Comparison of the Impressions of Chinese on Japanese, Chinese and Korean Dance Found by Text Mining ()
1. Introduction
1.1. Background
Dance is an art which utilizes various physical movements such as rhythmic movements and steps. It is possible for the performer to convey specific intentions and emotions to the viewers through these movements. Since dance can be used to communicate nonverbally, it can evoke sympathy and understanding of others to the viewers through the performer’s self-expression. Many studies have been conducted focusing on the emotions expressed and transmitted by dance.
1.2. Previous Studies: Different Impressions of Dances
Previous studies examined how dance can elicit the psychological and physical responses. According to the study by Sawada et al. (2003), dances expressing joy, grief, and anger had different characteristics of movement depending on the kinds of emotions, and viewers could distinguish the expressed emotions through viewing these dances. The relationship between emotional expression and characteristic movement was the focus of their study. In dance different types of movement elicit different emotional responses from viewers. Matsumoto (1987) proposed the seven kinds of dance movement based on emotion as the minimum units of body expression in dance. Matsumoto named each movement as “natural”, “happy”, “lonely”, “sharp”, “solemn”, “dynamic”, and “flowing” based on Japanese male students’ impressions towards dance movements. Shimizu et al. (2006) adopted the method called the “Evaluation Grid” which is used to discover attributes when evaluating beauty in dance. The result of the evaluation by Japanese males and females concluded that beauty is a result of good posture, balance and nimble movement.
Dancers can express their emotions and spectators can understand the expression through the dance movement because specific movements can be associated with specific emotions. It is important for the dancers to convey their emotions, beliefs, and values through body movement properly to spectators, but the quality of the dance performance matters as well. Dancers need to understand what information various dance movements and combinations convey to the spectators in order to facilitate non-verbal communication.
Previous research has revealed that dance has aspects of nonverbal communication. The characteristics of the traditional dances from each country can facilitate cross-cultural communication. The following studies can be cited on the dance of each of the three Asian countries. Masuko (1993) wrote the following of Japanese traditional dance: “While western temporal art flows in rhythm, Japanese temporal art stops rather than flows.”
According to Yao (2017), Chinese dance is an extension of the Chinese mindset which is reflected to the following three characteristics. The first is rounded movements like drawing a circle. Circles symbolise yin-and-yang as well as people’s cooperation and harmony with nature. Rounded movements not only express circles but also the above meaning and thought through dance. The second is being vivid with grace. Expression of life force and praise of life are important factors of Chinese traditional dance. The third is the combination of meanings and images. It is important to grasp not only body expression but also the meanings of the expression. Chinese traditional dance expresses people’s inner worlds and viewers can understand dancers’ feelings through appreciation of their body expression. The three characteristics of Chinese dance include movements that avoid straight lines. This choice of movement is used to symbolize the natural world, people’s harmony with nature, and the cooperative spirit of people. Qigong plays a major role in Chinese dance: energy, breath and sprit are incorporated into the movement. Chinese dance movement is to be enjoyed not just visually but also as symbolic expressions of the inner world.
Yoo (2007) said that Korean dance could be understood by its four unique qualities. The four unique qualities of Korean dance are “naturalness”, “circulation of ki energy”, “free dynamics and improvisation” and “sustained indirection.”
1.3. Previous Studies: Evaluation Method of Dance Works
Questionnaires using psychological scales have been mainly conducted for impression evaluation on dance works (Kinjo, 1973). Nonetheless, to understand the contents of thoughts and emotions generated by interacting with the work together with the details such as the transition, etc., it is crucial to refer to data in the forms of utterances and descriptions that reflect the viewers’ mind more deeply. However, the evaluation of works of art and expressions requires specialized knowledge, and it is difficult to share the evaluation contents between those within and outside the specialized circles. The text mining method brings a quantitative and objective analytical approach to evaluating such art, which will lead to the creation of knowledge that can be widely utilized regardless of the presence or absence of the specialized knowledge. When examining the psychological effects of traditional dance as a work of art, text mining from the viewers’ utterances and descriptions is expected to clarify the psychological processes generated from appreciating the dance in more detail.
Zheng et al. (2020) described a study using a text mining method on the impressions of Japanese, Chinese, and Korean dances for Japanese people. They analyzed the relationship between the characteristics of traditional dances in different countries and the psychological impacts that they have on the viewers. In the survey, Japanese male and female students were asked to watch the traditional dances and answer open-ended questions about their impressions. Based on the result of analyzing the open-ended questions using a text mining method, they showed that there are similarities and differences in the psychological effects of watching Japanese, Chinese, and South Korean traditional dances.
