KR20210043281A - Method of providing visual custom widgets through social media big data analysis and server performing the same - Google Patents

Abstract

Disclosed is a method for a management server to analyze social media data and provide an analysis result of the data to a user terminal. According to one of various embodiments of the present invention, the method for a management server to analyze social media data and provide an analysis result includes the following steps of: receiving a representative keyword set as a data analysis target from a user terminal; receiving selection information of a social media channel set as data analysis scope from the user terminal; receiving analysis period information set as a temporal criterion for data analysis from the user terminal; extracting content related with the representative keyword from data of the selected social media channel, based on the selection information of the social media channel; executing emotion classification with respect to each piece of the extracted content based on words included in each piece of the extracted content; and performing control such that a widget about a social media data analysis result is displayed on a website or application executed on the user terminal, based on the extracted content and the emotion classification information about each piece of the content. The emotion classification about each piece of the extracted content can be executed based on an emotion classification result of the words included in each piece of the content.

Description

Visual provision method of custom widget through social media big data analysis, and server performing it {METHOD OF PROVIDING VISUAL CUSTOM WIDGETS THROUGH SOCIAL MEDIA BIG DATA ANALYSIS AND SERVER PERFORMING THE SAME}

Various embodiments of the present invention relate to a method for visually providing a custom widget through social media big data analysis and a server that performs the same, and in detail, using big data in visually creating a widget containing data desired by a user. It's about how to do it.

With the development of social media on the Internet, the influence of social media is becoming more and more huge, and accordingly, attempts to analyze trends on social media are becoming more frequent.

In general, large corporations ask consulting companies to perform trend analysis on social media, and consulting companies perform data analysis on their own and generate and provide massive reports through human resources.

However, there was a point that SMEs, including start-ups, lacked funds to entrust a trend analysis request to a consulting company, and due to the characteristics of a shorter pivot cycle of a product or service being sold, it is necessary to perform trend analysis quickly, briefly, and somewhat inexpensively. Existed.

Accordingly, a number of services that perform a brief trend analysis on a specific keyword in real time on the Internet and provide the result have begun to emerge. These services provide functions such as analysis of the amount of mention of specific keywords and related keywords, and emotion analysis of related keywords.

However, there was also a lack of accuracy in terms of accuracy, such as always classifying specific words as specific emotions in relation to the analysis of emotions of related keywords for specific keywords by the user. For example, certain words may be interpreted as positive or negative depending on the situation, but such complexity could not be specifically resolved through big data.

Accordingly, the emotion of words is determined according to the situation through big data or user settings, and the analysis accuracy for the keywords set by the user is improved accordingly, and the visual image of the analyzed data is sent to the user through a custom widget. It was necessary to develop a method to provide.

Korean Patent Publication No. 10-2019-0023722

Various embodiments of the present invention aim to provide more various methods and increase the accuracy of data analysis in providing a custom widget to a user through social media big data analysis.

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

According to one of the various embodiments of the present invention for solving the above-described problem, the method of analyzing the social media data by the management server and providing the analysis result of the data to the user terminal is a representative set as a data analysis target from the user terminal. Receiving a keyword; Receiving selection information of a social media channel set as a data analysis range from a user terminal; Receiving analysis period information set as a temporal criterion for data analysis from a user terminal; Extracting content related to a representative keyword from the data of the selected social media channel based on the selection information of the social media channel; Performing emotion classification for each of the extracted contents based on words included in each of the extracted contents; And controlling the widget for the analysis result of social media data to be displayed on the website or application executed on the user terminal, based on the extracted content and emotion classification information for each content. The emotion classification for each content may be performed based on a result of emotion classification of words included in each content.

The emotion classification for each of the content extracted in relation to the representative keyword may be characterized in that the management server receives the emotion classification standard information for each word from the user terminal, and performs additionally based on the received emotion classification standard information for each word. have.

In the method of analyzing social media data of the management server and providing analysis results, the user receives emotion classification standard information for each word for a specific representative keyword from a user terminal, and then receives emotion classification standard information for each word of other users for the corresponding representative keyword. It may further include transmitting to the terminal.

In the process of performing emotion classification for each extracted content related to a representative keyword based on the emotion classification result of words included in each content, the emotion classification result for a specific word is other words in the content containing the corresponding word It can be determined based on the results of classifying their emotions.

