KR101602898B1 - Data visualization method and system using comment data for objects - Google Patents

Abstract

The present invention relates to a technology about a visualization method using expression elements collected from comment data of an object and a system thereof. For example, if an object is one content, the present invention relates to a technology for visualizing expression elements shown on a comment in which a consumer, who consumes the content, expresses emotions or opinions about the object. The present invention visually realizes expression elements in which emotions or opinions of a user are expressed wherein the emotions or the opinions are shown about an object in a user comment. The present invention analyzes objective information such as a manufacture company, a price, and the like that existing object information provides, and expression elements which express emotions or opinions that a user has while using the object. in which emotions or opinions are expressed wherein a user feels the emotions or the opinions while the user uses an object. Therefore, the present invention can provide information, which can be a reference of selecting an object, to a user who wants to newly use the object.

Description

TECHNICAL FIELD [0001] The present invention relates to a data visualization method and system using comment data of an object,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data visualization method and system using comment data of an object, and more particularly, to a technique of visualizing elements expressing feelings or opinions of a user or a consumer.

The present invention was derived from a research carried out by the Ministry of Education and the Korea Research Foundation as part of the Basic Research Project of the Humanities Society [Project Number: S-2013-A0403-00010, Title: Visualization of the system].

In general, users who consume content, such as movies, music, or literary works, or who use a product or service, use the content, product, or service ("object" And users who have not yet used the object or who want to get information about the object get the information by referring to the comments left by the users who have already used the object.

The user wants to obtain information about the object. Since the comment data on the object is composed based on the text, there is a problem that the time required for the user to obtain information about the object with reference to the comment data is considerably long. Particularly, when the amount of comment data for an object is large, a large number of users leave a comment, or when comment data is accumulated for a long time, a considerable effort is required even if the user reads the contents of the comment data.

Therefore, in order to cope with such a problem, research has been conducted on techniques for searching for comments or objects based on vocabularies on comment data, and for allowing users to shorten the search time for comments and objects.

An example of a method of retrieving contents using comment information on such contents is described in Korean Patent No. 10-0917784 entitled " Method and system for collective sensibility information retrieval based on comments on contents ".

The prior art described above collects comments attached to various contents on the Internet, creates a search database (hereinafter referred to as a DB), and provides a search method that displays objective and reliable ranking results for emotional queries using the search DB And a system. Especially, it is a technique to adjust the recommendation priority of an object by reflecting the frequency of emotional words appearing on the comments with respect to queries containing emotional words.

However, although the above prior art refers to a technique for retrieving emotional words from comments on an object, it does not mean that the overall emotion or opinion expected from the user for the object is effectively displayed, And there are limitations such as being recommended.

This is due to the fact that the above prior art adopts a text-based emotional word search, and it is required to develop a technique that can effectively show the overall emotion or opinion expected of a single object (content, product, or service).

Korean Patent No. 10-0917784 (registered on September 10, 2009)

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems of the conventional art, and it is an object of the present invention to visualize elements expressing emotions or opinions of a user appearing in existing user comments on an object (contents, goods, or services) The object of the present invention is to provide information that can be used as a criterion of object selection to a user who wants to newly use an object by analyzing emotions or opinions that a user uses and expresses not only objective information such as a manufacturer, .

The present invention relates to a method and system for intuitively visualizing the overall distribution of emotions or opinions expressed on an object by visualizing a plurality of elements expressing emotion or opinion on one object on the basis of a semantic distance And to provide the above objects.

Although the present invention can visualize emotion or opinion that is typically represented for one object, it is possible to visualize a plurality of expression elements on the basis of a relative meaning distance, so that the relative expression of a plurality of expression elements And a means for intuitively recognizing the distance and the distribution.

It is another object of the present invention to provide a means for visualizing a distribution based on a semantic distance by reflecting not only text but also various non-verbal elements expressing emotions or opinions such as emoticons and icons. In addition, since it can be freed from the restriction of text, it can provide a means of visualizing in a frame all the opinions or emotions expressed in various foreign languages.

In addition, the present invention can calculate the frequency of expression elements in which each emotion or opinion is expressed from comment data that can be obtained through various paths, for example, collected from a web site, It is intended to provide a graph.

According to an aspect of the present invention, there is provided a method of visualizing an expression element, the method comprising: extracting a plurality of expression elements from comment data collected for an object selected by a user; And visualizing the extracted plurality of presentation elements based on a distribution based on the semantic distance between the plurality of presentation elements.

The method of claim 1, further comprising measuring an extracted frequency in the comment data of the extracted expression elements, wherein visualizing the expression elements visualizes the extracted expression elements according to the frequency of the measured expression elements .

The step of extracting the expression elements may further include comparing the extracted expression elements with previously extracted expression elements and checking whether a new expression element is added to the extracted expression elements .

Wherein the step of visualizing the expression elements comprises the steps of: when a new expression element among the extracted expression elements is added, the step of visualizing one or more adjacent expressions having a semantic distance with the new expression element among the previously extracted expression elements, Determining a semantic location of the new presentation element based on the determined semantic distance from the one or more neighbor presentation elements.

The method may further include the step of determining the validity of the extracted expression element after extracting the expression elements, and if the extracted expression element is invalid for the object selected by the user, The method may further include the step of measuring the frequency extracted in the comment data of the extracted expression elements after extracting the expression elements, wherein the step of determining the validity of the extracted expression elements The validity of the extracted expression element may be determined by reflecting the measured frequency of the extracted expression element.

Identifying a frequency at which the extracted presentation element is extracted from an object other than the object selected by the user after the step of extracting the presentation elements, The method may further include the step of determining whether or not a predetermined number or more of objects have been extracted by a certain frequency or more. The method may further include the steps of: measuring the extracted frequency of the expression elements after extracting the expression elements; Wherein the step of visualizing the expression elements further comprises the step of visualizing the expression elements by reflecting the adjusted frequency of occurrence .

