KR101997992B1 - System for generating survey for social emotion survey based on emotion vocabulary and method thereof - Google Patents

Abstract

The present invention relates to a questionnaire generation system and method for a social emotional survey based on emotional vocabulary. According to the present invention, selecting each of the M words among the words located in the region for each region in the multi-dimensional emotional coordinate system to which the emotional words are mapped, and for each region, the structure containing each of the M vocabulary Generating M candidate questions of each; calculating a reliability of each of the candidate questions based on a result of responses of a plurality of subjects to each of the candidate questions; and M candidate questions for each region. A method for generating a questionnaire for a social emotional survey based on an emotional vocabulary includes selecting a candidate question having a reliability greater than or equal to a threshold as an effective question for each region.
According to the present invention, it is possible to quickly and easily generate valid questions for a social emotional survey based on the emotional vocabulary, and provides an advantage of increasing the reliability of the survey.

Description

System for generating survey for social emotion survey based on emotion vocabulary and method

The present invention relates to a questionnaire generation system and method for a social emotional survey based on emotional vocabulary, and more particularly, a questionnaire generation system and method for quickly and easily generating valid questions required for a social emotional survey It is about.

Emotion refers to complex emotions such as comfort, luxury, refreshment, and discomfort that respond to sensory stimuli. Emotion is a comprehensive feeling, so it is sometimes difficult to express clearly, and it has personal and varied characteristics.

As a method of measuring an individual's emotion, a biological sensor is used to measure physiological responses such as pulse wave (PPG), brain wave (EEG), and electrical skin response (GSR) to questions. As such, the method of measuring physiological factors may have an effect similar to that of measuring an individual's subjective emotion, such as an actual survey.

However, both methods have limitations in measuring individual emotions, and in particular, physiological measurement methods require a separate biosensor device, and thus are not suitable for measuring social emotions for a large number of subjects. Therefore, there is a need for a technique for designing a survey that can reliably measure social sensitivity without additional verification using biosensors.

The background technology of the present invention is disclosed in Korean Patent Laid-Open Publication No. 2014-0002238 (published Jan. 08, 2014).

It is an object of the present invention to provide a questionnaire generation system and method for an emotional vocabulary-based social emotion survey that can quickly and easily generate valid questions required for a social emotion survey.

According to the present invention, the method includes selecting M words among the words located in the area for each area in the multi-dimensional emotional coordinate system to which the emotional words are mapped, and for each of the areas, the M words are included in the structure. Generating M candidate questions respectively, calculating a reliability of each of the candidate questions based on a result of a plurality of subjects' responses to each of the candidate questions, and among the M candidate questions for each region. A method for generating a questionnaire for a social emotional questionnaire based on an emotional vocabulary includes selecting a candidate question having a reliability greater than or equal to a threshold as a valid question for each region.

In addition, the multi-dimensional emotional coordinate system, the two-dimensional emotional coordinate system including a first axis of the sociality dimension associated with the strength of the pessimism / optimism, and a second axis of the dimension of mutuality associated with the intensity of the passive / active And may be divided into four regions by the first and second axes.

In the generating of the candidate question, the candidate question may be generated by inserting the selected vocabulary in a blank position in a predetermined template question selected from a DB storing a plurality of template questions.

In the calculating of the reliability, the reliability may be calculated using a Cronbach alpha technique.

In the selecting of the valid question, the candidate question having the confidence level equal to or greater than a threshold value is selected for each region, and only the top L candidate questions among the first selected candidate questions are finalized as valid questions for each region. Can be selected.

The method may further include storing the selected valid question in a question store for the social emotion survey.

In addition, the present invention, a vocabulary selection unit for selecting each of the M vocabularies of the vocabulary located in each region in the multi-dimensional emotional coordinate system to which the emotional vocabularies are mapped, and for each of the regions, A candidate question generator for generating M candidate questions each having an included structure, a calculator for calculating reliability of each candidate question based on a result of a plurality of subjects' responses to each of the candidate questions, and each of the regions. It provides a question generating system for a social emotional survey based on the emotional vocabulary including a control unit for selecting a candidate question of the M candidate questions for which the reliability is greater than or equal to the threshold as a valid question for each region.