1.4. Purpose
Since the study on Japanese, Chinese, and Korean dances reported by Zheng et al. (2020) was aimed at Japanese people, when trying to extract the unique characteristics of each country’s dance, it is difficult to count in factors such as the nationality and cultural background of the viewers when the target viewers are only from one country, which was Japan. Therefore, it is desirable to collect and compare the data from each country’s viewers. This study collected data from Chinese people using open-ended questions in Chinese and compared them with the results obtained from the Japanese people to further examine the characteristics of each country’s dance.
2. Method
2.1. Participants
The participants of the study were 29 Chinese people (average age 29.4 years old, SD = 9.3, 13 men and 16 women). The participants were asked to search independently online for dances from Japan, China, and South Korea. They were then to complete a free-form questionnaire about their impressions and thoughts in Chinese. This survey was conducted from June to July 2020.
2.2. Method
In this study, the questionnaires were written in Chinese. The text-mining analysis was conducted using the free software KH Coder version 3 ( Higuchi, 2016, 2017). KH Coder is utilized for quantitative content analysis or text mining. The target languages of KH Coder are Japanese, English, Chinese (simplified, UTF-8), and other languages. We used co-occurrence analysis, self-organizing map (SOM), hierarchical clustering, multi-dimensional scaling (MDS), and correspondence analysis for content analysis.
In the co-occurrence network image, the extracted words are written inside of circles. The extracted words with interrelationships are connected by a line. The larger circles represent higher frequency words. The thicker lines represent words that appear more often together. In the case that the words are not connected by a line, there is no co-occurrence relation between the words even if the words are close to each other.
The SOM is a neural network proposed by Kohonen (1995), which can map input data to a two-dimensional plane by unsupervised learning. Similar input samples are automatically mapped closely on the SOM grid. Therefore, in the SOM of KH Coder, the words placed close to each other can be thought to have similar meanings.
Cluster analysis is a method for classification of various information by grouping information which have common characteristics. Data is classified by this method with distances between those belonging to the same cluster being the shortest while those between different clusters being the longest. The hierarchical cluster analysis of KH Coder displays clustering through a dendrogram in which the extracted words which have similar occurrence patterns form a cluster.
The MDS is a method for expression of relationships between words in a low-dimensional space by placing similar words near each other and different words far from each other. The MDS of KH Coder expresses relationships between the extracted words on a two-dimensional plane based on similarities of the occurrence patterns of the extracted words.
Correspondence analysis is a method for expression of relationships between two kinds of data such as survey items and attributes of participants on a two-dimensional plane. The correspondence analysis of KH Coder visualizes relationships between the extracted words on the two-dimensional plane based on occurrence frequency of each extracted word in each text.
3. Results
3.1. Regarding Impressions and Thoughts on Dance of Respective Countries
The top 10 most frequently-used words in impressions and thoughts on the traditional dances of various countries are listed in Table 1.
In the following figures, the appearance frequency of words in the descriptions about Japanese dances, Chinese dances and Korean dances are visualized with co-occurrence network, MDS, Correspondence analysis, and SOM, respectively. All of the extracted words in Chinese language appeared in Figures 1-12
![]()
Table 1. Impression and thought of dances in each country.
*JJ means Adjective (phrase) adnominal modifier usage.
are translated to English language (See Tables A1-A3 in Appendix part of this paper).
3.2. Impressions and Thoughts on Japanese Dances
In examining the results of the co-occurrence network of Japanese dance, relationships can be observed in the words “movement,” “facial expression,” “calmness,” and “slow.” Also, there is a relationship between the words “makeup” and “props (Figure 1).” In examining the MDS, the words “calmness,” “tempo,” and
![]()
Figure 1. Co-occurrence network of words in descriptions about Japanese dances.
![]()
Figure 2. MDS of words in descriptions about Japanese dances.
“slow” are in proximity, “classical,” “hand fan,” and “culture” are in proximity, and “person,” “quiet,” and “elegance” are in proximity (Figure 2). In examining the results of the correspondence analysis, the words “elegance,” “quiet,” and “slowness” are in proximity, the words “combination”, “Kimono”, and “classic” are in proximity, and the words “calm” and “visible” are in close proximity (Figure 3). In examining the SOM, the words “quiet”, “movement”, and “slowness” are located in the lower right corner, and the words “slow” and “tempo” are located in the central area of the map (Figure 4).