The method may further include the steps of: measuring the frequency at which the expression elements are extracted from the comment data after extracting the expression elements; comparing the frequency with which the expression elements appear in the object selected by the user, And adjusting the measured frequency by assigning a weight to the frequency at which the presentation element is extracted according to a comparison result between the frequency at which the presentation element appears in the object selected by the user and the measured frequency, have.

The step of extracting the expression elements may include the steps of searching whether or not the extracted expression elements are stored in a database in which standardized expression elements are stored in advance, and if the extracted expression elements are not stored in the database, Identifying a standardized expression element on the database closest to the expression element as a representative expression element of the extracted expression element, wherein the step of measuring the frequency includes: Wherein the step of adding the frequency extracted from the identified representative expression element to the frequency extracted in the comment data and visualizing the expression elements visualizes the representative expression element by reflecting the added frequency .

In addition, the step of visualizing the expression elements visualizes the expression elements using a multi-dimensional scaling map (MDS map) including the expression elements as a background.

A system for visualizing an expression element according to an exemplary embodiment of the present invention includes a storage element for storing comment data on an object selected by a user, an expression element extractor for extracting a plurality of expression elements from the stored comment data, And a visualization unit for visualizing the extracted plurality of presentation elements based on the distribution based on the semantic distance between the presentation elements.

According to the present invention, users who have already used an object can confirm intuitively the expression elements that the user feels before using the object by checking the expression elements of the object through the visualization graph, From the viewpoint of the user, there is an effect that people feel about the object, or the user who selects the object can easily select the desired object.

Also, since the visualization graph generated through the present invention is provided as a script program on a web site, it can be simultaneously provided to a large number of users.

In addition, since the present invention can be provided through a web page on a browser without installing a separate program, a developer can receive analysis results in real time without a new data management or distribution procedure every time the comment data is updated.

In addition, the present invention can intuitively confirm the reaction of public opinion on a policy when a government or a public agency announces a policy or a plan and people express their intention through the Internet on this.

In addition, it is possible to collect changes in response to public opinion by collecting public opinions about the incidents that occur in the Internet or companies that are generated from the outside, and it is possible to grasp the changes in public opinion in real time. It may respond.

In addition, by visualizing a plurality of elements expressing an emotion or opinion with respect to one object on the basis of a semantic distance, it is possible to intuitively provide a user with an overall distribution of emotions or opinions expressed to the object.

It is also possible to visualize emotions or opinions representative of one object. However, by visualizing a plurality of expression elements on the basis of relative meaning distances, the relative distances and distributions between a plurality of expression elements represented on the object can be intuitively .

In addition, it is possible to visualize the distribution based on the meaning distance by reflecting not only text but also various non-verbal elements capable of expressing emotions or opinions such as emoticons and icons, and can freely deviate from the constraint of text, It can also visualize within a single frame, including expressed opinions or emotions.

FIG. 1 is a diagram illustrating emotional vocabulary selected for producing an emotional lexical distribution map according to an embodiment of the present invention.
FIG. 2 is a diagram showing a maximum value of a TF-IDF score of each emotion word shown in FIG.
FIG. 3 is a diagram showing 36 emotion vocabularies finally selected among the emotion words shown in FIG.
4 is a diagram illustrating an emotional lexical distribution map according to an embodiment of the present invention.
FIG. 5 to FIG. 8 are views showing a representation element extracted from comment data of an object according to an embodiment of the present invention in a heat-map form.
9 is a flowchart illustrating a method of visualizing a presentation element according to an exemplary embodiment of the present invention.
FIG. 10 is a flowchart illustrating a method of visualizing measured expressions according to frequency of expression elements according to an exemplary embodiment of the present invention. Referring to FIG.
FIG. 11 is a flowchart illustrating a process for confirming whether a new vocabulary is added according to an embodiment of the present invention.
12 is a flowchart illustrating the addition of a new rendering element according to an embodiment of the present invention.
FIG. 13 illustrates a process of determining the validity of an expression element according to an exemplary embodiment of the present invention. Referring to FIG.
FIG. 14 is a diagram illustrating a process of determining the validity of an expression element according to an exemplary embodiment of the present invention, based on the frequency of expression elements.
FIG. 15 is a flowchart illustrating a process for determining the validity of an expression element according to an exemplary embodiment of the present invention. Referring to FIG.
FIG. 16 is a flowchart illustrating a method of controlling influence of an expression element when specific expression elements are concentrated according to an exemplary embodiment of the present invention. Referring to FIG.
17 is a flowchart illustrating a method of assigning weights when a specific expression element according to an embodiment of the present invention actually appears in a specific object is low.
18 is a flowchart illustrating a method of mapping an expression element according to an exemplary embodiment of the present invention to a previously stored standard expression element and measuring frequency.
19 illustrates a system for visualizing presentation elements in accordance with an embodiment of the present invention.
Figure 20 illustrates a system for identifying new presentation elements and visualizing presentation elements in accordance with an embodiment of the present invention.
21 illustrates a system for visualizing presentation elements by measuring and adjusting the frequency of presentation elements according to an embodiment of the present invention.
FIG. 22 is a detailed view of an expression element extracting unit according to an embodiment of the present invention.
23 to 27 are diagrams illustrating different visualization methods according to an embodiment of the present invention.
28 is a diagram showing a heat-map visualization method according to an embodiment of the present invention in a three-dimensional application.
29 is a diagram showing contour lines of a visualization method of an expression element according to an embodiment of the present invention.
Fig. 30 is a diagram showing the contour map shown in Fig. 29 in three dimensions.
31 to 33 are diagrams illustrating an application method based on a semantic map according to an embodiment of the present invention.

