KR101549188B1 - Apparatus and method for measuring brand image - Google Patents

Abstract

A brand personality measuring apparatus and a measuring method for enhancing the accuracy of measurement by extending the brand personality measurement item to a range of synonyms so that the subjective judgment of the researcher and the influence of the characteristics of the sample are less in the brand personality measurement item. The brand personality measuring apparatus according to one aspect of the present invention extracts synonyms of the measurement items from a hierarchical lexical semantic network for each of a plurality of measurement items for evaluating the personality of a brand, An extended metric group generation unit for generating an extended metric group; A sample collection unit for collecting a plurality of text samples related to the brand using a web crawler; A keyword extracting unit for extracting at least one keyword from the plurality of text samples; And an evaluation unit for evaluating the personality of the brand based on the extracted at least one keyword and the extended metric group.

Description

[0001] APPARATUS AND METHOD FOR MEASURING BRAND IMAGE [0002]

The present invention relates to a brand personality measuring apparatus and a measuring method. More particularly, the present invention relates to a brand personality measuring apparatus and a measuring method for improving the accuracy of measurement by analyzing contents by extending the brand personality measurement item.

Big data (Big Data) is an information technology that collects and analyzes valuable information and actively responds or predicts changes based on the generated knowledge. . Big data analysis has recently attracted attention because the cost of collecting and analyzing data has been drastically reduced compared to the past. As a result, analysis has become possible because of time and cost problems.

Recently, as the spread of smartphones has increased, a Location Based Service (LBS) has gained popularity because it accurately grasps the location of people and objects using wired and wireless communication networks and provides various contents and online and offline integrated services to users have. One of the most popular services in location-based services is using mobile to find popular restaurants nearby. The main business model of LBS is Customer Relationship Management (CRM). For effective CRM, it is important to understand customer needs and provide products or services that actual customers need.

In recent years, however, it has become increasingly difficult for clients to understand the needs of customers because of their diverse needs and complexity. In a highly competitive industry environment, information about customers who have a relationship with a company is very large, Since it is almost impossible to identify needs, the needs classification and classification based on big data analysis became necessary for customer needs analysis.

In this way, the need for customer needs analysis has increased, and the measurement of brand personality has been attracting attention and many studies are being carried out. However, it is pointed out that Aaker 's measure of brand personality, which is a representative method of brand personality measurement, is used in many studies, but there are limitations in generalization.

The development of the existing general measurement items investigated the previous research, extracted the measurement items for the relevant field, and organized them, and then they were verified through the expert group and the focus group. Then, This is because in many studies it is difficult to generalize due to the intervention of the subjective judgment of the researcher and the limitations of the sample group.

Therefore, it is necessary to measure how much the words used as brand personality measurement items are actually used in everyday life by using Big Data. .

1. Big Data ( Big Data )

As the technologies related to information and communication are continuously developed and the convergence with other industries is accelerated, a lot of data are being created around us. As smartphones become popular, and traffic is generated through tablet PCs, cameras, and game machines, traffic is also surging. In addition, the so-called 'big data age' is coming, in which the size and form of data varies, and the rate of increase of data is steep. In the meantime, Big Data has been discussed mainly in the category of management innovation. It was a move to collect and analyze vast amounts of data generated from various mobile devices and social media to understand the needs of customers and to introduce market needs into corporate management.

Big data is not new, but it has been from the past. We are excited about Big Data today because we are now able to analyze data that was too large to be analyzed in the past and we are discovering new values. Our data is being generated more quickly than in the past (Volume: the size of the data), more varieties (varieties of data), and close to real-time (Velocity) Big data analysis techniques using data have become real (Carter, 2011; Choi, Byung-joon et al., 2012). Definitions for Big Data Analysis vary by technology, size and methodology. From a technical point of view, it is defined as a next-generation technology and architecture designed to extract value from a wide variety of large-scale data at low cost, and to support ultra-high-speed acquisition, discovery and analysis of data (Carter, 2011) (McKinsey, 2011). From a methodological point of view, Big Data originally referred to only a huge data set that originally spanned tens to thousands of terabytes, but increasingly, related tools, platforms , And analytical techniques.

