KH Coder Research Articles

Elite poems and mass proverbs: exploring the historic landscape in Nanchang City, China

The rapid process of urbanisation has led to the erosion of the preservation of features of urban landscape. Research on historical landscapes in ancient urban planning has predominantly concentrated on the elite class, neglecting the broader population. Nanchang, a city with a history spanning over two millennia of urban development, is the focus of this study. The study integrates elite poems and mass proverbs through GIS spatial matching and KH Coder word frequency analysis to research the historic landscape in Nanchang City. The research uncovers that the elite class focuses on the historical, literary, and aesthetic aspects of Nanchang, while the masses emphasise urban structure, functionality, and aspirations. These dual perspectives underscore Nanchang’s evolution from inception to formalisation, highlighting its ability to cater to diverse societal needs. The efforts on the distribution of landscape spots and strict modulus reflect the adaptability and sustainability of Nanchang City.

Big Data Insights into Coastal Tourism: Analyzing Customer Satisfaction at Egyptian Red Sea Dive Resorts

This research aims to explore the relationship between customer satisfaction and various extracted factors at dive resorts in the Red Sea, Egypt, utilizing a hybrid methodology of descriptive and diagnostic analytics applied to online review data. Employing techniques such as KH coder for text analysis, exploratory factor analysis (EFA), and linear regression, this study systematically identifies key elements that influence customer satisfaction. Findings reveal that activities related to diving and marine life markedly enhance guest satisfaction, underscoring the critical role these aspects play in the overall appeal of Egyptian coastal tourism. Conversely, areas such as dining and amenities were identified as needing improvement. The originality of this study lies in its application of big data analytics to dissect and understand customer feedback in a sector-specific context, providing strategic insights for the sustainable advancement of coastal tourism in Egypt. By focusing on dive resorts, this research highlights their integral role in coastal tourism and offers a model for leveraging online customer reviews to enhance service quality and promote sustainable practices within the tourism industry, contributing to the overall growth and sustainability of coastal tourism.

Perceptions and responses to diseases among patients with inflammatory bowel disease: text-mining analysis of posts on a Japanese patient community website

Introduction: Patients with inflammatory bowel disease (IBD) are increasingly using online platforms to communicate with other patients and healthcare professionals seeking disease-related information and support. Free-text posts on these platforms could provide insights into patients’ everyday lives, which could help to improve patient care. In this proof-of-concept study, we applied text-mining to extract patient needs from free-text posts on a community forum in Japan, and holistically visualized the patients’ perceptions and their connections, and explored the patient characteristic-dependent trends in the use of words. Methods: Free-text posts written between May 11, 2020 and May 31, 2022 on the community forum were retrieved and subjected to text-mining analysis. Trends in the use of words were extracted from the posts for correspondence and co-occurrence network analyses using KH Coder open-source text-mining software. Results: Seventy-four posts were analyzed. Using text mining methods, we successfully extracted and visualized a variety of patient concerns and their connections. The correspondence and co-occurrence analyses revealed patient segment-dependent trends in the use of words. For example, patients with a disease duration of ≤5 yrs were more likely to use words related to emotions or their desire to change or quit their job, such as ‘anxiety’ and ‘resignation.’ Patients with a disease duration of >10 yrs were more likely to use words showing that they are finding ways to live with or accept their disease, and are getting used to the lifestyle but some patients continued to experience worsening disease. Conclusions: We found that free texts posted on an IBD community forum can be a useful source of information to capture the wide variety of thoughts of patients. Text mining procedures can help to visualize the relative importance of the topics identified from free text posts. Our findings of this proof-of-concept study will be useful for generating new hypotheses to better understand and address the needs of patients with IBD.

A case study on exploring the motivation of English-major EFL students in online exchange programs

The COVID-19 pandemic has had a profound impact on the education sector worldwide, prompting a swift transition to technology-driven learning solutions. One notable response to this challenge has been the introduction of online versions of study-abroad and international exchange programs. Considering the limited existing research on how EFL university students perceive online international exchange programs, this study investigated the motivation of EFL students toward such programs initiated in response to the cancellation of in-person exchanges. The survey among 84 students from a private women's university in Kansai in August 2021 revealed that an overwhelming majority (93%) expressed their desire to participate in online international exchange programs. This study gained further insights by employing the KH Coder text mining software (Higuchi, 2016) to analyze the free-text responses obtained from the questionnaire. The analysis of the responses unveiled their recognition of the distinct advantages offered by online exchange programs. Approximately half of the respondents indicated their willingness to engage in non-credit programs, while a similar percentage conveyed that academic credits could enhance their motivation to participate. Notably, the analysis of the free text revealed that the latter group's seemingly extrinsic motivation was largely self-determined, with their primary goal being program participation. These findings underscore the potential for online international exchanges to be well-received by students, particularly when academic credits are awarded. Additionally, correspondence analysis highlights that the desire to earn academic credits is a topic of interest across all school year levels.

