Hostname: page-component-6bf8c574d5-t27h7 Total loading time: 0 Render date: 2025-02-21T13:37:05.678Z Has data issue: false hasContentIssue false

Analysis of National Cybersecurity Strategies of G20: objectives, latent themes, latest trends and comparisons

Published online by Cambridge University Press:  18 February 2025

Hasan Çifci
Affiliation:
İstanbul Aydın Üniversitesi, Küçükçekmece/İstanbul, Türkiye
Esma Ergüner*
Affiliation:
Başkent Üniversitesi, Etimesgut/Ankara, Türkiye
*
Corresponding author: Esma Ergüner; Email: [email protected]

Abstract

This study analyzes National Cyber Security Strategies (NCSSs) of G20 countries through a novel combination of qualitative and quantitative methodologies. It focuses on delineating the shared objectives, distinct priorities, latent themes, and key priorities within the NCSSs. Latent dirichlet allocation topic modeling technique was used to identify implicit themes in the NCSSs to augment the explicitly articulated strategies. By exploring the latest versions of NCSS documents, the research uncovers a detailed panorama of multinational cybersecurity dynamics, offering insights into the complexities of shared and unique national cybersecurity challenges. Although challenged by the translation of non-English documents and the intrinsic limitations of topic modeling, the study significantly contributes to the cybersecurity policy domain, suggesting directions for future research to broaden the analytical scope and incorporate more diverse national contexts. In essence, this research underscores the indispensability of a multifaceted, analytical approach in understanding and devising NCSSs, vital for navigating the complex, and ever-changing digital threat environment.

Type
Research Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2025. Published by Cambridge University Press

Policy Significance Statement

This research presents valuable insights and vision for policymakers by examining G20 nations’ National Cyber Security Strategiesusing a combination of qualitative and quantitative analysis. The study provides a broad understanding of cybersecurity dynamics of developed countries through a detailed analysis of shared goals, distinct focus areas, and underlying themes in their cybersecurity strategy documents. By using topic-modeling algorithm to uncover implicit focus items and themes, the findings reveal unstated policy statements to show the broader strategic landscape. Policymakers can use this analysis to address both common and country-specific cybersecurity challenges by considering widely adopted strategies in the developed states. Such an effort by policymakers will contribute international cooperation and policy alignment in cybersecurity domain.

1. Introduction

The study of National Cyber Security Strategies (NCSSs) has attracted significant academic attention due to their influential role in shaping national cybersecurity policies and practices. Studies related to NCSSs can be classified as (i) country-based studies that produce outputs contributing to the development, revision, or in-depth examination of their own NCSSs (Montasari, Reference Montasari and Montasari2023), and (ii) studies focused on the comparative analysis of specific countries’ NCSSs to identify differences and similarities (Iova and Watashiba, Reference Iova and Watashiba2023). The countries involved in such studies can be members of any union (EU, NATO, BRICS, etc.) or can often be selected countries that are thought to have more developed NCSSs (Ovchinnikova and Upadhyay, Reference Ovchinnikova and Upadhyay2023; Štitilis et al., Reference Štitilis, Pakutinskas and Malinauskaitė2017; Newmeyer, Reference Newmeyer2015; Sabillon et al., Reference Sabillon, Cavaller and Cano2016; Shafqat and Masood, Reference Shafqat and Masood2016; ITU, 2018a, 2018b; OECD, 2012; Luiijf et al., Reference Luiijf, Besseling and Graaf2013).

We focused on the NCSSs of the G20 countries since they represent the majority of the world’s economy and population. Additionally, the diversity within the G20 in terms of information technology infrastructure and cybersecurity policies adds significant values for strategy analysis. Therefore, examining the G20’s cybersecurity policies offers important new perspectives on global trends, priorities, and prospective areas for future international collaboration.

To understand the cybersecurity strategies of G20 countries we tried to find the answers to the following research questions (RQ):

RQ1: What are the shared objectives and distinct priorities of G20 cybersecurity strategies?