![]()
Figure 3. Correspondence analysis of words in descriptions about Japanese dances.
![]()
Figure 4. SOM of words in descriptions about Japanese dances.
3.3. Impressions and Thoughts of Chinese Dances
In examining the results of the co-occurrence network of Chinese dance, relationships can be observed in words “diversity,” “color,” “gorgeous,” “music,” and “dress.” Also, there is a relationship between the words “dance,” “nationality,” and “tradition (Figure 5).” In examining the MDS, the words “dance,” “nationality,” and “China” are in proximity, “dress,” “gorgeous,” and “tradition” are in proximity (Figure 6). In examining the results of the correspondence analysis,
![]()
Figure 5. Co-occurrence network of words in descriptions about Chinese dances.
![]()
Figure 6. MDS of words in descriptions about Chinese dances.
the words “dance,” “movement,” and “turn” are in proximity, and the words “color,” “dress,” “diversity,” and “gorgeous” are in close proximity (Figure 7). In examining the SOM, the words “color,” “diversity,” and “dress” are located in the lower right corner, and the words “gorgeous,” and “comparison,” are located in the lower left corner of the map, the words “turn,” “many,” and “music,” are located in the upper right corner of the map (Figure 8).
![]()
Figure 7. Correspondence analysis of words in descriptions about Chinese dances.
![]()
Figure 8. SOM of words in descriptions about Chinese dances.
3.4. Impressions and Thoughts on Korean Dances
In examining the results of the co-occurrence network of Korean dance, relationships can be observed in the words “tempo,” “fast,” “music,” and “group (Figure 9).” Also, there is a relationship between the words “vivid,” “clothes,” “movement,” and “actors.” In examining the MDS, the words “dance,” “group,” “whole,” are in proximity, “classical,” as well “group dance” and “many” are in proximity (Figure 10). In examining the results of the correspondence analysis,
![]()
Figure 9. Co-occurrence network of words in descriptions about Korean dances.
![]()
Figure 10. MDS of words in descriptions about Korean dances.
the words “fast,” “music,” “group,” and “tempo” are in proximity, and the words “dance,” “movement,” “whole,” and “dress” are in close proximity (Figure 11). In examining the SOM, the words “tempo,” “group,” and “music” are located in the upper-left corner, and the words “vivid,” “clothes,” and “actors” are located in the lower central area of the map (Figure 12).
![]()
Figure 11. Correspondence analysis of words in descriptions about Korean dances.
![]()
Figure 12. SOM of words in descriptions about Korean dances.
4. Impressions of Japanese, Chinese, and Korean Dance: Discussion Based on the Results of Analysis
The characteristics in the description of each country’s dance are discussed based on the results of the co-occurrence network and MDS, correspondence analysis, and SOM.
With Japanese dance, in the co-occurrence network, the words “movement,” “facial expression,” “calmness,” and “slowly” are indicated to be related, and in the MDS, the words “calmness,” “tempo,” and “slow” are in proximity.
From this result, the characteristics of Japanese dance of slow movements, tempo, and facial expressions are understood to present a quiet impression.
In another proximity group found through the MDS, the words “person,” “quiet,” and “elegance” are in proximity. In the correspondence analysis, the words “elegance,” “quiet,” and “slowness,” are in proximity, therefore Japanese dance is understood to present not only slow motion and quiet facial expressions but also to contain a sense of elegance.
In particular, “slow” and “sluggish” are the impressions that Japanese participants also described in the study by Zheng et al. (2020), indicating a point of agreement. This study targeting Chinese participants showed relationships between the words “facial expression,” movement and impressions of Japanese dance, which is different from the previous study. Japanese dance has a feature of not showing excessive facial expression ( Koga, 2012), which could influence the Chinese participants’ impressions of Japanese dance.
In the co-occurrence network, the words “makeup” and “props” are also observed to be related, and in the MDS, “classical,” “hand fan,” and “culture” are in proximity. In the correspondence analysis, the words “blend,” “Kimono,” and “classic” are in close proximity. The Japanese style, Japanese culture and tradition can be sensed in clothes and props.