A quantitative content analysis of topical characteristics of the online COVID-19 infodemic in the United States and Japan

BackgroundThe COVID-19 pandemic has spurred the growth of a global infodemic. In order to combat the COVID-19 infodemic, it is necessary to understand what kinds of misinformation are spreading. Furthermore, various local factors influence how the infodemic manifests in different countries. Therefore, understanding how and why infodemics differ between countries is a matter of interest for public health. This study aims to elucidate and compare the types of COVID-19 misinformation produced from the infodemic in the US and Japan.MethodsCOVID-19 fact-checking articles were obtained from the two largest publishers of fact-checking articles in each language. 1,743 US articles and 148 Japanese articles in their respective languages were gathered, with articles published between 23 January 2020 and 4 November 2022. Articles were analyzed using the free text mining software KH Coder. Exploration of frequently-occurring words and groups of related words was carried out. Based on agglomeration plots and prior research, eight categories of misinformation were created. Lastly, coding rules were created for these eight categories, and a chi-squared test was performed to compare the two datasets.ResultsOverall, the most frequent words in both languages were related to health-related terms, but the Japan dataset had more words referring to foreign countries. Among the eight categories, differences with chi-squared p ≤ 0.01 were found after Holm-Bonferroni p value adjustment for the proportions of misinformation regarding statistics (US 40.0% vs. JP 25.7%, ϕ 0.0792); origin of the virus and resultant discrimination (US 7.0% vs. JP 20.3%, ϕ 0.1311); and COVID-19 disease severity, treatment, or testing (US 32.6% vs. JP 45.9%, ϕ 0.0756).ConclusionsLocal contextual factors were found that likely influenced the infodemic in both countries; representations of these factors include societal polarization in the US and the HPV vaccine scare in Japan. It is possible that Japan’s relative resistance to misinformation affects the kinds of misinformation consumed, directing attention away from conspiracy theories and towards health-related issues. However, more studies need to be done to verify whether misinformation resistance affects misinformation consumption patterns this way.

Flat Recovery Process in Mental Health: A Qualitative Analysis of Autoethnography

This study aims to provide a new perspective on the concept of recovery by qualitatively analyzing the experiences of mentally ill people from the Japanese perspective through autoethnography. Traditionally, the recovery process has been viewed as a series of “ups and downs” that may not apply to individuals with chronic mental health issues. A qualitative autoethnography analysis used personal diary entries from 20XX through 20XX + 4. Using text mining with KH Coder and cluster analysis, I first classified and visualized the data from the diary content and then, after confirming the content chronologically, examined the validity of the concept of the recovery process based on the theoretical framework of the Japanese cultural perspective. External validity was ensured through triangulation with the author’s work. Coding revealed four to six groups of mental health recovery keywords for each year. Confirmation of the results according to the life history also revealed problems with the conventional recovery process concept and advantages of the flat recovery process based on the Japanese cultural perspective.

A Comprehensive Analysis of the per- and poly-fluoroalkyl substances (PFAS) research landscape through AI-assisted text mining

Per- and poly-fluoroalkyl substances (PFAS) have been widely used in various industrial applications due to their unique properties. This study aims to provide a comprehensive analysis of PFAS research trends using a novel approach combining text mining techniques and large-scale language models (LLMs). PFAS-related scientific literature published from 1980 to 2024 was gathered from Scopus, and KH Coder and Claude 3 were used to perform the analysis. The results showed a significant increase in research output and a clear shift in research topics over the past 40 years. Whereas in the past, the focus was on analytical methods, more recently, the emphasis has been on environmental fate, toxicity assessment, alternative compounds, and regulation. With Claude 3, research areas can now be identified without reviewing the results of expert text mining. Comparisons of AI-extracted trends with insights from traditional review articles showed strong agreement, confirming the effectiveness of this approach. These findings suggest the need for continued interdisciplinary research on PFAS such as the development of remediation strategies, elucidation of health effects, and evidence-based policymaking. This study showed the possibility of integrating text mining and LLM for a comprehensive analysis of research trends, which will accelerate future research and development strategies.