Big data analysis differs from existing management information analysis in that the existing management information analysis is based on refined data based on a certain form, whereas the big data analysis is based on data such as visit history of website, communication information of social media The analysis is based on data whose forms are different and unstructured. Therefore, Big Data Analysis uses advanced distributed processing techniques and statistical techniques to facilitate the handling of very large data with low structuring levels (Gruman, 2010). Big data analysis is widely used in the CRM field. As soon as a customer makes a payment at a store at a large discount store, the purchase history data of the customer is analyzed and a receipt printing a customized coupon for the customer is displayed. Even if a plurality of consumers purchase the same product, Each other coupon is outputted and the coupon is presented by presenting the coupon reflecting the customer's individual buying tendency in real time, thereby greatly increasing the additional purchase rate through use of coupon and increasing the marketing reaction rate of the customer, thereby leading to a profit increase.

Recently, big data analysis is used as a technique to read consumer mind by analyzing information gathered from SNS (Social Network Service) such as Twitter and Facebook, so that marketing means to increase company's sales and profits and crisis It is attracting attention as a management means.

2. Restaurant Brand Personality ( Restaurant Brand Personality )

Brand personality is a human characteristic associated with a brand. If the brand is a person, it is "a person of a certain personality." Consumers can choose the brand based on the evaluation of each brand's attributes (beliefs), or based on the unique image of each brand (Lee, HakSik, et al., 2006). Brand personality accumulates in consumers' minds through involuntary and involuntary efforts, rather than by voluntary or intentional paths. The brand personality can be formed from an image of a main user of the brand, an image of the company, an image of an advertisement model of the brand, and the like. For example, it is the ad model that reminds me of the strong masculine image of Marlboro cigarettes.

Brand personality metrics were presented by Aaker. Aaker is divided into four groups of 1,200 people based on the US Census Bureau's population census, extracted by gender, age, income, race and region, and then sent to each group in the mail with toothpaste, soap, discount stores, tires, credit cards, After selecting nine top brands of household necessities such as toys, coke, broadcasting, and fast food, a questionnaire consisting of 114 items was mailed and measured. It can be divided into three categories: Sincerity (down-to-earth, honest, wholesome, cheerful), Excitement (daring, spirited, imaginative, up-to-date) outdoorsy, and tough), 15 attributes (Facet Name), and 42 items (Traits).

Siguaw & Austin (1999) applied Aaker's brand personality scale to restaurants, and at Cornell University students, nine restaurants of quick service, casual dining, and upscale were used to examine the brand personality of restaurants by competence, sincerity, excitement, sophistication , and ruggedness.

Austin et al. (2003) used McDonald's, Burger King, Wendy's, Chili's, T.G.I. Brandness and brand personality were measured by six quick service restaurants such as Friday, Applebee's, and three upscale restaurants such as John Thomas Steakhouse, Dano's on Cayuga and Renee's American Bistro. Musante et al. (2008) suggested that Aaker's 42 measurements were reduced to 18 items suitable for quick service and casual dining.

The results of this study are as follows. First, the effect of brand personality on brand preference, brand loyalty, and e-Word of Mouth (Magnolia & Thelen, 2008) (2009) found that brand personality influenced emotions, satisfaction, and brand loyalty. Lin & Huang (2012) found that brand personality in the Taiwanese coffee brand chain was associated with repurchase intentions And a significant influence. According to Han Seong-ho (2011), domestic studies are conducted using Aaker's measurement items as they are, using some of Aaker's measurement items, using research items of other researchers, and developing independent restaurant brand personality measurement items have.

In many studies, brand personality metrics proposed by Aaker were excluded, added or changed according to the needs of researchers. Austin et al. (2003) used Reliability, Intelligent, and Successfulness of Excitement, Spirited, Imaginative, Up-to-date and Competence, and Musante et al. (2008) measured the Outdoorsy and Roughness of Excitement Lee et al. (2009) did not measure the upclass of Sincerity's Down-to-earth, Cheerful, Excitement Daring, Competence of Reliability, and Sophistication.

The reason that the brand personality measurement items suggested by Acker are excluded, added, or changed according to the needs of the researchers is that it is difficult to generalize the research results due to the subjective judgment of the researchers and the limitations of the sample .