Other objects and features of the present invention will become apparent from the following description of embodiments with reference to the accompanying drawings.

Preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

However, the present invention is not limited to or limited by the embodiments. Like reference symbols in the drawings denote like elements.

FIG. 1 is a diagram illustrating emotional vocabulary selected for producing an emotional lexical distribution map according to an embodiment of the present invention.

The present invention relates to a visualization method and system using an expression element collected from comment data of an object. The object is an object including a feeling, such as a movie, a product, a novel, a game, And can visualize emotions appearing from comments or reviews of objects that contain human emotions.

As an embodiment of the present invention, the object can be also described as a visualization method and system using comment data for a movie, which is limited to a movie.

The comment data of the movie can utilize the data collected through the web service constructed according to the user, or the comment data accumulated in the large portal and the fan club bulletin board can be individually collected using the program.

As an embodiment of the present invention, a web crawler capable of collecting data may be used to automate emotional vocabulary collection including user's emotion from comment data of a movie. The crawler may be a large portal (Naver, ) It is possible to collect comments and comments of certain movies on the movie homepage in the form of non-refined data, process the collected data into data that can be used for research, and extract emotional vocabulary by analyzing the refined data. Accordingly, the emotional vocabulary collected through the crawler can be linked with the situation of watching a movie, so that a movie suitable for a user's motivation can be recommended later.

To visualize the frequency of emotion vocabulary in a movie, the position of each emotion word should be specified on a two-dimensional plane. For this, the positional coordinates on two dimensions can be derived by using correlation of emotion words. In order to construct a distribution map of emotional vocabulary, we classify only emotional vocabulary that can be perceived when watching movies among 834 emotional terms by referring to the study on the appropriateness and experience frequency of Korean emotional terms of Han Dukwoong and Kang Hyeja (2000). At this time, only one doctoral specialist in Korean Language and Literature Department of Ajou University and two inventors of the present invention chose only emotional vocabulary to collect opinions, and finally selected 100 emotional vocabularies.

In addition, in addition to emotional vocabulary selection through expert analysis, a survey was conducted to select the final emotional vocabulary based on the 100 emotional vocabularies selected to select the emotional vocabulary that users most feel when watching movies. The survey was conducted by a group of 30 media students at Ajou University. After a brief explanation of the emotions that could be felt when watching a movie, when the viewer views a movie about 100 emotional vocabularies obtained through expert analysis, To the extent that they can feel it. In the actual questionnaire, "I thought about the movie stories of various genres you have watched so far, and then what is the degree of feeling the following emotional vocabulary when watching the movie?" I responded with a Likert-type scale, 1 point meaning 'no relevance at all' and 7 point meaning 'very relevant'.

In this study, we conducted a specialist analysis and user survey to collect the best vocabulary that can be felt when watching movies in order to meet the recommendation of movies using user motivation. In order to select relevant emotional vocabularies based on the information, 32 averaged emotional vocabularies (below 4.00 'normal') were further removed through average analysis, and 68 emotional vocabularies Respectively.

FIG. 1 is a diagram showing 68 emotional vocabularies suitable for the selected movie recommendation.

FIG. 2 is a diagram showing a maximum value of a TF-IDF score of each emotion word shown in FIG.

FIG. 2 is a flowchart illustrating a method for deriving a TF-IDF score of each emotional vocabulary appearing in a movie comment or a review, in order to further remove emotional vocabulary having little influence by comparing actual movie data with the 68 emotional vocabularies described in FIG. And the maximum value of the TF-IDF score that can appear in each emotion vocabulary.

In this case, TF (word frequency, term frequency) means a value indicating how often a specific word appears in a document, DF (Document Frequency) means the number of documents in which a specific word appears, inverse document frequency).

FIG. 3 is a diagram showing 36 emotion vocabularies finally selected among the emotion words shown in FIG.

In FIG. 2, the ratio of TF-IDF score is 0.8% or less in all movies in the case of 'astonishing' among the respective emotion vocabulary derived from the TF-IDF score, , The ratio of the TF-IDF score in at least one movie is 42%.

FIG. 3 is a diagram showing the 36 selected emotion vocabularies after eliminating the emotional vocabulary of which the ratio of the TF-IDF score is less than 10%.

4 is a diagram illustrating an emotional lexical distribution map according to an embodiment of the present invention.

In order to derive the semantic distances between the 36 emotion vocabularies shown in FIG. 3 on the two-dimensional plane, the distance between similar or different emotional vocabularies based on 36 emotional vocabularies was analyzed to analyze the correlation Multi-Dimensional Scaling (MDS) can be used.

At this time, multidimensional scaling analysis is a technique related to statistics which calculates the relative distances between individuals and is expressed as a distance relative to a plane that can be perceived by humans, and is a background technique for measuring similarities and dissimilarities in data in information visualization.

The advantage of the multidimensional scaling method is that it can create semantic maps of individuals who know only relative distances and can create semantic maps based on physical distances as well as psychological distances.

In order to analyze the multidimensional scale according to an embodiment of the present invention, a semantic distance survey was conducted on 36 emotion vocabularies, with a total of 20 persons as experimental subjects, 11 male and 9 female students in their twenties in Gyeonggi Province and Seoul, The questionnaire was composed of 36 questionnaires (68x68) arranged along the horizontal axis and vertical axis. The questionnaire survey was conducted using the Rickert scale, which gives 3 points when the distance between the emotional vocabulary is the closest and 3 points when it feels the farthest distance. Respectively. Figure 4 shows the Metric MDS based on the 68 emotional vocabulary selected for the semantic distance between 36 movies and 36 emotional vocabularies, based on data recorded by 20 people.