Artificial intelligence literacy: a proposed faceted taxonomy

PurposeThe purpose of this paper is to propose a taxonomy of artificial intelligence (AI) literacy to support AI literacy education and research.Design/methodology/approachThis study makes use of the facet analysis technique and draws upon various sources of data and information to develop a taxonomy of AI literacy. The research consists of the following key steps: a comprehensive review of the literature published on AI literacy research, an examination of well-known AI classification schemes and taxonomies, a review of prior research on data/information/digital literacy research and a qualitative and quantitative analysis of 1,031 metadata records on AI literacy publications. The KH Coder 3 software application was used to analyse metadata records from the Scopus multidisciplinary database.FindingsA new taxonomy of AI literacy is proposed with 13 high-level facets and a list of specific subjects for each facet.Research limitations/implicationsThe proposed taxonomy may serve as a conceptual AI literacy framework to support the critical understanding, use, application and examination of AI-enhanced tools and technologies in various educational and organizational contexts.Practical implicationsThe proposed taxonomy provides a knowledge organization and knowledge mapping structure to support curriculum development and the organization of digital information.Social implicationsThe proposed taxonomy provides a cross-disciplinary perspective of AI literacy. It can be used, adapted, modified or enhanced to accommodate education and learning opportunities and curricula in different domains, disciplines and subject areas.Originality/valueThe proposed AI literacy taxonomy offers a new and original conceptual framework that builds on a variety of different sources of data and integrates literature from various disciplines, including computing, information science, education and literacy research.

Research topics in occupational medicine, 1990-2022: A text-mining-applied bibliometric study.

Occupational health has been influenced by societal and industrial changes. This study aimed to clarify topic trends in occupational health research in 1990-2022. We conducted a text-mining-adjusted bibliometric study using research titles in occupational health. Data on research titles and the years of publication were collected from 26 peer-reviewed journals on PubMed. Using morphological and correspondence analyses in text mining, we structured research topics into multiple categories and visualized the relationship between all categories and publication years. Statistical analyses were conducted using the text mining software - KH Coder 3.0. We obtained 48 645 articles containing 714 890 words in their titles. The research topics were classified into 4 categories and 17 subcategories, of which those of occupations; countries; non-intervention; psychosocial factors; lifestyle factors; safety; symptoms; therapy and care; and productivity have recently shown an increasing trend. In contrast, the subcategories of risk, chemical factors, disease, and organ damage showed decreasing trends. Chemical factors, which were the main topics in the 1990s, included risk, organ damage, and disease. Productivity, the main topic in the 2020s, co-occurred with lifestyle factors, symptoms, and intervention. Focal areas in occupational-health research shift according to societal trends. Occupational-health research has primarily analyzed issues in developed countries with capitalist values and may not have sufficiently covered issues in developing countries. It is imperative for policymakers and public funding bodies to determine priorities for investigation in the field.

Transferability of Effective Consultation Skills for Decision-Making Support in the Voluntary Surrender of Older Adult Drivers' Licences in Super-Aged Japan.

Previously, we identified eight effective consultation skills to support decision-making in the voluntary surrender of older adult drivers' licences in super-aged Japan. This study aimed to clarify the transferability of these skills. We collected text data by interviewing 11 safe-driving counsellors (four police officers, four clerical staff and three nurses) in the License Division of the National Police Agency from February to March 2022. Interviews were semi-structured and conducted by telephone or email. During the interview, participants were asked to recall their experiences as counsellors providing decision-making support to older drivers and to compare their experiences with the eight consultation skills. We analysed the content of the responses by quantitative text analysis with KH Coder 3 software. As a characteristic of the words and phrases used by counsellors in their narratives about consultation skills, the most frequently extracted words from among 3147 words were think, parties and family, and promote had the highest mediation centrality. The eight subgraphs were 'Respecting the will of relevant parties from their standpoint is natural', 'Listening attentively and empathetically to relevant parties is effective', 'Presenting objective data to guide decisions is successful', 'Showing cognitive functioning test results is often effective', 'Counselors with medical expertise can elicit positive counseling outcomes', 'Intervention by medical or police counselors facilitates the decision to surrender voluntarily', 'Counseling skills need to be improved' and 'A diagram of the 8 skills is helpful for inexperienced counselors'. The results suggest that the eight consultation skills have similarities and are transferable. This transferability might contribute to practical application or cohort follow-up study research. These skills can be incorporated into counsellor training, and counsellors can be expected to use these skills in the future. Regardless of the safe-driving counsellor's years of experience, the skills can help them provide uniform and accurate support in decision-making regarding the voluntary surrender of older adult drivers' licences. These skills are a promising approach to help older adults lead safe and secure lives as they age.