According to the questionnaire generation system and method for the social emotional survey according to the present invention, it is possible to quickly and easily generate valid questions for the social emotional survey based on the emotional vocabulary and increase the reliability of the survey. to provide.

1 is a diagram showing a question generating system for a social emotional questionnaire based on emotional vocabulary according to an embodiment of the present invention.
2 is a view for explaining the emotional coordinate system used in the embodiment of the present invention.
3 is a view for explaining a method for generating a questionnaire using the system of FIG. 1.

DETAILED DESCRIPTION Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention.

The present invention relates to a questionnaire generation system and method for a social emotional survey based on emotional vocabulary, and when a subject's emotion (emotion) is to be measured through a survey, a valid question capable of measuring the related emotion better We present a questionnaire generation technique for social sentiment survey that provides quick and easy creation of (questions).

Here, surveying emotions or emotions among a plurality of subjects is expanded to the concept of social emotion rather than individual emotions. Thus, embodiments of the present invention use the term 'social emotion' rather than simple personal emotion. Social emotional vocabulary was modeled on SNS. KakaoTalk and Naver blogs and Facebook recruited emotional words into emotional words.

1 is a diagram showing a question generating system for a social emotional questionnaire based on emotional vocabulary according to an embodiment of the present invention. Referring to FIG. 1, a question generating system 100 according to an exemplary embodiment of the present invention includes a vocabulary selecting unit 110, a candidate question generating unit 120, an operation unit 130, and a controller 140.

The vocabulary selector 110 selects each of the M words among the words located in the area for each area in the multi-dimensional emotional coordinate system to which the emotional words are mapped. The selected vocabulary is used to generate candidate questions. The vocabulary mapped to each area of the emotional coordinate system may correspond to ordinary and emotional vocabulary.

2 is a view for explaining the emotional coordinate system used in the embodiment of the present invention. 2 is a two-dimensional emotional coordinate system, the first axis (X axis) of the sociality dimension related to the strength of the pessimism / optimism, and the second axis of the dimension of mutuality related to the intensity of the neglect / aggressiveness. It is divided into four quadrants (areas), including (Y axis).

The first quadrant (①) is an area that reveals optimism and positive emotion, the second quadrant (②) is an area that reveals the pessimism and positive emotion, and the third quadrant (③) is an area that shows the emotion of pessimism and negative The fourth quadrant (4) is an area that reveals optimism and negative emotions. Here, the vocabulary existing in the same quadrant (area) is different in intensity but exhibits similar sensitivities of properties or properties.

The X and Y axes of the emotional coordinate system each have a scale of -10 to +10 relative to the origin, and the larger the value, the greater the related intensity. A vocabulary with a more pessimistic property is placed toward the left side with x = 0, a vocabulary with a strong optimism property toward the right side, a vocabulary with a strong active property toward the top side with a y = 0, and a negative property toward the bottom Strong vocabulary is placed.

In this way, a plurality of vocabularies are arranged at each position (quadrant) of the emotional coordinate system at positions corresponding to the strengths of the pessimism / optimism or the neglect / aggressiveness. That is, the vocabulary associated with the optimistic / active characteristic is arranged in the first quadrant, and the vocabulary associated with the pessimistic / active characteristic is arranged in correspondence with the respective intensity in the second quadrant. The same is true for the remaining areas.

The vocabulary selecting unit 110 selects M vocabularies for each area of the emotional coordinate system. Accordingly, M words are selected per area, and the selected words are used to make candidate questions for each area. Here, the vocabulary may be randomly selected in each region or may be selected according to a preset rule. In addition, for each region, vocabularies that are separated from the origin by more than a set distance may be selected and various embodiments may exist.