As a result, the emotional vocabularies related to the representative words "Happy" and "Surprise" were distributed in the positive direction of the X axis and the representative words "Anger", "Disgust" Related emotional vocabulary. The emotional vocabulary related to the representative words "Fear" and "Surprise" was distributed in the positive direction of the Y axis and the emotion vocabulary related to the representative words "Sad" and "Boring" The vocabulary was distributed.

As a result, positive emotional vocabularies are distributed in the positive (+) direction of the X axis, and negative emotional vocabularies are distributed in the negative (-) direction of the X axis due to the nature of the emotional vocabulary.

In addition, positive (+) direction of the Y axis is the distribution of emotional vocabularies that are dynamic (which can take a relatively large gesture when feeling emotion), and static (negative (Which can take a small gesture).

 The vocabularies associated with the representative words 'Happy', 'Sad', 'Anger', 'Fear', 'Disgust', and 'Boring' can be seen as distinct clusters of words, Happy 'and' Fear ', respectively. This can be interpreted as a result of the fact that when a user watches a movie, "an incredible feeling occurs due to a joyful joy" and "when an amazing feeling occurs due to a sudden appearance of fear" is dominant.

FIG. 5 to FIG. 8 are views showing a representation element extracted from comment data of an object according to an embodiment of the present invention in a heat-map form.

In order to visualize the emotional vocabulary extracted from the comment data for the movie described with reference to Figs. 1 to 4, the frequency of the emotional vocabulary constituting the MDS Map is required. We compare emotional vocabulary dictionaries with selected comment data through the upper process and measure emotional vocabulary frequency in each movie.

In addition, the TF-IDF score is calculated and adjusted to lower the weight of a specific vocabulary that appears frequently, regardless of the nature of the movie. Finally, the visualization can be done using the TF-IDF score of each selected emotion vocabulary.

 The final visualization graph can be represented as a heat-map consisting of small cells in a square with the background of the emotional vocabulary MDS Map. At this time, all the cells are initialized to a value of 0, and the cell value increases according to the TF-IDF score of the emotion vocabulary located in the corresponding cell. As the value of the cell increases, the color changes to a different color, so that the TF-IDF score of the emotion vocabulary can be ascertained high or low. Also, the cell with the numerical value affects the numerical value of the surrounding cell, so that the graph is shown as a topographic map.

5 is a graph visualizing the distribution of the emotional vocabulary appearing in the comment data of visitors to the movie 'Snowy National Railroad'. As shown in FIG. 5, in the movie "Snowy trains", the audience responds that they are interesting and wonderful, and the feeling that they are sad and boring is also showing a high frequency. Indeed, if you look at the movie comments on the Snowy Trains, you can see many disappointed audiences in the movie.

FIG. 6 is a visualization of the movie "Purgatory and murder" in the form of a heat map, and the highest emotion vocabulary among the visitors' feelings in the horror movie "Purgatory and murder" is "surprising" The frequency of emotional vocabulary related to fear is high.

FIG. 7 is a diagram visualizing the movie 'Don Clamamy' in the form of a heat map. The distribution of the viewer's emotional vocabulary on the 'Don Clamami' And 'outrageous', indicating that they are frequent.

FIG. 8 is a visualization of the movie "Warang Sound" in the form of a heat map. In the case of Warang Son, the audience's emotions are highly sad and touching.

Through the above example, it can be seen that the comment data collected from the comment after the movie is viewed matches the genre and story characteristics of the movie, and the emotional vocabulary pattern appears.

As an embodiment of the present invention, a method of extracting an emotional vocabulary using a comment data of a movie and visualizing the emotional vocabulary has been described. However, in the case of an idea, an intention, an evaluation, It may also be the subject of emotional reactions such as cognitive activities, emotions, emotions, desires, and attitudes, such as opinions, arguments, and refutations.

In addition, the object to which the present invention can be applied includes human emotions, emotions, desires, attitudes and the like in an emotional part, and includes thinking, intention, evaluation, opinion, argument, refutation and the like in the recognition part. Also, the relationship part includes cultural contents, human relations (communication, conflict), social relations (multicultural homo hundred etc.), and relationship with technology (cultural retention, etc.).

9 is a flowchart illustrating a method of visualizing a presentation element according to an exemplary embodiment of the present invention.

A method of visualizing an expression element includes extracting a plurality of expression elements from comment data collected for an object selected by the user (S910), and extracting a plurality of expressions extracted based on the distribution based on the semantic distance between the extracted plurality of expression elements Visualize the elements (S920).

At this time, the comment data of the object means all comment data including the feelings of the people such as a review of a movie, a review of a product, a review of a novel, a review of a game, a review of a travel,

In addition, the expression elements include words, paragraphs, emoticons, etc. expressing emotions of people extracted from the comment data.

In addition, a method of visualizing based on the distribution based on the semantic distance between a plurality of expression elements is based on a multi-dimensional scaling map (MDS map) described in Figs. 1 to 4, map shape or contour shape.

FIG. 10 is a flowchart illustrating a method of visualizing measured expressions according to frequency of expression elements according to an exemplary embodiment of the present invention. Referring to FIG.

In the method of visualizing the presentation element, a plurality of presentation elements are extracted from the collected comment data for the object selected by the user (S910), the frequency extracted in the comment data of the extracted presentation elements is measured (S930) (S920) a plurality of extracted expression elements based on a distribution based on a semantic distance between a plurality of expression elements. At this time, the extracted expression elements are classified into a shape of a heat map, a contour line, etc. based on a multi-dimensional scaling map (MDS map) including expression elements according to the frequency of the measured expression elements. Can be visualized.

In addition, when the expression elements are extracted and the frequency of the expression elements is measured, when the standardized expression elements are not standard types, they are mapped to the expression elements of the standard type stored in the dictionary, The frequency of the comment data of the object can be measured.

FIG. 11 is a flowchart illustrating a process for confirming whether a new vocabulary is added according to an embodiment of the present invention.