The emotion classification for each of the extracted contents may be additionally performed based on a result of emotion classification for reaction information input by users for each content.

The method for analyzing social media data by the management server and providing analysis results includes calculating a numerical relationship between each of the words included in the extracted content and a representative keyword, and words included in the extracted content based on the calculated association. It may further include the step of performing the emotion classification for the.

The social media data analysis and analysis result providing method of the management server includes: generating positive/negative mention amount trend information related to the representative keyword on the basis of the emotion classification result of the content extracted in relation to the representative keyword; Analyzing the positive/negative mention amount trend information to detect whether or not a change amount greater than a preset reference occurs in the positive/negative mention amount trend; And extracting words related to the occurrence cause based on the frequency of occurrence of words included in the extracted content in relation to the representative keyword and the result of emotional classification of the words when a change amount greater than a preset criterion occurs in the positive/negative mention trend. It may further include a step.

The step of performing emotion classification for each of the extracted contents based on the words included in each of the extracted contents includes determining an industry group to which the representative keyword belongs, and for each of the extracted contents based on a criterion set according to the determined industry group. It may be characterized by performing emotion classification.

In the step of performing emotion classification for each of the extracted contents based on the criteria set according to the determined industry group, the emotion classification of words is performed based on the emotion classification criteria for each word set according to the industry group, and the extracted contents based on the result It may be to perform emotion classification for each.

According to another embodiment of the present invention, setting information for receiving a representative keyword set as a data analysis target from a user terminal, selection information of a social media channel set as a data analysis range, and analysis period information set as a temporal criterion for data analysis Collection unit; A content extraction unit for extracting content related to a representative keyword from the data of the selected social media channel based on the selection information of the social media channel; An emotion classification performing unit for performing emotion classification on each extracted content based on words included in each of the extracted contents; And a widget manager configured to control a widget for a social media data analysis result to be displayed on a website or application executed on a user terminal, based on the extracted content and emotion classification information for each content, wherein the emotion classification execution unit The process of performing emotion classification for each of the contents extracted from is performed based on a result of emotion classification of words included in each of the contents, and a management server may be initiated.

According to an embodiment of the present invention, by analyzing the extracted social media content and words included therein according to various criteria, it is possible to perform customized analysis and improve the accuracy of analysis.

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

1 is a conceptual diagram schematically showing the overall configuration of a social media big data analysis platform according to an embodiment of the present invention.
2 to 7 are diagrams illustrating screens displayed on a user terminal according to an embodiment of the present invention.
8 is a block diagram schematically showing the configuration of a management server according to an embodiment of the present invention.
9 is a flowchart illustrating a process in which a management server provides content analysis data according to a user's request in the form of a widget according to an embodiment of the present invention.

The terms used in the present specification are for describing exemplary embodiments and are not intended to limit the present invention. In this specification, the singular form also includes the plural form unless specifically stated in the phrase. As used herein, “comprises” and/or “comprising” do not exclude the presence or addition of one or more other elements other than the mentioned elements. Throughout the specification, the same reference numerals refer to the same elements, and "and/or" includes each and all combinations of one or more of the mentioned elements. Although "first", "second", and the like are used to describe various elements, it goes without saying that these elements are not limited by these terms. These terms are only used to distinguish one component from another component. Therefore, it goes without saying that the first component mentioned below may be the second component within the technical idea of the present invention.

When a part of the specification is said to "include" a certain component, it means that other components may be further included rather than excluding other components unless specifically stated to the contrary. In addition, terms such as "... unit" and "module" described in the specification mean a unit that processes at least one function or operation, which may be implemented as hardware or software or a combination of hardware and software. .

Hereinafter, a description of various embodiments of the present invention will proceed with reference to the drawings.

1 is a conceptual diagram schematically showing the overall configuration of a social media big data analysis platform 10 according to an embodiment of the present invention.

Referring to FIG. 1, the social media big data analysis platform 10 may include a management server 100, a user terminal 200, and an external server 300.

The management server 100 may collect contents from various types of social media channels, receive various setting information including a representative keyword from the user terminal 200, and extract only contents related to the representative keyword among the collected contents. I can.

The management server 100 extracts content related to the representative keyword from the data provided by the social media channel, based on the representative keyword set by the user, and performs data analysis based on the information included in the extracted content. A data analysis result may be provided, and the data analysis result may be displayed in the form of various widgets on the user terminal 200.