2 illustrates a verb-type word for convenience of description, the present invention is not necessarily limited thereto, and the word may represent a part-of-speech such as a noun, an adverb, or an adjective having the same property. For example, the word “cool” described in the first quadrant of FIG. 2 may mean words including other parts of speech such as “cool”, “cool”, “cool”, and the like.

The candidate question generator 120 generates M candidate questions having a structure including M words each for each region of the emotional coordinate system. Accordingly, M candidate questions are generated for each region in the emotional coordinate system.

The two-dimensional emotional coordinate system shown in FIG. 2 has a total of four areas. If 10 words (M = 10) are selected for each area, 10 candidate questions are generated for each area (quadrant), so a total of 40 candidate questions are generated.

In this case, each candidate question (candidate item) has a structure including a corresponding vocabulary, and may be generated in the following manner. The candidate question generator 120 may select a predetermined template question from a DB in which pre-generated template questions are stored, and generate a candidate question by inserting a corresponding vocabulary at a blank position in a selected template question. have. The DB is a space in which a plurality of pre-made template questions are stored, and each stored template question has a form in which a part of sentences is blanked.

The calculator 130 calculates the reliability of each candidate question based on the response results of the plurality of subjects to each candidate question. If there are 50 subjects, 50 responses are obtained for each candidate question. In this case, the response result may correspond to one of scores classified from 'not at all' to 'very very'.

For example, in this example, a 7-likert scale response with scales ranging from 1 (no disagreement) to 7 (very disagreeing) when examining subject responses to candidate questions. Can be used.

Reliability judgment in the survey mainly uses the Cronbach Alpha (Cronbach's α) technique. The Kronbach alpha technique is used to evaluate which subjects (subjects' emotions) measure items that make up the subject. As a result of the reliability evaluation, the question items (questions) with low reliability are eliminated because they are judged as variables that lower the reliability in measuring the object.

As such, the Cronbach's alpha method is based on the correlation values between the measurement items and improves the overall reliability of the measurement tool by finding and removing the lower measurement items that lower the reliability.

The Cronbach's alpha technique means that the higher the alpha value α, the higher the reliability. α has a value from 0 to 1, and the closer to 1, the higher the reliability. In general, if it is 0.7 or more, it is selected as data with some reliability.

As such, the Kronbach alpha test is for evaluating the reliability of a questionnaire and is used to find and remove questions that impair reliability in the questionnaire. In other words, the Kronbach alpha test is used to exclude the inferior reliability of the data, and is used to obtain the overall reliability of the questionnaire items and the reliability of individual items. Of course, it is also possible to obtain the reliability of each item based on the result of obtaining the overall reliability of the state after subtracting the corresponding item from the entire item. This Cronbach's alpha test corresponds to a known technique, and thus a detailed description thereof will be omitted.

The controller 140 selects candidate questions having a confidence level equal to or greater than a threshold value of 0.7 among M candidate questions for each region as valid questions for each region. In this case, after selecting candidate candidates having a reliability higher than or equal to a threshold for each region, only the top L candidate questions among the first selected candidate questions may be finally selected as valid questions for each region.

According to the embodiment of the present invention, only the upper L (ex, two) data among the data having a reliability greater than 0.7 may be selected as valid data. Accordingly, two valid questions may be selected for each region.

Thereafter, the controller 140 stores the valid question selected for each region in the question store for the social emotion survey. Stored valid questions correspond to highly reliable questions, which are used as reliable questions for future social sentiment surveys, and can produce reliable survey results.

3 is a view for explaining a method for generating a questionnaire using the system of FIG. 1.

First, the vocabulary selecting unit 110 selects M vocabularies among the vocabularies located in each region for each region in the multidimensional emotional coordinate system (S310).

The candidate question generator 120 generates M candidate questions having a structure including M vocabularies, respectively, in each region (S320). In this case, a question automatic generation program for completing a sentence by inserting a corresponding vocabulary in a blank part of a template question selected in the DB may be used.