Social role of the ‘Bow-Wow Patrol’ in urban areas of Japan: a qualitative study

The ‘Bow-Wow Patrol (BWP)’, established to avert local crime by encouraging dog walking, may help build social relationships among individuals through dog walking. However, details of its social influence remain unclear. Thus, this study aimed to elucidate the social roles of BWP in the urban areas of Japan. A total of 18 BWP organisation members from two Tokyo municipalities were recruited using snowball sampling between November 2021 and July 2022. In an interview, participants were asked about their perceptions of changes in their social relationships through the BWP. Qualitative content analysis was conducted after text mining using the KH Coder software. The mean age of the participants was 63.9 years, and 83.3% engaged in BWP 7 days a week. The content analysis revealed six social roles of the BWP: establishment of social networks with schools as the centre, establishment of loose networks inside and outside the BWP organization, enhancement of a sense of trust among neighbours, enhancement of a sense of trust in communities, norms of reciprocity among dog owners, and dog and owner characteristics. This study found that the BWP in urban Japan strengthens social cohesion and expands social networks among dog walkers, encouraging them to continue walking.

Exploring the components of “Kando” and the factors evoking it during motorcycle riding

This study focused on the physical activity of riding a motorcycle with a mechanical motor or engine and “kando” – a difficult concept to translate into English and explain to a Western audience. Kando is a ubiquitous Japanese expression used when people are emotionally moved. This study elucidated the components of kando regarding motorcycle riding and the factors that evoke it. Data were collected from 23 motorcyclists through online focus-group interviews and analyzed using the KH Coder. Six factors, “convenience,” “inconvenience,” “bodily,” “interaction,” “variety,” and “stimulation,” were identified as the unique aspects of motorcycle riding that evoke kando. Kando’s objects included “motorcycle as a vehicle,” “nature,” and “humans.” Kando was produced through the combination of the six unique factors and the three aforementioned objects; moreover, kando while riding a motorcycle comprised 10 factors, “delight,” “excitement,” “stimulation of the five senses,” “overcoming,” “achievement,” “warmth,” “sense of healing,” “exhilaration,” “sense of release,” and “sense of bonding.” Motorcyclists experience kando through motorcycle riding as a physical activity. The findings could contribute to further elucidating the term kando and motorcycle development.

Comparison of Virtual Learning Community Construction between China and the United States Based on KH Coder Text Data Mining

Virtual learning communities, as an innovative online education model, have gained significant attention worldwide. The virtual learning communities in the United States are characterized by advanced educational models, innovative technological applications, and rich practical experiences. Exploring and drawing lessons from the construction of virtual learning communities in the United States is of great value to China. This paper utilizes KH Coder as an analysis tool and conducts literature searches on the Web of Science for U.S. literature and on the CNKI database for Chinese literature. The analysis covers the period from 2019 to 2024, and includes frequency and network co-linearity analyses. The research findings indicate the following: First, Chinese online education research emphasizes the importance of technological application and innovation in the field of education. Second, Chinese online teaching research has certain limitations in terms of research scope. Third, there are certain issues in the dissemination of research findings in Chinese online teaching research. Based on the research results, the following adjustments should be made for Chinese virtual learning communities: First, the comparison of research literature between China and the United States shows that the development of virtual learning communities requires diverse collaborations. Second, the development of virtual learning communities requires continuous technological innovation. Third, the comparison of research literature between China and the United States reminds us that the development of virtual learning communities needs to focus on social interaction. Fourth, the development of virtual learning communities also requires teacher support and participation. Fifth, the development of virtual learning communities needs to encourage students' active participation and autonomous learning.