RQ2: What are the latent themes in G20 cybersecurity strategies, as revealed by topic analysis, but not explicitly stated in the strategy documents?

RQ3: What are the predominant and common keywords in G20 cybersecurity strategies?

To answer these research questions, we used a mix of qualitative and quantitative approaches by utilizing expert visions and advanced analytical techniques. A sophisticated machine learning technique called topic modeling was used as quantitative analysis to uncover latent themes and patterns. Together with qualitative analysis, topic modeling provided a unique and objective way to interpret the complex nature of global cybersecurity strategies.

In the literature, there are studies comparing NCSSs of countries and international alliances. Most of the analysis is qualitative, that is primarily based on expert analysis while few using quantitative methods such as topic modeling. This study is the most recent analysis in the domain since it incorporates the newest versions of national cybersecurity strategies. Whereas previous studies mostly focused on discovering thematic similarities or common issues, our study goes further by revealing shared objectives, distinct priorities, latent themes, and new trends in the strategies.

The findings of the study present vital information for cybersecurity community about cybersecurity strategies to develop better-informed and globally aligned strategies. The comparative analysis of the G20 countries’ strategies also helps to improve awareness of common cybersecurity trends that have influence on future policies in this essential domain.

2. Literature review

Around the world, governments developed strategies to deal with problems in digital security, stop threats, reduce risks, prevent attackers, and make sure their digital systems are strong and safe to help progress in business, technology and society (ITU, 2018a; ENISA, 2016). Organizations such as International Telecommunication Union (ITU), European Union Agency for Cybersecurity (ENISA), Organization for Economic Co-operation and Development (OECD), NATO, and Oxford University have shared guidance resources to support nations in building good cybersecurity strategy and plans. Table 1 presents the most common cybersecurity indices and frameworks that can be used as guidance to develop NCSSs.

Table 1. Common National Level Cybersecurity indices and frameworks

National Cybersecurity Strategies are usually revised and updated about every four to six years (United Nations, 2024a). This periodic update process is necessary for adapting to the continually changing landscape of cyber threats and advances in technology. Given the natural cycle of renewal, our literature review began with studies published from 2018 onwards.

Azmi et al. (Reference Azmi, Tibben and Win2018) analyzed widespread cybersecurity frameworks using document analysis and identified shared actions, pillars, and lifecycle processes to present a general model for organizations and governments. Sunkph et al. (Reference Sunkph, Ramjan and Ottamakorn2018) examined the implementation of international cybersecurity policies in ASEAN (Association of Southeast Asian Nations) countries (Indonesia, Malaysia, the Philippines, Singapore, Thailand, and Vietnam) and highlighted the importance of regional collaboration to address escalating cyber threats. Gorka (Reference Gorka2018) analyzed the aims of the NCSSs of Visegrad Group (Czechia, Poland, Slovakia, and Hungary) to reveal similarities and differences among these states.

Kovacs (Reference Kovacs2019) analyzed the cyber security strategies of the EU, NATO, and four EU countries (Austria, Czechia, Hungary, and the UK) to determine possible elements and conclusions of the strategies. Baezner and Cordey (Reference Baezner and Cordey2019) compared the cybersecurity strategies of six European countries and identified eight shared challenges, including similarities and differences in their approaches.

Enescu (Reference Enescu2020) analyzed EU member states’ cybersecurity strategies and highlighted their alignment with EU initiatives and areas of divergence. Santisteban et al. (Reference Santisteban, Ocares and Andrade-Arenas2020) analyzed 20 NCSSs to determine the strategies of mostly targeted nations in the cyber domain. Handbook of International Cybersecurity (Tikk and Kerttunen, Reference Tikk, Kerttunen, Tikk and Kerttunen2020) presents various perspectives on international cybersecurity and covers cybersecurity concepts and frameworks, national and regional strategies, and global approaches to cybersecurity.