 Therefore, there has been a request for a device and method for measuring the personality of a brand person whose subjective judgment and the influence of the characteristics of the sample are less influenced by the personality measurement item of the researcher.

The object of the present invention is to provide a brand personality measuring apparatus and a measuring method for enhancing the accuracy of measurement by extending the brand personality measurement item to a range of synonyms so that influence of the subjective judgment of the researcher and the characteristics of the sample less affects the brand personality measurement item do.

The brand personality measuring apparatus according to one aspect of the present invention extracts synonyms of the measurement items from a hierarchical lexical semantic network for each of a plurality of measurement items for evaluating the personality of a brand, An extended metric group generation unit for generating an extended metric group; A sample collection unit for collecting a plurality of text samples related to the brand using a web crawler; A keyword extracting unit for extracting at least one keyword from the plurality of text samples; And an evaluation unit for evaluating the personality of the brand based on the extracted at least one keyword and the extended metric group.

At this time, the web crawler may be a wrapper-based web crawler.

The keyword extracting unit may further include a morpheme analyzing unit for analyzing morphemes for each of the text samples and extracting an adjective among the morphemes included in the text sample as the keyword for the text sample.

In addition, the extended metric group generation unit may prevent the previously designated vocabulary from being included in the extended metric group as a synonym of the evaluation item.

Furthermore, a pre-designated vocabulary can be a vocabulary inappropriate for a measurement item because it has various meanings.

The evaluation unit may evaluate the personality of the brand by collating the expanded measurement item group with the keyword and calculating the frequency of use for each measurement item.

The storage unit may further include a storage unit for storing the plurality of measurement items and the extended measurement item group.

Also, the hierarchical lexical semantic network may be WordNet.

In addition, the plurality of measurement items may be the 42 brand personality measurement items of the acker.

The method for measuring the brand personality according to another aspect is a method for extracting synonyms of the measurement items from a hierarchical lexical semantic network for each of a plurality of measurement items for evaluating the personality of a brand, ; Collecting a plurality of text samples related to the brand using a web crawler; Extracting at least one keyword from the plurality of text samples; And evaluating the personality of the brand based on the extracted at least one keyword and the extended metric group.

At this time, the collecting step can be performed by a wrapper-based web crawler.

The extracting of the keyword may include analyzing the morpheme for each of the text samples; And extracting an adjective of the morpheme included in the text sample as the keyword for the text sample.

Determining whether a predefined vocabulary is included as a synonym in the expanded metric group and excluding the predefined vocabulary from the group if the predefined vocabulary is included in the extended metric group; .

In addition, in the evaluation step, the individuality of the brand can be evaluated by collating the expanded measurement item group with the keyword and calculating the frequency of use for each measurement item.

The effect of the subjective personality judgment and the characteristics of the sample on the brand personality measurement items is extended to the range of the synonyms so that the personality measurement device and the measurement method of improving the accuracy of the measurement can be implemented have.

That is, even if the measurement items of the brand personality are not based on the words commonly used by the general consumers, such as the measurement items of the existing ackers, the measurement items are expanded so that the brand personality is measured based on the words used by the general consumers , And the accuracy of the brand personality measurement is improved.

1 is a block diagram showing an example of a brand personality measuring apparatus,
2 is a block diagram showing another example of the brand personality measuring apparatus,
FIG. 3 is a flowchart showing an example of a method for measuring brand personality,
4 is a table showing an example of a sample collection technique,
FIG. 5 is a table showing the results of analyzing the actual use of the existing brand personality measurement items,
6 is a table illustrating brand personality measurement items extended through WordNet,
FIG. 7 is a table comparing actual usage frequency of the brand personality measurement items and actual usage frequency of the brand personality measurement items extended through WordNet.

BEST MODE FOR CARRYING OUT THE INVENTION The present invention will be described in more detail with reference to the drawings.

1 is a block diagram showing an example of a brand personality measurement apparatus.

1, the brand personality measurement apparatus 10 includes an extended measurement item group generation unit 10, a sample collection unit 12, a keyword extraction unit 14, and an evaluation unit 16.