The management server 100 is a big data based on machine learning in a process of extracting content related to a representative keyword from data provided by a social media channel and performing data analysis based on the extracted content. Analysis can be performed. The management server 100 may continuously and effectively change the algorithm used for content extraction and data analysis through learning, and accordingly, the data analysis performance or efficiency of the management server 100 may increase.

The user terminal 200 may access the management server 100 to transmit a desired analysis range to the management server 100 and receive an analysis result. The user may use the user terminal 200 to input a representative keyword to be analyzed, selection information of a social media channel set as a data analysis range, and analysis period information to be transmitted to the management server 100.

The user terminal 200 may include a memory storing information necessary for an operation, a central processing unit such as a CPU that performs various operations required for an operation, an input/output device, and the like.

The user terminal 200 includes all kinds of handheld-based wireless devices that can be connected to a web server through a network, such as a mobile phone, a smart phone, a personal digital assistant (PDA), a portable multimedia player (PMP), and a tablet PC. It may include a communication device, and is one of digital devices having computing power by mounting a microprocessor and having a memory means such as a personal computer (eg, a desktop computer, a notebook computer, etc.), a workstation, a PDA, a web pad, etc. May be.

Although not shown in FIG. 1, an administrator terminal (not shown) that is separated from the user terminal 200 may communicate with the management server 100 to change various settings of the management server 100.

The external server 300 is a server that exists separately from the management server 100, and according to an embodiment, the management server 100 may collect data provided by various social media channels from the external server 300. That is, the external server 300 may be a concept including a plurality of servers managed by various subjects, and may be servers related to various communities, portals, and SNS.

The management server 100 can collect news articles, SNS posts, community posts, and blog posts through the external server 300. For example, SNS posts can be used in various ways such as Twitter™, Instagram™, and Facebook™. It can be classified as a kind of SNS post.

According to an embodiment of the present invention, a communication network used by each subject operating in the social media big data analysis platform 10 to communicate may be configured regardless of its communication mode, such as wired and wireless, for example. For example, it may be implemented in various communication networks such as a local area network (LAN), a metropolitan area network (MAN), and a wide area network (WAN). Preferably, the communication network according to an embodiment of the present invention may be a known World Wide Web (WWW) or the like.

2 to 7 are diagrams showing screens displayed on the user terminal 200 according to an embodiment of the present invention.

The user terminal 200 may access the management server 100 through a specific website or a specific application to receive a data analysis result of social media, and the received data analysis result is a widget through the display of the user terminal 200. It can be displayed in a form.

Referring to FIG. 2, a user may create a dashboard for a subject he or she desires, and add widgets desired to be displayed in the dashboard. In this specification, the term'dashboard' may mean a user interface function that integrates and manages various information on a web site or application, and enables users to navigate, and'widget' refers to data analysis results in various diagrams and diagrams. , May mean a unit screen displayed by numerical values, etc. In the present invention, various types and a plurality of widgets may be included in one dashboard, and a user may determine which widgets are to be included and in what manner.

Referring to FIG. 2, among various widgets that can be displayed through data analysis on a specific representative keyword,'total review, issuer ranking, mention trend, hashtag ranking, campaign performance analysis, comment summary, total mention, positive The status of'Total Negative References, Trend of Positive Negative References, and Positive Negative Keyword Ranking' is shown.

3 illustrates an example in which widgets corresponding to'general review' and'issuer ranking' are displayed on the user terminal 200.

Referring to FIG. 3, a general review for a specific representative keyword is shown, and the amount of reference of the keyword, the keyword mentioned together, the hashtag used together, the positive/negative related keywords, etc. may be displayed on the'general review' related widget. .

Looking at the'issuer ranking' widget screen of FIG. 3, keywords determined to be related to the representative keyword set by the user may be ranked and displayed in the order of the amount of mention.

FIG. 4 illustrates an example in which widgets corresponding to'referred amount trend' and'hash tag ranking' are displayed on the user terminal 200.

Looking at the widget screen of the'reference amount trend' of FIG. 4, the number of times that the representative keyword set by the user is mentioned in the social media channels selected by the user may be displayed by date. Such a display date may fluctuate according to the analysis period set by the user.