9 extracts a plurality of expression elements from the collected comment data for the object selected by the user (S910), and compares the extracted expression elements with the extracted expression elements (S1110 ). In operation S1120, a plurality of expression elements extracted based on the distribution based on the semantic distances between the plurality of extracted expression elements are visualized in operation S920. can do. In this case, the process of finding the meaning of a new expression element can be performed through a technique such as context-based analysis.

12 is a flowchart illustrating the addition of a new rendering element according to an embodiment of the present invention.

In step S921, it is determined whether a new expression element is added in step S921. If a new expression element is added, a new expression element and one or more adjacent expression elements (S922).

In this case, the predetermined criterion may be that of the previously extracted expression elements, which is closest to the new expression element, or may be a combination of the extracted expression elements and the new expression element within the range of r You can also use the case as a reference.

Then, the semantic position of the new expression element is determined based on the determined semantic distance from the adjacent expression elements (S923), and the new found expression element is visualized (S924).

At this time, as the similarity between the new expression element and the neighboring expression element becomes similar, the position may be determined by weighting so that the meaning distance between the new expression element and the adjacent expression element becomes closer to each other. That is, when the first adjacent expression element is similar to the meaning of the new expression element than the second neighbor expression element, the distance between the new expression element and the first neighbor expression element is smaller than the distance between the new expression element and the second adjacent expression element The position of the new presentation element can be determined so that it is shorter than that of the new presentation element. In this case, the semantic similarity between expression elements can be obtained through context-based analysis, or it can be obtained by various methods such as questionnaires for a large number of people.

FIG. 13 illustrates a process of determining the validity of an expression element according to an exemplary embodiment of the present invention. Referring to FIG.

In the method of visualizing the expression elements shown in FIG. 9, a plurality of expression elements are extracted from the collected comment data about the object selected by the user (S910), and the validity of the extracted expression elements is determined (S1310). At this time, if the extracted expression element is not valid for the object selected by the user, the invalid expression element is removed (S1320).

In operation S920, a plurality of expression elements from which the invalid expression elements are removed from the plurality of expression elements extracted based on the distribution based on the semantic distance between the extracted plurality of expression elements may be visualized. As a criterion for determining that a specific expression element is invalid, the meaning of the specific expression element is significantly different from other expression elements, the frequency of the specific expression element is remarkably less than the reference value, (In this case, it may be a mechanical repetition, rather than a genuine review, or an announcement, etc.).

FIG. 14 is a diagram illustrating a process of determining the validity of an expression element according to an exemplary embodiment of the present invention, based on the frequency of expression elements.

In the method of visualizing the expression elements described in FIG. 13, a plurality of expression elements are extracted from the collected comment data about the object selected by the user (S910), and the extracted frequency in the comment data of the extracted expression elements is measured S1410).

Then, the validity of the extracted expression element is determined using the frequency of the expression element (S1310). At this time, if the extracted expression element is not valid for the object selected by the user, the invalid expression element is removed (S1320).

FIG. 15 is a flowchart illustrating a process for determining the validity of an expression element according to an exemplary embodiment of the present invention. Referring to FIG.

In the method of visualizing the presentation element shown in FIG. 9, a plurality of presentation elements are extracted from the collected comment data about the object selected by the user (S910), and the extracted expression elements are extracted from the objects other than the object selected by the user (S1510). Thereafter, it is determined whether or not the extracted expression elements have been extracted by more than a predetermined number of objects among the objects other than the object selected by the user (S1520). If the extracted expression elements are extracted from a certain number or more of objects, The weight is adjusted (S1530). Then, a plurality of extracted expression elements are visualized based on the distribution based on the semantic distance between the extracted expression elements (S920).

Accordingly, if the expression element appears equally to all objects (contents) without discrimination, it can be regarded as invalid.

FIG. 16 is a flowchart illustrating a method of controlling influence of an expression element when specific expression elements are concentrated according to an exemplary embodiment of the present invention. Referring to FIG.

In the method of visualizing the presentation element shown in FIG. 9, a plurality of presentation elements are extracted from the collected comment data about the object selected by the user (S910), and the frequency of extraction of the presentation elements is measured (S1610). Thereafter, the measured frequency is weighted according to the measured frequency, and the measured frequency is adjusted (S1620). At this time, the presentation elements can be visualized by reflecting the adjusted frequency (S920).

Accordingly, when a specific expression element appears to be concentrated on a specific content, the influence of the specific expression element can be adjusted by adjusting the weight. In other words, if the specific expression elements appear to be overly concentrated, the influence of the other expression elements may be underestimated.

17 is a flowchart illustrating a method of assigning weights when a specific expression element according to an embodiment of the present invention actually appears in a specific object is low.

In the method of visualizing the expression elements shown in FIG. 9, a plurality of expression elements are extracted from the collected comment data about the object selected by the user (S910), and the frequency of extraction of the expression elements from the comment data is measured (S1710). Thereafter, the frequency with which the presentation element appears in the object selected by the user is compared with the measured frequency (S1720), and the presentation element is extracted according to the comparison result between the frequency at which the presentation element appears in the object selected by the user and the measured frequency. The frequency is weighted and the measured frequency is adjusted (S1730).

Accordingly, a frequency at which a specific expression element actually appears in a specific object (content / movie) is compared with a frequency at which comment data appears, and a low weight value can be given when the frequency of occurrence in the comment data is low.

18 is a flowchart illustrating a method of mapping an expression element according to an exemplary embodiment of the present invention to a previously stored standard expression element and measuring frequency.

In the method of visualizing the presentation element shown in FIG. 10, when a plurality of presentation elements are extracted from the collected comment data for the object selected by the user (S910), the extracted presentation elements in the database, in which the standardized presentation elements are stored in advance, (S911). If the extracted expression elements are not stored in the database, the standardized expression elements on the database closest to the extracted expression elements are identified as representative expression elements of the extracted expression elements (S912).