The expanded metric group generation unit 10 extracts synonyms of the measurement items from the hierarchical lexical meaning network 110 for each of a plurality of measurement items for evaluating the individuality of the brand, Create a group of metrics.

A plurality of measurement items can use 42 measurement items suggested by Aaker. In addition, the brand personality measurement apparatus 1 may further include a storage unit 100 so that the measurement items and the extended measurement item groups for each measurement item are stored.

The sample collecting unit 12 collects a plurality of text samples 200 related to a specific brand through the network 20 such as the Internet using a web crawler.

The web crawler utilized by the sample collecting unit 12 can be applied to any web crawler that can quickly and accurately collect web documents of a certain subject. Typically, a wrapper crawler may be used. However, the present invention is not limited thereto, and general clorers, focused crawlers, topical crawlers, and the like can be widely used.

At this time, analyzing the structure of the collection target web site in advance helps the web crawler to accurately collect the text sample.

The keyword extracting unit 14 extracts at least one keyword from a plurality of text samples.

In this case, the keyword for measuring the personality of the brand is suitable as an adjective, which is a part of speech that can express the personality of the brand.

Accordingly, the keyword extracting unit 14 may further include a morpheme analysis unit 140 for analyzing morphemes for each of the text samples, and may extract an adjective among the morphemes included in the text sample as keywords for the text sample.

The evaluating unit 16 evaluates the personality of the brand based on the at least one extracted keyword and the analogy of the measurement item and the measurement item included in the extended measurement item group.

Specifically, the evaluating unit 16 evaluates the individuality of the brand by collating the expanded measurement item group with the keyword and calculating the frequency of use for each measurement item.

On the other hand, the extended measurement item group generation unit 10 can prevent the previously designated vocabulary from being included in the extended measurement item group as a synonym for the evaluation item. This is to exclude vocabulary that is judged to be inadequate as a measurement item to measure brand personality by having various meanings.

In addition, the extended metric group generation unit 10 uses the hierarchical lexical semantic network 110 to generate the "extended metric group" so as to include the measurement items up to the synonyms of the metric.

In the embodiment of FIG. 1, the hierarchical lexicon semantic network 110 existing outside the brand personality measurement apparatus 1 is connected through a network such as the Internet.

As described above, the hierarchical lexical semantic network 110 existing outside the apparatus 1 may utilize a public database such as a word net (http://wordnet.princeton.edu/).

Meanwhile, embodiments in which the hierarchical lexicon semantic network is integrated and utilized in the apparatus 1 are also possible.

2 is a block diagram showing another example of the brand personality measurement apparatus.

The embodiment of FIG. 1 uses a hierarchical lexical semantic network 110 existing outside the brand personality measurement apparatus 1 to be connected via a network such as the Internet, whereas in the embodiment of FIG. 2, the hierarchical lexical semantic network 112 is provided in the inside of the apparatus 1. [

3 is a flowchart showing an example of a method for measuring the brand personality.

In order to measure a personality of a specific brand, an expanded group of measurement items is generated for each measurement item by extracting synonyms about the measurement items from the hierarchical lexical semantic network for each of a plurality of measurement items (300).

The generated group of measurement items can be stored in the storage unit 100 connected to the extended measurement item generation unit 10. [

On the other hand, it may be further determined whether or not a predetermined vocabulary is included as a synonym in the expanded metric group, and if so, excluded from the group. For example, words such as "good" have a variety of meanings, and therefore, they are judged to be inappropriate adjectives for evaluating the personality of a brand.

 At the same time, a step of collecting a plurality of text samples for evaluating a particular brand is performed (31). It has been previously described that the sample collecting unit 12 of the apparatus 1 can collect a plurality of text samples 200 related to a specific brand through the network 20 such as the Internet using a web crawler.

Once the plurality of text samples have been collected, at least one keyword is extracted from the plurality of text samples (320).

The extracted keywords are data for evaluating the personality of the brand, together with the measurement items (or synonyms of the measurement items) included in the extended measurement item group.

To this end, in step 320, a morpheme is analyzed for each of the text samples, and an adjective among the morphemes included in the text sample is extracted as the keyword for the text sample.