Looking at the'hash tag ranking' widget screen of FIG. 4, the hashtags used in the contents of the social media channel in which the representative keyword set by the user is mentioned are displayed. In FIG. 4, a widget is configured in the form of a diagram whose size changes according to the frequency of use of a hashtag.

As shown in Fig. 4, the widgets'referred amount trend' and'hash tag ranking' are visually displayed through graphs and diagrams of different types, respectively, each widget is analyzed through the management server 100 on the user terminal 200. The method of displaying the data may be set differently according to the user's selection. For example, the user can change the display method of a specific widget by selecting one of various methods such as a bar graph, a line graph, a bubble chart, a diagram, a table, and a word cloud.

5 illustrates an example in which widgets corresponding to'campaign performance analysis' and'reference summary' are displayed on the user terminal 200.

Looking at the'Campaign Performance Analysis' widget screen of FIG. 5, when a campaign is progressed for a specific period for a representative keyword set by a user, a data analysis result on the campaign performance may be displayed.

Looking at the widget screen of the'reference summary' of FIG. 5, some of the content of the social media channel including the representative keyword set by the user may be displayed. According to an embodiment, in the display of the comment summary, whether the content is extracted from a social media channel and a result of emotional classification of the corresponding content may be displayed. The emotion classification result according to an embodiment may be determined as one of three classification items of positive, negative, and neutral.

6 illustrates an example in which widgets corresponding to'total mention amount' and'positive negative total mention amount' are displayed on the user terminal 200.

Looking at the'total mention amount' widget screen of FIG. 6, the number of contents referring to the representative keyword set by the user may be classified and displayed for each social media channel.

Looking at the'positive and negative total mention amount' widget screen of FIG. 6, the number of contents mentioning the representative keyword set by the user can be classified and displayed by social media channels, and the number of contents classified by each social media channel May be classified and displayed based on the emotion classification result.

7 illustrates an example in which a widget corresponding to a'positive and negative mention amount trend' is displayed on the user terminal 200.

Looking at the'positive negative mention amount trend' widget screen of FIG. 7, the trend of the number of contents of the social media channel classified based on the emotion classification result may be displayed in the form of a line graph.

The user may set the display format or analysis period of the widgets displayed in FIGS. 3 to 7, and may select at least some of various social media channels to set the analysis range. Although not shown through FIGS. 3 to 7, a'ad manager' widget that displays advertisement campaign statistics of a specific SNS through various indicators may be additionally utilized.

8 is a block diagram schematically showing the configuration of the management server 100 according to an embodiment of the present invention.

Referring to FIG. 8, the management server 100 includes a setting information collection unit 110, a content extraction unit 120, an emotion classification performance unit 130, a data analysis unit 140, a widget management unit 150, and a communication unit ( 160), the storage unit 170, and the controller 180 may be included.

The setting information collection unit 110 may receive a representative keyword set as a target for data analysis from the user terminal 200, and set as a temporal criterion for data analysis and selection information of a social media channel set as a data analysis range. Analysis period information may be additionally received.

The user may select at least some of the plurality of social media channel lists so that only the contents disclosed through the social media selected by the user are analyzed. For example, a user may set to perform data analysis on a specific representative keyword only for one SNS channel he uses. In addition, the user can set the analysis period information so that the analysis is performed only within the time period desired by the user.

The content extraction unit 120 may extract content related to a representative keyword from data provided by the selected social media channel based on selection information of a social media channel set by a user. In addition, the content extraction unit 120 may extract only the uploaded content within the analysis period set by the user.

The emotion classification performing unit 130 may perform emotion classification on each of the extracted contents based on words included in each of the contents extracted by the contents extracting unit 120. The emotion classification may refer to an operation of classifying emotions inherent in a specific content based on the type of words included in the corresponding content, the frequency of occurrence of words, and the like. Emotion classification may be classified into various types, such as fear, anger, sadness, joy, and dissatisfaction, for example, and may be simply classified as positive, normal, and negative.

The emotion classification performing unit 130 may be classified into detailed configurations of a content classification unit 131, a word classification unit 132, and a trend analysis unit 133.

The content classification unit 131 performs emotion classification for each content in units of content extracted by the content extraction unit 120, and this may be performed based on the classification result of the word classification unit 132.