Thereafter, when the frequency extracted in the comment data of the extracted expression elements is measured (S930), the extracted frequency of expression elements in the comment data is added to the frequencies extracted in the comment data, And visualizes a plurality of expression elements extracted based on the distribution based on the semantic distance between the extracted expression elements (S920), the representative expression elements are visualized by reflecting the sum frequency.

Accordingly, when the expression element is not a standard type, it is mapped to a standard-type expression element stored in the emotional vocabulary dictionary (in a pre-stored database), and in the mapped emotional vocabulary dictionary, The frequency can be measured.

19 illustrates a system for visualizing presentation elements in accordance with an embodiment of the present invention.

A system 1900 for visualizing presentation elements may be, for example, a computing system, and includes a storage device 1910 and a processor 1920. The processor 1920 includes an expression element extraction unit 1930, a frequency measurement unit 1940, a validity determination unit 1950, and a visualization unit 1960.

The storage device 1910 stores the comment data for the object selected by the user and the expression element extraction unit 1930 extracts a plurality of expression elements from the comment data stored in the storage device 1910 and outputs the extracted expression data to the visualization unit 1960 And visualizes a plurality of extracted presentation elements based on the distribution based on the semantic distance between the extracted plurality of presentation elements.

In addition, when the frequency extracted in the comment data of the expression elements extracted by the frequency measurement unit 1940 is measured, the visualization unit 1960 may visualize the extracted expression elements according to the frequency of the measured expression elements, , The validity determination unit 1950 determines the validity of the extracted expression element, and if the extracted expression element is not valid for the object selected by the user, it may remove the invalid expression element.

The frequency measurement unit 1940 also identifies the frequencies at which the presentation elements extracted by the presentation element extraction unit 1930 are extracted from objects other than the object selected by the user, It is possible to judge whether or not a certain number or more of objects have been extracted at a certain frequency or more.

At this time, the comment data of the object means all comment data including the feelings of the people such as a review of a movie, a review of a product, a review of a novel, a review of a game, a review of a travel,

In addition, the expression element includes a word, a paragraph, an emoticon, etc. expressing the feelings of people extracted from the comment data, and based on the distribution based on the semantic distance between the plurality of expression elements, the multidimensional scale analysis Based on a map (Multi-Dimensional Scaling map: MDS map), it can be visualized in the shape of heat-map or contour.

Figure 20 illustrates a system for identifying new presentation elements and visualizing presentation elements in accordance with an embodiment of the present invention.

A system 1900 for identifying new presentation elements and visualizing presentation elements includes a storage device 1910 and a processor 1920. The processor 1920 includes an expression element extraction unit 1930, an expression element comparison unit 1970, a new expression element verification unit 1980, and a visualization unit 1960.

The storage device 1910 stores the comment data for the object selected by the user and the expression element extraction unit 1930 extracts a plurality of expression elements from the comment data stored in the storage device 1910 and outputs the extracted expression data to the visualization unit 1960 And visualizes a plurality of extracted presentation elements based on the distribution based on the semantic distance between the extracted plurality of presentation elements.

At this time, the expression element comparison unit 1970 compares the extracted expression elements with the expression elements extracted from the expression element extraction unit 1930, and the new expression element verification unit 1980 checks whether the extracted expression elements are new Check whether an expression is added.

21 illustrates a system for visualizing presentation elements by measuring and adjusting the frequency of presentation elements according to an embodiment of the present invention.

A system 1900 for visualizing presentation elements includes a storage device 1910 and a processor 1920. The processor 1920 includes an expression element extraction unit 1930, an expression element comparison unit 1970, a new expression element verification unit 1980, and a visualization unit 1960.

The storage device 1910 stores the comment data for the object selected by the user and the expression element extraction unit 1930 extracts a plurality of expression elements from the comment data stored in the storage device 1910 and outputs the extracted expression data to the visualization unit 1960 And visualizes a plurality of extracted presentation elements based on the distribution based on the semantic distance between the extracted plurality of presentation elements.

The frequency measurement unit 1940 identifies the frequencies at which the presentation elements extracted by the presentation element extraction unit 1930 are extracted from objects other than the object selected by the user, It is possible to determine whether or not the object has been extracted more than a certain frequency from a predetermined number or more of objects.

At this time, the frequency adjuster 1990 may weight the frequency of the presentation element according to the frequency of the presentation element according to the frequency of the measured presentation element, and adjust the frequency of the presentation element.

At this time, the visualization unit 1960 can visualize the presentation elements by reflecting the adjusted frequency in the frequency adjustment unit 1990.

Accordingly, even when a specific expression element appears overly concentrated on a specific content, the influence of the expression element can be controlled by lowering the weight.

Also, the frequency measurement unit 1940 measures the frequency of extraction of the presentation elements from the stored comment data, and compares the frequency with which the presentation elements appear in the object selected by the user and the identified frequency. At this time, Can adjust the frequency of extracting the expression elements by assigning weights to the extracted frequencies of the expression elements according to the comparison result between the frequency of occurrence of the expression elements in the object selected by the user and the identified frequency.

Accordingly, a frequency at which a specific expression element actually appears in a specific object (content / movie) is compared with a frequency at which comment data appears, and a low weight value can be given when the frequency of occurrence in the comment data is low.

FIG. 22 is a detailed view of an expression element extracting unit according to an embodiment of the present invention.

The expression element extraction unit 1930 shown in FIGS. 19 to 21 includes an expression element search unit 1931 and an expression element identification unit 1932.