The evaluation factor for evaluating the personality of a particular brand is that the part -

In the step 330 of evaluating the personality of the brand, by comparing the expanded measurement item group with the extracted keywords, the frequency of use of the keywords included in the extended measurement item group related to each measurement item is calculated, Personality.

In order to measure the brand personality, the details of the measurement items and the collection of the samples to be compared with the measurement items are described in more detail below.

1. Measurement Items and Measurement Method

Restaurant brand personality measurement system can use Content Analysis which extends Aaker's brand personality measurement items to WordNet and substitutes it into collected data using web crawler.

The content analysis is a method of observing and measuring communication contents when the researcher is not able to directly access the survey subject, so as to analyze other data left by the respondent in order to reconfirm the data even though they collected the data through questionnaires or interviews This is a useful method in some cases.

To do this, the actual frequency of use of the existing metrics was measured using the five factors (Factor), Facet Name, and the actual use of the 42 traits in contrast to the adjectives extracted from the reviews collected through the Web crawler Calculate the frequency.

Expansion of measurement items is done by inputting brand personality measurement items into WordNet and extracting words (synonyms) similar to each measurement item.

The actual usage frequency of the expanded metrics is calculated by comparing the metric items extended through WordNet with the adjectives extracted from the review.

2. Sampling and Collection Methods

It can collect and analyze using the data generated from the location-based service.

Key location-based services include yelp.com, a Web and app-based service listed on the Nasdaq in 2011, and zagat.com, a book and Internet-based service recently acquired by Google from Yale University in the 1970s, ), And foursquare.com, a popular mobile-based service.

In order to secure the validity and the representative of the survey, Yelp.com (http://www.yelp.com), which is the largest and most popular service, targets 62 Korean restaurants registered in the Chicago area, 497 Chinese restaurants, A total of 158 reviews per Japanese restaurant can be collected and used as a sample.

A web crawler can be used as a method of collecting samples.

The Web crawler is a dedicated program agent for automatically collecting data such as reviews, reviews, etc. on the Internet automatically. The range of data that can be collected varies depending on the type.

Types of web crawlers include general crawlers, focused crawlers, topical crawlers, and wrapper crawlers.

Wrapper-based web crawlers are often used to quickly and accurately collect web documents of a certain subject, such as satisfaction ratings among various web crawlers. At this time, the structure of the web site to be collected should be analyzed in advance so that the web crawler can access the data and collect the desired data accurately.

In FIG. 4, an example of the sample collection technique is summarized in the table.

3. Characteristics of the sample

Yelp.com users fill out e-mail, postal code, date of birth, etc. when signing up for a review. For Facebook users, ID can be linked without membership.

When a member writes a review, he or she can use the General Guidelines, Review Guidelines, Photo Guidelines, User Profile Guidelines, Talk Guidelines, (Messaging Guidelines), Event Guidelines (Event Guidelines), and Business Guidelines (Business Guidelines) are provided, and the spell check function is also provided to maintain a certain level of review quality.

Compared to Yelp.com users and US Internet users, Yelp.com users have a high percentage of women, a high percentage of women aged 18 to 44, a high percentage of children without children, and many households with income over $ 150,000. , College graduates, and many Asian and Hispanic students.

4. Analysis Results - Analysis of actual use of existing brand personality measurement items

Aaker (1997) collected 36,536 reviews through the web crawler to find out whether the brand personality measurement items used by actual consumers were used, extracted 1,216,494 words through a morpheme analyzer, And 329,239 adjectives, and prepared for 42 Aaker brand personality measurement items.

FIG. 5 is a table showing the results of analyzing the actual use of the existing brand personality measurement items.

Aaker's brand personality metrics were used 7,266 times in actual reviews.

The results of this study are summarized as follows: 1) The wholesome, sincere, down-to-earth, sentimental, up-to-date, independent, spirited, leader, secure, hardworking, intelligent, technical, glamorous, feminine, outdoorsy, masculine, daring, upperclass, goodlooking, small-town, family-oriented 2 times, imaginative 4 times, confident 6 times, corporate 8 times.