Thereafter, the calculator 130 calculates the reliability of each candidate question based on the response results of the plurality of subjects to each candidate question (S330). In this case, it may be determined that the question of which the reliability is lower than the threshold is not suitable for the question for the social emotion survey.

Next, the controller 140 selects candidate questions having a reliability greater than or equal to a threshold among the M candidate questions for each region as valid questions for each region (S340). In this case, only the top two valid questions may be selected among candidate questions whose reliability is greater than or equal to the threshold. Subsequently, the controller 140 stores valid questions selected for each area in a question repository for a social emotion survey (S350), so that it can be used for the actual survey later.

According to the questionnaire generation system and method for the social emotion survey according to the present invention as described above, it is possible to quickly and easily generate valid questions for the social emotion survey based on the emotional vocabulary and increase the reliability of the survey That provides the benefits.

Although the present invention has been described with reference to the embodiments shown in the drawings, this is merely exemplary, and it will be understood by those skilled in the art that various modifications and equivalent other embodiments are possible. Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

100: survey generation system 110: vocabulary selection unit
120: candidate question generation unit 130: arithmetic unit
140: control unit

Claims (12)

In the question generating method of the question generating system for the social emotional survey based on the emotional vocabulary,
Selecting M words among the words located in the area for each area in the multi-dimensional emotional coordinate system to which the emotional words are mapped;
For each region, each of the M candidate questions having a structure including each of the M words is generated, and the selected words are inserted at a blank position in a predetermined template question selected from a DB storing a plurality of template questions. Generating the candidate question;
Calculating a reliability of each of the candidate questions based on a response result of the plurality of subjects to each of the candidate questions;
Selecting a candidate question of which the reliability is equal to or greater than a threshold value among the M candidate questions for each area as a valid question for each area; And
Storing the selected valid question in a question store for the social emotion survey;
The multidimensional emotional coordinate system,
It is a two-dimensional emotional coordinate system including a first axis of the sociality dimension associated with the strength of the pessimism / optimism, and a second axis of the mutuality dimension associated with the intensity of the passive / active, wherein the first and second Questionnaire generation method for a social emotional survey based on emotional vocabulary divided into four areas by axis. delete delete The method according to claim 1,
Computing the reliability,
A method for generating a questionnaire for a social emotional survey based on emotional vocabulary that calculates the reliability using Cronbach alpha. The method according to claim 1,
The step of selecting the valid question,
After selecting the candidate questions with the reliability above the threshold for each region first, and then selecting the top L candidate questions among the first selected candidate questions as valid questions for each region. How to create a survey for you. delete A vocabulary selection unit for selecting M vocabularies among vocabularies located in each of the regions in the multi-dimensional emotional coordinate system to which the emotional vocabularies are mapped;
For each region, each of the M candidate questions having a structure including each of the M words is generated, and the selected words are inserted at a blank position in a predetermined template question selected from a DB storing a plurality of template questions. A candidate question generator to generate the candidate question;
A calculator configured to calculate a reliability of each candidate question based on a response result of a plurality of subjects to each candidate question; And
And a controller configured to select candidate questions having a reliability greater than or equal to a threshold among the M candidate questions for the respective areas as valid questions for the respective areas, and to store the selected valid questions in a question repository for a social emotion survey. ,
The multidimensional emotional coordinate system,
It is a two-dimensional emotional coordinate system including a first axis of the sociality dimension associated with the strength of the pessimism / optimism, and a second axis of the mutuality dimension associated with the intensity of the passive / active, wherein the first and second Questionnaire generation system for the social emotional survey based on the emotional vocabulary divided into four areas by the axis. delete delete The method according to claim 7,
The calculation unit,
Questionnaire generation system for a social emotional survey based on emotional vocabulary for calculating the reliability using the Cronbach alpha technique. The method according to claim 7,
The control unit,
After selecting the candidate questions with the reliability above the threshold for each region first, and then selecting the top L candidate questions among the first selected candidate questions as valid questions for each region. Questionnaire generation system for. delete

Images