Jacuch’s study (Reference Jacuch2021) compared the cybersecurity strategy of Poland and those of five other nations: the UK, the US, France, Lithuania, and Estonia. He examined strategic documents from these countries to identify key similarities and differences to provide recommendations for strengthening Poland’s cybersecurity framework. In another study to audit national strategies, Sabillon (Reference Sabillon2021) examined the best practices of 10 major countries in terms of building effective cybersecurity strategies and policies.

The Organization of American States (OAS) (2022) analyzed the cybersecurity strategies of its 35 member states and compared approaches and common challenges. Falch et al. (Reference Falch, Olesen, Skouby, Tadayoni and Williams2023) focused on the Nordic-Baltic region and compared strategies in terms of threat assessments, risk management, and international cooperation. Odebade and Benkhelifa (Reference Odebade and Benkhelifa2023) conducted a comparative study of the NCSSs of 10 nations across Europe, Asia-Pacific, and North America. They found a common emphasis on protecting critical infrastructure to improve public-private partnerships. Ali et al. (Reference Ali, Razzaq, Abbass, Yousaf and Shan2024) use various analysis techniques for evaluating and improving cybersecurity posture of countries. The study compares the cybersecurity of underdeveloped countries (Pakistan) against developed countries (Lithuania, Estonia, Singapore, Spain, and Norway) and suggests frameworks and recommendations.

Topic modeling is increasingly being used for cybersecurity from malware detection to document analysis (Bechor and Jung, Reference Bechor and Jung2019; Samtani et al., Reference Samtani, Chinn, Larson and Chen2016). Adams et al. (Reference Adams, Carter, Fleming and Beling2018) utilized topic modeling to analyze textual descriptions of attack patterns and estimate topic distributions to help experts with attack evaluations. Bechor (Reference Bechor and Jung2019) explored the intersection of cybersecurity and data science by applying LDA topic modeling to scholarly articles published between 2012 and 2018. Ignaczak et al. (Reference Ignaczak, Goldschmidt, Da Costa and Righi2022) systematically reviewed 83 studies to propose a taxonomy of cybersecurity activities supported by text mining and topic modeling. Barik et al. (Reference Barik, Misra, Konar, Kaushik and Ahuja2022) conducted a systematic literature review (SLR) on the use of text mining in cybersecurity and analyzed 516 papers published between 2015 and 2021. They emphasized the potential of text mining methods to contribute to and improve cybersecurity efforts.

One of the first studies of topic modeling used to analyze the similarities and differences between NCSSs can be attributed to Kolini and Janczewski (Reference Kolini and Janczewski2017). They gathered and evaluated 60 NCCSs that were created between 2003 and 2016. Additionally, they suggested that the topic modeling approach can be utilized as an automated way for textual analysis and evaluation of national strategies. Song et al. (Reference Song, Kim, Bae and Kim2021) employed topic modeling to analyze the NCSSs of the US, UK, Japan, and the EU to discover the policy changes over time and to revise South Korea’s NCSS.

Even though the prior studies contributed to the literature by analyzing NCSSs, there is a significant gap regarding the detailed analysis of G20 nations’ strategies, both qualitatively and quantitatively. Previous research was often focused on specific regions or groups of countries such as ASEAN countries (Sunkph et al., Reference Sunkph, Ramjan and Ottamakorn2018), the Visegrad Group (Gorka, Reference Gorka2018), or selected European nations (Baezner and Cordey, Reference Baezner and Cordey2019; Kovacs, Reference Kovacs2019). Among the research that used topic modeling, the study by Song et al. (Reference Song, Kim, Bae and Kim2021) focused on the US, UK, Japan, and the EU only for policy changes over time and Kolini and Janczewski (Reference Kolini and Janczewski2017) analyzed NCSSs up to 2016, but their study did not focus on the G20 nations and is now outdated, as NCSSs have since been updated. In short, the literature lacks analyses of shared perspectives, distinct priorities, and latent themes in current G20 cybersecurity strategies. To address these gaps in the literature, we formulated research questions as presented in the introduction section of this manuscript.