In the attribute, Upperclass, Down-to-earth 4 times, Intelligent 8 times, Successful, Reliable 18 times, Charming 36 times.

5. WordNet  Extension of existing brand personality metrics through

In order to expand the 42 brand personality measurement items, it is necessary to input synonyms (words of similar meaning) of brand personality measurement items by inputting them into online service of WordNet (http://wordnetweb.princeton.edu/perl/webwn) developed by Princeton University, As a result, 42 measurement items were extended to 197 adjectives.

6 is a table illustrating brand personality measurement items extended through WordNet.

The most expanded metrics through WordNet are sentimental (15), smooth (14), tough (12), spirited (12), and secure (11).

On the contrary, there are no family-oriented, small-town, wholesome, original, cheerful, exciting, successful, upperclass, goodlooking, outdoorsy, and masculine.

In the expansion process, when a word is included in several items, it is spirited, dependable, audacious, good, honest, reliable, rugged, smooth, true, and similar notation is levelheaded, level- Sure-footed is surefooted and integrated into one word to eliminate word duplication.

6. Analyze the actual usage of the expanded metric

The actual use analysis of brand personality items expanded through WordNet was analyzed in the same way as the analysis of actual usage frequency of existing items. A total of 197 brand - name personality metrics were used in the actual review 16,959 times.

First, 128 items (including 5 items) were used in the extended brand personality measurement items, and 70 items were used at least once. The most frequently used items were good (27,874), okay (5,552), bad (3,755), friendly (3,246), new (2,916), okay (2,555), authentic There were hard (1,056), cool (968), real (958), fine (788), bland (786), main (596), unique (588) and trendy Good, bad, new, and okay are words that have various meanings.

The results of this study are as follows. First, we analyze the composition ratio of the items, attributes, and factors to the measurement of the item. (99.30%), game (98.20%), inventive (94.70%), positive (94.20%), positive, corporate, earthy, upperclass, goodlooking, small-town, family- ) and authentic (92.20%) accounted for more than 90%. The attributes were 99.6% Ruggedness, 96.3% Tough and 80.6% Reliable.

Third, the distribution of factors, attributes, and measurement items of the extended brand personality were examined by Excitement (6,636, 41.3%), Sincerity (4,965, 30.6%), Competence (2,212, 14.5%), Ruggedness %) And Sophistication (1,037, 6.1%), respectively. In the attribute, Excitement occupied 70.8% of Spirited, but it was relatively well used in Up-to-date (10.5%), Imaginative (10.0%) and Daring (8.7%). Competence was also highly reliable (80.6% (14.2%) and Successful (5.2%), respectively. However, Sincerity was concentrated on Cheerful (68.0%) and Honest (28.7%), Wholesome (3.2%), Down-to-earth (0.2%) were less used, Ruggedness was concentrated on Tough (96.3% 3.7%) were less used, and sophistication was also concentrated on Charming (99.6%) and Upperclass (0.4%) was less.

7. Comparison of the explanatory power and frequency of use of brand personality measurement items

The results of comparing the actual frequency of use of the brand personality measurement items expanded through WordNet with the Variance Explained of the brand personality measurement items of the Aaker, the brand personality measurement items, Respectively.

The five variables of the brand personality measurement items proposed by Aaker were Sincerity (26.5%), Excitement (25.1%), Competence (17.5%), Sophistication (11.9%) and Ruggedness .

However, the frequency of using the items in the reviews of the actual people was found to be two factors of Sincerity (61.4%) and Excitement (34.5%).

Sincerity decreased from 61.4% to 30.6%, excitement from 34.5% to 41.3%, competence from 0.6% to 14.5%, and sophistication from the first to the second. 0.6% to 6.1%, and the ruggedness from 2.9% to 7.5%.

8. Conclusion

Big data analysis in the food service industry is being actively pursued through location based services. Business and academics are interested in big data because they are in big data.

In accordance with an embodiment of the present invention, we have examined how much existing measurement tools are used in everyday life using Big Data, expanded the measurement items using WordNet, and examined the actual usage status of them.

The results are summarized as follows.