According to an embodiment of the present invention, the content classification unit 131 additionally receives emotion classification criterion information for each word from the user terminal 200 in performing emotion classification for each extracted content related to a specific representative keyword. And, based on the received emotion classification criterion information for each word, emotion classification may be performed.

That is, the user may determine what emotions a specific word should be classified for a specific representative keyword inputted by the user through his/her user terminal 200.

In addition, the content classification unit 131 receives emotion classification criterion information for each word for a specific representative keyword from the user terminal 200, and then stores emotion classification criterion information of other users with respect to the representative keywords in the user terminal 200. Can be delivered to. That is, when a user inputs a word for which he/she wants to set the emotion classification criterion for a specific representative keyword on a website or application executed on the user terminal 200, other users set the word for the same representative keyword in some way. You will be able to check if there is.

According to an embodiment of the present invention, the content classification unit 131 may perform emotion classification for each content based on an additional emotion classification result for reaction information input by users with respect to the contents to be analyzed. .

The response information input by users may mean various kinds of feedback on specific content, and this may include recommendations, dislikes, likes, dislikes, comments, related articles, hashtags, and the like.

The content classification unit 131 according to an exemplary embodiment is an industry group to which the representative keyword belongs in performing emotion classification for each extracted content based on words included in each of the extracted content in relation to the representative keyword set by the user. May be determined and sentiment classification for each extracted content may be performed based on a criterion set according to the determined industry group.

The word classifier 132 may classify emotions for each word included in specific content.

The word classification unit 132 may classify emotions into various types such as fear, anger, sadness, joy, dissatisfaction, etc., in performing emotion classification for each word, and simply into positive, negative, and normal types. You can also classify your emotions.

According to an embodiment, the word classifier 132 may perform classification by numerically quantifying a probability or score that a corresponding word belongs to a specific emotion, rather than simply classifying emotions. For example, the word classification unit 132 calculates the probability that a specific word corresponds to'positive' by 80%, the probability that corresponds to'normal' by 15%, and the probability that to'negative' is 5%, It can be done with'positive'. In contrast, the word classification unit 132 sets a score for a specific word such as 80 points for'positive', 15 points for'normal', and 5 points for'negative', and performs the final classification. May be. Such probability information or score information may be used as basic data when the content classification unit 131 performs emotion classification on content including a plurality of words.

In performing emotion classification for words, the word classifier 132 according to an embodiment of the present invention determines the emotion classification for a specific word based on the emotion classification result of other words in the content including the word. I can. That is, a specific word may be affected by the result of emotional classification of other words in the content containing the word. As such, the result of emotional classification between words may affect each other. The result of classification of emotions may be more accurate.

The word classifier 132 according to an embodiment of the present invention calculates and calculates the relationship between each of the words included in the specific content to be analyzed and a representative keyword, and based on the calculated association, Emotion classification for words can be performed.

For example, if a specific news article extracted for a representative keyword set by a user is subject to emotion classification, words in the same sentence as the representative keyword in the relevant news article are calculated with high relevance, and keywords that are opposed to the representative keyword Words in a sentence such as, can be calculated to have low relevance, and can also calculate the relevance through the distance between a specific word and a representative keyword within the sentence.

The word classification unit 132 according to an embodiment, in the process of determining an industry group to which a representative keyword belongs and performing emotion classification for each content based on a criterion according to the determined industry group, the content classification unit 131 The emotion classification of words may be performed based on the emotion classification criteria for each word set accordingly. In this case, the content classification unit 131 may perform the emotion classification of the content based on the emotion classification result of the words classified by the word classification unit 132 according to the criteria according to the industry group.

As described above, since there is an emotion classification criterion for each word set according to the industry group, even if the word is the same, the emotion classification result for the corresponding word may appear differently depending on which industry group the representative keyword is classified into.

The trend analysis unit 133 may generate positive/negative mention amount trend information based on a result of emotional classification of content extracted in connection with a specific representative keyword. This may be generated in the same manner as the example of the'positive negative mention amount trend' widget screen of FIG. 7.

The trend analysis unit 133 according to an embodiment classifies the frequency of occurrence of words included in the extracted contents related to the representative keyword and the emotions of the corresponding words when a change amount greater than a preset reference occurs in the positive/negative mention amount trend. Based on the results, words related to the cause of the change can be extracted.