Towards a ‘synergy’ of text mining and critical discourse analysis: a corpus-assisted discourse study of imagining China in Hong Kong political discourse

Abstract This study proposes to incorporate text mining into critical discourse analysis (CDA) to give a corpus-assisted discourse study of the particular ways of imagining Hong Kong’s relations to China in the public speeches of three former Chief Executives in the two decades after its handover. With the computer-assisted text-mining tool KH Coder, this study combines the methods of quantitative text mining and qualitative discourse analysis to examine their preferential ways of imagining China at different levels of discourse: (1) topics/themes, (2) discursive strategies, and (3) linguistic means and realizations. It generates illuminating findings concerning their preferential ways of imagining Hong Kong’s relations to China as well as the analytic potential of incorporating text mining into CDA. It is expected that it can lead to more studies towards this endeavour.

A longitudinal qualitative study on Iranian university lecturers' identity reconstruction during the COVID-19 pandemic through text-mining: a dynamic approach

This research addressed the critical gap in understanding the procedure of professional identity reconstruction among university lecturers in Iran during the pandemic. Over the course of two academic semesters, this study utilized a dynamic theoretical framework to investigate the intricacies of how 41 Iranian university English teachers reconstructed their professional identities. Data were collected through a dynamic methodological framework, comprising: semi-structured interviews conducted monthly, focus group interviews held every 3 months, and participants' reflective journal entries collected every 2 weeks. Text mining techniques and content analysis, through KH Coder 3.0, were employed to explore the nuanced aspects of teachers' identity reconstruction, focusing on contextual elements, agency, and pedagogical practices. The use of KH Coder 3.0 software for word co-occurrence and word association analysis revealed that university lecturers reconstructed their professional identities in response to the challenges posed by two main factors: (1) adapting to new educational contexts and (2) overcoming difficulties in teacher–student communication caused by the abrupt shift from traditional in-person education to online schooling. Through their practical experiences, educators emerged as proactive problem solvers, receptive adapters, and confident professionals. These findings carry implications for educational institutions and teacher educators, emphasizing the importance of prioritizing pedagogical systems that encourage teachers to excel in hybrid teaching contexts, fostering proficiency in online communication and virtual teaching methods.

Computerized Content Analysis of TripAdvisor Reviews: A Case Study of a Street Food Restaurant With One Michelin Star

Online review platform is a digital communication for reviewers to communicate with future customers. TripAdvisor reviews of a Thai street food restaurant were examined in the periods of before (2012-2017) and after (2018-2023) being awarded one Michelin Star. Two digital tools were employed: Web Scraper and KH Coder. Positive aspects of the reviews were found. Most of the reviews rose from 4 bubble ratings at 42.53% during 2012-2017 to 5 bubble ratings at 42.43% during 2018-2023. Frequent words and co-occurrence network highlighted the signature dishes of the restaurant and quality of food as well as pinpointed the need for a table reservation and the waiting time. The findings are significant to the customer service experience, specifically pertaining to dining and table assistance. Future studies can expand to multiple online restaurant reviews or a nation’s cuisine using computerized content analysis.

Scenery deconstruction: a new approach to understanding the historical characteristics of Nanjing cultural landscape

The “Eight Scenic Views Paintings” represent crucial visual materials for investigating the history of cultural landscapes. However, traditional methods of interpreting materials struggle to discern the inherent connections between different landscape elements. This study proposes an approach for deconstructing historical images, taking the example of the Forty Scenic Views in the Late Ming Dynasty in Nanjing, China. To explore the co-occurrence structure, hierarchical clustering, and correlation features among various elements, various digital humanities quantification methods were applied, including spatial analysis of ArcGIS, co-occurrence and clustering of KH Coder, and correlation analysis of SPSS. This study reveals the characteristics of the landscape construction of Nanjing in the Late Ming: natural landscape as the foundation, artificial landscape as the core, and advocating tradition as the fashion. It also uncovers the landscape order: mountains, waters, and scenic views interweaved and coexisted, as well as nature and culture intertwined and clustered. In addition, multiple information graphs of the subordinate and co-occurrence relationships of 20 landscape elements were constructed, 5 landscape paradigms were extracted, and 36 pairs of related relationships were discovered, deepening the historical understanding of the urban landscape construction of Nanjing in the Late Ming. This paper puts forward the idea of analyzing historical images by digital method, which provides some essential and detailed historical basis for explaining the value of cultural landscape heritage and shaping contemporary urban landscape.