First, customers who use restaurants do not use many of the words used in the existing measurement items. Of the 42 Aaker measurements, 17 items such as wholesome, sincere, and down-to-earth were not used in real life, and 8 items such as daring, upperclass, and goodlooking were rarely used. This shows that the measurement items that measure brand personality differ from those used in real life.

Second, as a result of expanding 42 items through WordNet, 31 items including sentimental, spirited, smooth, tough, and secure were expanded to 197 items. This shows that when adding or replacing the existing measurement items in the research process, it can be done through the WordNet vocabulary database based on the subjective judgment of the researcher or the words used by the general consumers instead of the opinions of the minority groups.

Third, extended measures measure brand personality better than existing ones. When the existing measurement items were used as they were, they were shifted to Sincerity and Excitement. However, when expanded through WordNet, the competence, sophistication, and ruggedness also increased in a balanced manner, showing the frequency of use similar to the explanatory power level proposed by Aaker. This shows that over time, metrics need to be added or changed to achieve the original purpose.

The results of this study are as follows: First, it is difficult to generalize the results of the research because of the subjective judgment of the researchers and the limitations of the sample on the measurement items (Aaker, 1997; Austin et al., 2003; Ryu & Jang, 2008) We have investigated how many consumers actually use the measurement items that have been used in many researches, and extended the measurement tools to near actual use.

Second, the fusion of interdisciplinarity is achieved by combining the measurement of brand personality, which has been studied extensively in the field of food service management, with the Big Data and WordNet, which are engineering technologies, to produce expanded results.

Third, we use big data generated through the Internet and mobile to measure the brand's perception of the consumer.

Claims (14)

An extended metric group generation unit for extracting synonyms of the measurement items from the hierarchical lexical semantic network for each of a plurality of measurement items for evaluating the personality of the brand and generating an expanded group of measurement items for each measurement item, ;
A sample collection unit for collecting a plurality of text samples related to the brand using a web crawler;
A keyword extracting unit for extracting at least one keyword from the plurality of text samples;
And an evaluation unit for evaluating the individuality of the brand based on the extracted at least one keyword and the extended metric group,
Wherein the keyword extracting unit further includes a morpheme analyzing unit for analyzing morphemes for each of the text samples,
And extracting an adjective among the morphemes included in the text sample as the keyword for the text sample . The method according to claim 1,
Wherein the web crawler is a wrapper-based web crawler. delete The method according to claim 1,
Wherein the extended metric group generation unit does not include a pre-designated vocabulary as a synonym of a plurality of metrics for evaluating the personality of the brand in the extended metric group. 5. The method of claim 4,
The pre-designated vocabulary is a vocabulary inappropriate for a measurement item because it has various meanings. The method according to claim 1,
Wherein the evaluation unit evaluates the individuality of the brand by collating the expanded measurement item group with the keyword and calculating the frequency of use for each measurement item. The method according to claim 1,
And a storage unit in which the measurement item group generating unit stores the plurality of measurement items and the extended measurement item group. The method according to claim 1,
Wherein the hierarchical lexical semantic network is WordNet. The method according to claim 1,
Wherein the plurality of measurement items are the 42 brand personality measurement items of the acker. Extracting synonyms of the measurement items from a hierarchical lexical semantic network for each of a plurality of measurement items for evaluating the individuality of a brand, and creating an expanded group of measurement items for each measurement item;
Collecting a plurality of text samples related to the brand using a web crawler;
Extracting at least one keyword from the plurality of text samples; And
And evaluating the personality of the brand based on the extracted at least one keyword and the extended metric group,
The extracting of the keyword may include analyzing a morpheme for each of the text samples; And
And extracting an adjective from the morpheme included in the text sample as the keyword for the text sample. 11. The method of claim 10,
Wherein the collecting step is performed by a wrapper-based web crawler. delete 11. The method of claim 10,
Determining whether a predefined vocabulary is included as a synonym in the extended metric group and excluding the predefined vocabulary from the group if the predefined vocabulary is included as a synonym in the step of creating the expanded metric group, Personality measurement method. 11. The method of claim 10,
Wherein the evaluating step evaluates the personality of the brand by collating the expanded group of measurement items with the keyword and calculating the frequency of use for each measurement item.

Images