For example, when the ratio of positive/negative mentions in relation to a specific keyword changes by 20% or more within 3 days, the trend analysis unit 133 detects such a change and causes such a change due to a certain factor. You can analyze whether or not. The trend analysis unit 133 is based on the date when the ratio of the positive/negative mention amount changes and the change ratio by date, the frequency of appearance of words included in the contents corresponding to the corresponding dates, and the result of emotional classification of the words. , Words that have influenced the positive/negative mention ratio can be extracted as the cause of occurrence.

The data analysis unit 140 may perform a data analysis task necessary to display various widgets shown through FIGS. 2 to 7. The data analysis unit 140 may utilize the classification result of the above-described emotion classification execution unit 130, and'total review, issuer ranking, reference amount trend, hashtag ranking, campaign performance analysis, comment summary, total reference amount , A data analysis for generating a'positive and negative total mention amount, a positive and negative mention amount trend, and a positive and negative keyword ranking' widget may be performed.

The data analysis unit 140 may utilize big data and may perform data analysis through machine learning or deep learning.

The widget management unit 150 may support a function of displaying result data derived from the data analysis unit 140 in the form of various widgets on the user terminal 200. The widget management unit 150 receives display format information for each widget from the user terminal 200 and creates a custom widget in various visual methods such as a bar graph, a line graph, a bubble chart, a diagram, a table, and a word cloud. can do.

The communication unit 160 enables the management server 100 to communicate with the user terminal 200 and the external server 300. The communication network used by the communication unit 160 to perform communication may be configured regardless of its communication mode such as wired or wireless, for example, a local area network (LAN) or a metropolitan area network (MAN: Metropolitan). Area Network) and a wide area network (WAN).

The storage unit 170 serves to store information collected, generated, and processed in various configuration units of the management server 100. That is, the storage unit 170 may store setting information input by a user, information collected from various social media channels, analysis information of social media contents, and the like. The storage unit 170 may include, for example, a memory, a cache, a buffer, and the like, and may include software, firmware, hardware, or a combination of at least two or more of them.

The control unit 180 includes a setting information collection unit 110, a content extraction unit 120, an emotion classification performing unit 130, a data analysis unit 140, a widget management unit 150, a communication unit 160, and a storage unit 170. ) Can perform the function of controlling the data flow between. That is, the control unit 180 according to an embodiment of the present invention includes a setting information collection unit 110, a content extraction unit 120, an emotion classification performing unit 130, a data analysis unit 140, and a widget management unit 150. , Each of the communication unit 160 and the storage unit 170 may be controlled to perform a unique function.

In FIG. 8, the setting information collection unit 110, the content extraction unit 120, the emotion classification performing unit 130, the data analysis unit 140, the widget management unit 150, and the communication unit 160 function as the control unit 180. Since the configuration is classified as, it may be integrated and configured as one controller 180.

9 is a flowchart illustrating a process in which the management server 100 provides content analysis data according to a user's request in the form of a widget according to an embodiment of the present invention.

Referring to FIG. 9, the management server 100 may receive a representative keyword to be analyzed, selection information of a social media channel, and analysis period information from the user terminal 200 (S901).

The management server 100 may then extract only contents related to the set representative keyword from within the data provided by the social media channel selected by the user (S903).

The management server 100 may receive emotion classification criterion information for each word from the user terminal 200 before performing emotion classification on the extracted content and words included in the corresponding content (S905).

Based on the reference information received in step S905, the management server 100 may perform emotion classification on words included in the contents extracted in step S903 (S907), and based on the emotion classification result for the words. Emotion classification for each of the extracted contents may be performed (S909).

The management server 100 may finally perform data analysis based on the emotion classification information of the contents, transmit the data analysis information to the user terminal 200 (S911), and receive it on the display unit of the user terminal 200. Various widgets may be displayed based on one data analysis information.

As described above, according to various embodiments of the present invention, when the management server analyzes social media data and provides the analysis result, more accurate data analysis may be possible by performing emotion classification on the content itself or words in various ways. have.

The steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or a combination thereof. Software modules include Random Access Memory (RAM), Read Only Memory (ROM), Erasable Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), Flash Memory, hard disk, removable disk, CD-ROM, or It may reside on any type of computer-readable recording medium well known in the art to which the present invention pertains.