The expression element search unit 1931 searches whether or not the extracted expression element is stored in the database in which the standardized expression element is stored in advance and the expression element identification unit 1932 determines whether the extracted expression element is stored in the database , And identifies the standardized expression element on the database closest to the extracted expression element as the representative expression element of the extracted expression element.

At this time, the frequency measurement unit 1940 adds the extracted frequency of the extracted expression elements to the frequency of the identified representative expression elements in the comment data, and the visualization unit 1960 adds the representative expression elements to the frequency And visualized by reflecting on the frequency.

Accordingly, when the expression element is not a standard type, it is mapped into a representation element of a standard type stored in an emotional vocabulary dictionary (in a pre-stored database), and in the comment data of each object on the basis of a standard expression element on the mapped emotional vocabulary dictionary Can be measured.

23 to 27 are diagrams illustrating different visualization methods according to an embodiment of the present invention.

23 is a graph showing the present invention in the form of a scatter plot in addition to a heat map form according to an embodiment of the present invention. In this case, as the frequency of the expression vocabulary increases, the color becomes redder . 24 is a graph in the form of Small Multiples.

FIG. 25 shows the present invention in the form of contour lines. In this case, the higher the frequency, the higher the value is, and the higher the height, the higher the frequency. Fig. 26 is a diagram showing a choropleth map. Fig. In this case, the present invention can be realized not only in the shape of a rectangle, but also in the form of a natural terrain or a natural object such as a part of a map.

FIG. 27 is a diagram showing the present invention with statistical cartograms. In this case, when each expression word comes out according to the comment data or the opinion selected by the user, a high-frequency expression word is given to the user So that a high-frequency expression word selected by the user can be displayed on the map.

28 is a diagram showing a heat-map visualization method according to an embodiment of the present invention in a three-dimensional application.

Although the heat-map visualization described in the present invention is shown on a two-dimensional plane, it can be transformed into a three-dimensional form such as three-dimensional while maintaining the same property.

Angle, pixel size, and color can be adjusted according to the frequency of the expression elements. FIG. 28 is a diagram showing a heat-map form in the three-dimensional form according to the present invention.

29 is a diagram showing contour lines of a visualization method of an expression element according to an embodiment of the present invention.

FIG. 29 is a diagram showing a two-dimensional contour line according to the frequency of the expression vocabulary. The color and size of the contour line can be adjusted according to the frequency of the expression vocabulary.

Fig. 30 is a diagram showing the contour map shown in Fig. 29 in three dimensions.

FIG. 30 is a diagram showing the two-dimensional contour map shown in FIG. 29 in a three-dimensional contour line according to the frequency of the presentation lexicon, and the color, height and size of the contour line can be adjusted according to the frequency of the presentation lexicon.

31 to 33 are diagrams illustrating an application method based on a semantic map according to an embodiment of the present invention.

FIG. 31 shows an embodiment that utilizes positioning of a multi-dimensional scaling map (MDS map) used in the present invention. The MDS map positioning is an existing positioning Dimensional positioning based on various attributes shown on the MDS map.

For example, Audi, located in the second quadrant, is located in the same quadrant of BMW as the automotive enterprise, but it is more closely related to the future-oriented image. Likewise, the SM in the fourth quadrant is closer to the relaxed image than the KIA.

Such positioning using the MDS Map can be utilized not only for the corporate image but also for image positioning of goods, characters, and characters as shown in FIGS. 32 and 33, according to the features appearing on the MDS.

The embodiments of the present invention described above are centered on expression elements (elements including vocabulary, emoticon, emotion, evaluation, and opinion) extracted from comment data on one object (content). However, the idea of the present invention is not limited to the case where the expression elements of the comment data for one object are intuitively expressed in one emotional map.

That is, according to another embodiment of the present invention, a menu for providing an editing menu by a user or a comparison analysis function for two or more objects may be provided. At this time, the user can select the first object and the second object to compare the presentation elements in the review for the first object with the presentation elements in the review for the second object. At this time, the comparison menu can perform a set operation (union, intersection, difference set) between the expression elements in the review for the first object and the expression elements in the review for the second object, A re-draw menu may also be provided that allows visualization again for a union or difference set.

In addition, not only one visualization data exists for the same object, but there may be a visualization version according to two or more time versions (or time layers) according to time series version control, . ≪ / RTI > At this time, attributes of each node (expression element) according to time can be represented by area, color, etc., and can reflect frequency, concentration, and the like. For example, the heat-map can be an example of this.

The visualization method of the presentation element according to an exemplary embodiment of the present invention can be implemented in the form of a program command 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 configured for the present invention 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 devices described above may be configured to operate as one or more software modules to perform the operations of the present invention, 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 .

1910: Storage device
1920: Processor

Claims (21)