In Chinese dance, within the co-occurrence network the words “diversity,” “color,” “gorgeous,” “music,” and “dress” are related and similarly in the MDS, correspondence analysis, SOM, etc. the words “gorgeous” and “clothes” are in proximity. Thus, in Chinese dance, it is considered that the colorful clothes and the nature of the music contribute to a luxurious impression that exists especially in Chinese dance. In the study of Zheng et al. (2020), the word “splendor” was also mentioned frequently, which is consistent with the Japanese participants.
Next, focusing on other words, in the co-occurrence network, relationships are observed in the words “dance,” “nationality,” and “tradition,” and in the MDS, the words “dance,” “nationality,” and “China” are in proximity, thus it can be understood that nationality and tradition can be perceived in the dance of their own country. In correspondence analysis, the words “dance,” “movement,” and “turn” were in proximity, and the words “turn,” “many,” and “music” were also grouped in the SOM, concluding it is conceivable that numerous turning motions are a feature of Chinese dance.
As mentioned above, according to Yao (2017), one of the characteristics of Chinese dance is “a rounded movement that seems to draw an imaginary circle”. The appearance of words such as “movement” and “turning” in Chinese dance seems to indicate that such movement is more frequent in Chinese dance than in the dances of the other countries.
The previous study of Zheng et al. (2020) showed relationships between movement of Chinese dance and an intense impression and emotion. In contrast, this study showed that there was no relationship between the emotional impressions and the movement of Chinese dance. The Chinese participants were presumed to have difficulty having the intense impression of the familiar dance of their country and associating the dance with specific emotion.
Regarding Korean dance, in the co-occurrence network, the words “vivid,” “clothes,” “movement,” and “actors” are related and also in the SOM, the words “vivid,” “clothes,” and “actors” are grouped, and so it is easy to present the viewers with an impression of “vividness” that is different from the gorgeous of Chinese dance. Focusing on other words in the co-occurrence network, the words “tempo,” “fast,” “music,” and “group” are related, and other analysis demonstrated in the MDS, “dance,” “group,” “whole,” “classical,” “group dance” and “many” are in proximity. In correspondence analysis, words such as “fast,” “music,” “group,” and “tempo” are in proximity. In examining the SOM, the words “tempo,” “group,” and “music” are located in the upper left. From this, it can be considered a characteristic of Korean dance is the fast tempo and group-oriented.
5. General Discussion
In observing the characteristics of each country’s dance, Japanese dance has slow movements with facial expressions maintaining calmness and an elegant image. Chinese dance can be said to make Chinese people perceive their culture and traditions and, compared to other countries, based on the characteristics of the costume it has a luxurious image. Korean dance is group dancing, has a fast tempo, and based on the characteristics of the costumes, a vivid image is considered to be presented to the viewers.
Regarding Chinese and Korean dances, the study on the impressions of Japanese, Chinese, and Korean dances by Zheng et al. (2020) with Japanese people as the participants, concluded that Japanese dance presents a peaceful and slow image. Chinese dance is said to present a gorgeous and dynamic image, and Korean dance presents a vivid image due to the characteristics of the costumes, which is almost entirely consistent with the result at this time. Therefore, from the results of this study, it was found that, beyond the nationality and cultural background of the viewers, Japanese, Chinese, and Korean dances have characteristics that allow viewers to have similar impressions.
In the study by Zheng et al. (2020), the Japanese participants had an impression of “intense” for the Chinese dance, but the word “intense” was not found in the impressions of Chinese participants. It is possible that the reason behind it is that the participants in each country are more accustomed to the dance of their own country to some extent, which became the basis for their judgment. In particular, the Japanese might have felt that Chinese dance movement was “intense” because they compared it with the movement and tempo of Japanese dance.
Based on our results, we have shown that there are shared impressions expressed by the dance of each country to participants of different nationalities. At the same time, it also became clear that participants with different nationalities could also have different impressions.
From this result, it was suggested that investigating the impression of traditional performing arts such as dance may lead to understanding of different cultures.
Appendix
![]()
![]()
Table A1. Chinese-English Translation of Extracted Words about Japanese Dances for Figure 1 to Figure 4. Here “〇” indicates that the word appears in the figure.
![]()
![]()
Table A2. Chinese-English Translation of Extracted Words about Chinese Dances for Figure 5 to Figure 8. Here “〇” indicates that the word appears in the figure.
![]()
Table A3. Chinese-English Translation of Extracted Words about Korean Dances for Figure 9 to Figure 12. Here “〇” indicates that the word appears in the figure.