Mining Text Value in Japanese Translation Teaching Based on KH Coder Text Data

Abstract Vocabulary serves as the cornerstone of Japanese translation instruction, with the identification of high-frequency terms within educational texts yielding significant pedagogical insights. This study employs the KH-Coder software to extract feature words frequently used in Japanese translation teaching materials. Through a combination of covariate network analysis, sentiment analysis, and hierarchical clustering, the paper delves into the intrinsic textual value of these educational resources. Furthermore, it integrates both generative and discriminative approaches to conduct dependency parsing of Japanese translation. By incorporating probabilistic functions, the methodology enhances contextual understanding, ensuring that each annotation unit within a sentence is optimally tagged. Analysis of the CGTT-35 corpus through KH-Coder reveals a list of the top 50 high-frequency words in Japanese translation teaching, each appearing over 100 times. This enables students to use these frequent words and sentence structures as cues to master common phrases and familiarize themselves with core content. The approach has proven effective in enhancing students’ receptivity to Japanese translation instruction and their comprehension capabilities, offering valuable implications for teaching strategies.

Which vocabulary learning strategies are important and useful for Japanese university students? A text-mining approach

A growing body of vocabulary learning strategies (VLSs) studies focuses on accumulating empirical evidence on the relationship between frequency of VLS use and vocabulary acquisition by using the frequency-based measures of VLSs. Our study by contrast, examined which VLSs are perceived as important and useful, and how they are employed by Japanese university students learning English as a foreign language (N = 40). Further, we investigated whether university students’ perceptions and use of strategies vary between higher- (n = 16) and lower-proficiency groups (n = 24). To this end, we used the KH Coder software and produced co-occurrence networks to categorize the words extracted from the results of an open-ended questionnaire. The results revealed that Japanese university students perceived cognitive strategies related to word form, sound, association learning, and metacognitive strategies as important and useful, and employed them frequently. Results from the two proficiency groups showed that while the higher-proficiency group perceived the importance of metacognitive strategies, these tendencies were not observed in the lower-proficiency group. Given the possibility that learners’ proficiency influences the use of VLSs in the vocabulary learning process, considerations and suggestions for future VLSs studies and teaching are discussed.

Causes of Countermeasures Against Serious Accidents Caused by Health Hazards of Chemical Substances, Based on Cases of Occupational Accidents in Japan

The purpose of this study was to summarize the typical and specific causes and risk reduction measures of serious accidents in Japan caused by chemical substances in terms of the 10 types of human health hazards in the UN GHS (Globally Harmonized System of Classification and Labelling of Chemicals) classification. The list of "Cases of Major Accidents Caused by Chemical Substances" published on the "Safety in the Workplace Website" of the Ministry of Health, Labor and Welfare (MHLW) was linked to the "Results of GHS Classification by the Government" of the National Institute of Technology and Evaluation (NITE). Analysis was conducted by health hazards in order to obtain reference examples of measures taken against health hazards in response to the revision of the law on autonomous chemical substance management in 2022. Using the text mining tool KH Coder ver. Three using cluster analysis, we grouped causes and measures among serious disasters by health hazard, illustrated co-occurrence networks, and extracted typical examples of each in a co-occurrence network. Representative causes of and measures against occupational accidents caused by health hazardous properties were summarized from the extracted typical cases. Although few occupational accidents were caused by the health hazards of chemicals, when all health hazards were classified into acute toxicity, acute health hazard, and chronic health hazard, contact was a clear cause for many of the acute toxicity and acute health hazards, such as corrosiveness and sensitization. However, many occupational accidents were caused by the physical hazardous aspects of the chemical substances or by the safety aspects of the workplace. Causes of occupational accidents due to health hazards included unsafe behavior and unsafe conditions, or lack of understanding of the physicochemical properties of a substance and specific reactions such as mixing or hydrolysis of the substance. Typical risk reduction measures for health hazards included equipment to prevent human contact with health hazardous substances, systems to control unsafe behavior, promoting understanding of chemical reactions, and providing information about chemicals to all persons involved in testing, research, and subcontract work. The data of occupational accident cases of "death or more than 4 days lost from work" were used. Most of the cases were related to acute health hazards, but relatively few to chronic health hazards, and few occupational accidents were caused by health hazards. Most of the occupational accidents were caused by health hazards due to the physical hazards of chemical substances and inadequate workplace safety. In light of the above, it is necessary in autonomous chemical substance management to first take risk reduction measures for workplace safety and chemical physical hazards, then to prevent contact with acute toxicity and acute health hazards such as corrosiveness and sensitization, and, lastly, to take risk reduction measures for chronic health hazards, using allowable concentrations and controlled concentrations as indices.