatical errors are included in the text, the emoticon corresponding to the text content can be recommended.

FIG. 6 is a diagram for explaining an emoticon recommendation method according to a specific embodiment of the present invention.

The user terminal transforms each emoticon keyword into a bit stream for all emoticons in the system, generates an emoticon signature (S601), and stores it in the database (S603).

When text is input to the user terminal (S605), the user terminal converts the text into a bit string to generate a text signature (S607), and calculates the similarity between the text signature and the emoticon signature (S609).

If the threshold value for the similarity is 40%, the user terminal determines whether the similarity is 40% or more (S611). If the threshold is 40% or less, the user terminal outputs only the text (S613). When the degree of similarity is 40% or more, the user terminal outputs emoticon in descending order of similarity (S615). The user selects an emoticon (S617), and the user terminal simultaneously outputs the text and the selected emoticon (S619).

The above-described technical features may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions recorded on the medium may be those specially designed and constructed for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks, and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware device may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

As described above, the present invention has been described with reference to particular embodiments, such as specific elements, and specific embodiments and drawings. However, it should be understood that the present invention is not limited to the above- And various modifications and changes may be made thereto by those skilled in the art to which the present invention pertains. Accordingly, the spirit of the present invention should not be construed as being limited to the embodiments described, and all of the equivalents or equivalents of the claims, as well as the following claims, belong to the scope of the present invention .

Claims (14)

delete delete delete delete delete delete delete delete In a method for recommending a user terminal's emoticon,
Generating a hash value for at least one or more keywords assigned to the emoticon by a user using a hash function;
Generating an emoticon signature of K (where K is a natural number) bit string using the hash value according to a predetermined rule; And
Displaying at least one emoticon selected from the emoticon set according to the text signature of the text input from the user and the similarity of the emoticon signature,
The generating of the hash value may include generating the hash value for each unit character of the keyword segmented according to the N-gram,
The step of generating the emoticon signature
Performing a division operation for dividing the hash value by the K; And
And changing the value of the K bit string, which is a first value, to a second value,
Wherein the second value is a value obtained by changing a value of a digit corresponding to a remaining value according to the division operation from the first value
How to recommend emoticons.
10. The method of claim 9,
The step of displaying the emoticons
Determining whether the similarity of the text signature and the emoticon signature is equal to or greater than a threshold value; And
Displaying emoticons for emoticon signatures above the threshold;
The emoticon recommendation method.
10. The method of claim 9,
The step of displaying the emoticons
And displaying the emoticons in descending order of the degree of similarity
How to recommend emoticons.
10. The method of claim 9,
The step of displaying the emoticons
Activated according to the request of the user
How to recommend emoticons.
10. The method of claim 9,
Converting the hash value of the text according to the preset rule, and generating the text signature
The emoticon recommendation method further comprising:
delete

Images