In the above, embodiments of the present invention have been described with reference to the accompanying drawings, but those skilled in the art to which the present invention pertains can be implemented in other specific forms without changing the technical spirit or essential features. You will be able to understand. Therefore, the embodiments described above are illustrative in all respects, and should be understood as non-limiting.

10: Social media big data analysis platform
100: management server
200: user terminal
300: external server

Claims (10)

In a method for a management server to analyze social media data and provide the analysis result of the data to a user terminal,
Receiving a representative keyword set as a data analysis target from a user terminal;
Receiving selection information of a social media channel set as a data analysis range from a user terminal;
Receiving analysis period information set as a temporal criterion for data analysis from a user terminal;
Extracting content related to a representative keyword from the data of the selected social media channel based on the selection information of the social media channel;
Performing emotion classification for each of the extracted contents based on words included in each of the extracted contents; And
Based on the extracted content and emotion classification information for each content, controlling a widget for a social media data analysis result to be displayed on a website or application executed on a user terminal,
The emotion classification for each of the extracted contents is performed based on the emotion classification results of words included in each of the contents.
The method of claim 1,
The sentiment classification for each of the extracted contents related to the representative keyword,
A method for analyzing social media data and providing analysis results of a management server, characterized in that the management server receives emotion classification criteria information for each word from a user terminal, and performs additionally based on the received emotion classification criteria information for each word.
The method of claim 2,
After receiving the emotion classification criterion information for each word for a specific representative keyword from the user terminal,
A method of analyzing social media data and providing analysis results of a management server, further comprising the step of transmitting, to a user terminal, emotion classification criterion information for each word of other users for the corresponding representative keyword.
The method of claim 1,
In performing emotion classification for each extracted content related to a representative keyword based on the emotion classification result of words included in each content,
A method for analyzing and providing analysis result of social media data of a management server, characterized in that the result of the emotion classification for a specific word is determined based on the result of the emotion classification of other words in the content including the corresponding word.
The method of claim 1,
The emotion classification for each of the extracted contents is,
A method of analyzing social media data and providing analysis results of a management server, characterized in that it is additionally performed based on an emotion classification result for reaction information input by users for each content.
The method of claim 1,
Calculate the relationship between each of the words included in the extracted content and the representative keyword numerically,
A method of analyzing social media data and providing analysis results of a management server, further comprising the step of performing emotion classification on words included in the extracted content based on the calculated association.
The method of claim 1,
Generating positive/negative mention amount trend information related to the representative keyword based on a result of emotional classification of content extracted in connection with the representative keyword;
Analyzing the positive/negative mention amount trend information to detect whether or not a change amount greater than a preset reference occurs in the positive/negative mention amount trend; And
Extracting a word related to the cause of occurrence, based on the frequency of appearance of words included in the extracted content in relation to the representative keyword and the result of emotional classification of the words, when a change amount greater than a preset criterion occurs in the positive/negative mention trend A method of analyzing social media data and providing analysis results of the management server further comprising a.
The method of claim 1,
The step of performing emotion classification for each of the extracted contents based on words included in each of the extracted contents,
A method for analyzing social media data and providing analysis results of a management server, characterized in that the industry group to which the representative keyword belongs is determined, and emotion classification for each extracted content is performed based on a criterion set according to the determined industry group.
The method of claim 8,
The step of performing emotion classification for each of the extracted contents based on the criteria set according to the determined industry group,
A method for analyzing social media data and providing analysis results of a management server, in which the emotion classification of words is performed based on the emotion classification criteria for each word set according to the industry group and emotion classification for each extracted content based on the result .
In a management server that analyzes social media data and provides the analysis result of the data to a user terminal,
A setting information collection unit configured to receive a representative keyword set as a data analysis target from a user terminal, selection information of a social media channel set as a data analysis range, and analysis period information set as a temporal reference for data analysis;
A content extraction unit for extracting content related to a representative keyword from the data of the selected social media channel based on the selection information of the social media channel;
An emotion classification performing unit for performing emotion classification on each extracted content based on words included in each of the extracted contents; And
Based on the extracted content and emotion classification information for each content, it includes a widget management unit that controls to display a widget for a social media data analysis result on a website or application executed on the user terminal,
The process of performing the emotion classification for each of the contents extracted by the emotion classification performing unit is performed based on a result of emotion classification of words included in each of the contents.

Images