Extracting a plurality of expression elements from the collected comment data for an object selected by the user; And
Visualizing the extracted plurality of presentation elements based on a distribution based on a semantic distance between the extracted plurality of presentation elements;
Lt; / RTI >
Wherein a distance in a plane or space of each of the visualized plurality of presentation elements is determined based on a semantic distance between each of the plurality of presentation elements. The method according to claim 1,
Measuring the extracted frequency in the comment data of the extracted presentation elements;
Further comprising:
The step of visualizing the presentation elements
And visualizing the extracted presentation elements according to the frequency of the measured presentation elements. The method according to claim 1,
After extracting the presentation elements,
Comparing the extracted expression elements with previously extracted expression elements; And
Determining whether a new expression element among the extracted expression elements has been added
Further comprising the steps of: The method of claim 3,
The step of visualizing the presentation elements
Determining one or more neighboring expression elements having a semantic distance with the new expression element among the previously extracted expression elements within a certain standard when a new expression element among the extracted expression elements is added;
Determining a semantic position of the new presentation element based on the determined semantic distance from the adjacent presentation elements;
Lt; / RTI >
Wherein determining the semantic position of the new presentation element comprises:
When the semantic distance between the first adjacent expression element and the new expression element among the one or more adjacent expression elements is closer to the semantic distance between the second adjacent expression element and the new expression element among the one or more adjacent expression elements, Wherein the visualized position of the new presentation element is determined to be closer to the visualized position of the first adjacent presentation element than the visualized position of the second adjacent presentation element. The method according to claim 1,
After extracting the presentation elements,
Determining validity associated with the selected object of the extracted presentation element; And
Removing the invalid expression element if the extracted expression element is invalid for the object selected by the user;
Further comprising the steps of: 6. The method of claim 5,
After extracting the presentation elements,
Measuring the extracted frequency in the comment data of the extracted presentation elements;
Further comprising:
The step of determining the validity of the extracted expression element
Wherein the validity of the extracted presentation element is determined by reflecting the measured frequency of the extracted presentation element. The method according to claim 1,
After extracting the presentation elements,
Identifying a frequency at which the extracted presentation element is extracted from objects other than the object selected by the user; And
Determining whether the extracted presentation element has been extracted at a certain frequency or more from a number of objects other than the object selected by the user;
Further comprising the steps of: The method according to claim 1,
After extracting the presentation elements,
Measuring a frequency at which the expression elements are extracted; And
Assigning a weight to the measured frequency according to the measured frequency, and adjusting the measured frequency;
Further comprising:
The step of visualizing the presentation elements
And visualizing the presentation elements by reflecting the adjusted frequency. The method according to claim 1,
After extracting the presentation elements,
Measuring a frequency at which the expression elements are extracted from the comment data;
Comparing the frequency with which the expression element appears in the object selected by the user and the measured frequency; And
Adjusting the measured frequency by assigning a weight to the extracted frequency of the presentation element according to a result of comparison between the frequency at which the presentation element appears in the object selected by the user and the measured frequency,
Further comprising the steps of: 3. The method of claim 2,
The step of extracting the presentation elements
Searching whether the extracted presentation element is stored in a database in which standardized presentation elements are stored in advance; And
Identifying a standardized presentation element on the database closest to the extracted presentation element as a representative presentation element of the extracted presentation element if the extracted presentation element is not stored in the database;
Lt; / RTI >
The step of measuring the frequency includes
Adding the extracted frequency of the extracted expression elements in the comment data to the frequency of the extracted representative expression elements extracted in the comment data,
The step of visualizing the presentation elements
And visualizing the representative expression elements by reflecting the summed frequency. 3. The method of claim 2,
Wherein visualizing the presentation elements comprises:
Wherein the presentation elements are visualized using a multi-dimensional scaling map (MDS map) including the presentation elements and visualized by reflecting a meaning distance between the presentation elements. A computer-readable recording medium having recorded therein a program for executing the method according to any one of claims 1 to 11. A storage device for storing comment data for an object selected by a user;
An expression element extracting unit for extracting a plurality of expression elements from the stored comment data; And
A visualization unit for visualizing the extracted plurality of presentation elements based on a distribution based on a semantic distance between the extracted plurality of presentation elements;
Lt; / RTI >
Wherein a distance in a plane or space of each of the visualized plurality of presentation elements is determined based on a semantic distance between each of the plurality of presentation elements. 14. The method of claim 13,
A frequency measurement unit for measuring a frequency extracted in the comment data of the extracted expression elements;
Further comprising:
The visualization unit
And visualizes the extracted presentation elements according to the frequency of the measured presentation elements. 14. The method of claim 13,
An expression element comparison unit comparing the extracted expression elements with previously extracted expression elements; And
A new expression element checking unit for checking whether a new expression element among the extracted expression elements has been added;
The data visualization system further comprising: 14. The method of claim 13,
A validity determining unit for determining validity associated with the object selected by the user of the extracted presentation element and removing the invalid presentation element when the extracted presentation element is invalid for the object selected by the user;
The data visualization system further comprising: 17. The method of claim 16,
A frequency measurement unit for measuring a frequency extracted in the comment data of the extracted expression elements;
Further comprising:
The validity determination unit
Wherein the validity of the extracted presentation element is determined by reflecting the measured frequency of the extracted presentation element. 14. The method of claim 13,
Wherein the extracted expression element identifies a frequency at which the extracted expression element is extracted from another object other than the object selected by the user and determines whether the extracted expression element is extracted at a certain frequency or more from a predetermined number or more of objects other than the object selected by the user A frequency measuring unit for determining whether or not the received signal is a signal;
The data visualization system further comprising: 14. The method of claim 13,
A frequency adjuster for measuring frequency of extraction of the expression elements, weighting frequency of the expression elements according to the measured frequency, and adjusting frequency of the expression elements;
Further comprising:
The visualization unit
And visualizes the presentation elements by reflecting the adjusted frequency. 14. The method of claim 13,
A frequency measurement unit measuring a frequency at which the expression elements are extracted from the stored comment data and comparing the frequency with which the expression elements appear in the object selected by the user and the measured frequency; And
A frequency adjustment unit for assigning a weight to the extracted frequency of the expression elements according to a comparison result between the frequency at which the expression elements appear in the object selected by the user and the measured frequency,
The data visualization system further comprising: 15. The method of claim 14,
The expression element extracting unit
An expression element search unit for searching whether or not the extracted expression elements are stored in a database in which standardized expression elements are stored in advance; And
A representative expression element identification unit for identifying a standardized expression element on the database closest to the extracted expression element as a representative expression element of the extracted expression element if the extracted expression element is not stored in the database, ;
Lt; / RTI >
The frequency measurement unit
Adding the extracted frequency of the extracted expression elements in the comment data to the frequency of the extracted representative expression elements extracted in the comment data,
The visualization unit
And wherein the representative visualization element is visualized by reflecting the summed frequency.

Images