3. Materials and methods

This study contains 10 primary steps as shown in Figure 1.

Figure 1. Study steps.

Initially, we sourced the NCSS documents for all G20 countries on the internet. For NCSS documents not in English, specifically those from Brazil and China, we utilized Google Translate for translation.

Explicitly stated objectives in strategy documents were then extracted manually by reading the documents. The set of declared cybersecurity goals and priorities of each G20 nation was obtained after this step.

To discover implicit themes and priorities within these strategies, we created Python code for topic analysis using latent dirichlet allocation (LDA). We chose LDA algorithm since it is widely used effective topic modeling algorithm to uncover hidden themes or topics within a set of documents (Blei et al., Reference Blei, Ng and Jordan2003). Each document was analyzed separately as its own corpus to reflect its unique context with lengths ranging from 2500 words to 32,000 words and this is sufficient data for meaningful topic modeling. We adjusted the number of topics based on the explicit themes found in each document to match its content. With this approach, we avoided choosing the number of topics randomly. The code was written in Python using NLTK and Gensim libraries. The text was prepared by tokenizing and lemmatizing, removing stop words, and filtering out non-alphabetic words. By using LDA, we were able to go beyond a simple keyword analysis and uncover the latent themes and objectives in G20 cybersecurity strategies.

After employing the LDA topic modeling algorithm on each strategy document, we identified topics consisting of sets of keywords with associated weights. To create meaningful statements from these keywords, we used the GPT-4 advanced language model (OpenAI 2024). In other words, LDA-generated keywords with their weights were given into GPT-4 to generate preliminary topic statements for each country. Since the context of NCSSs has nuances that could be misled by an AI model, we reviewed and revised the GPT-4 generated topic statements. With this iterative process, we generated and then revised latent topics to ensure they accurately reflected the context of each country’s strategy.

Now then, we had all the data to start with the detailed analysis of the strategies. We first performed an analysis to compare explicitly stated objectives in the G20 strategies. Then we analyzed the implicit objectives of the strategies discovered by topic modeling. With the help of these two different analyses, we obtained comparisons of explicit and implicit objectives and unstated latent focus areas and concerns of the G20 countries.

Furthermore, we analyzed the predominant keywords of the strategies, extracted from LDA topics, to get information about the focal areas and thematic emphasis within each nation’s cybersecurity policy.

Lastly, we identified common keywords in the strategies, extracted from LDA topics. This step was important to understand the common focus areas of the strategies.

4. Findings and results

In this section, we provided the research questions and their answers.

RQ1: Shared objectives and distinct priorities in G20 cybersecurity strategies

We extracted the objective statements manually from the G20 strategies. We found that there are 25 distinct objectives as shown in Table 2 (Argentina Government, 2023; Australian Government, 2020; Brasil Presidency of the Republic, 2020; Government of Canada, 2018; The Cyberspace Administration of China, 2016; European Commission, 2020; ANSSI, 2015; German Federal Ministry of Interior, 2016; Indian Ministry of Communication and IT, 2013; Indonesian Government, 2023; Italian Government, 2022; Government of Mexico, 2017; Japanese Government, 2021; Russian Federation, 2016; Kingdom of Saudi Arabia, 2020); South Korea National Security Office, 2019; Republic of Turkey, 2020; UK Government, 2016; The White House, 2023; South Africa State Security Agency, 2012). The objectives are organized based on the number of their adoptions by these countries.

Table 2. Objectives in the NCSSs of G20 countries

1 Shows the number of countries which have those keywords.

The most widely accepted objectives, adopted by 10–17 countries, include protecting critical infrastructures, networks, and data; establishing national coordination and ecosystems; engaging in international cooperation; supporting cybersecurity industry, R&D, and innovation; and improving cybersecurity capabilities. The acceptance of these objectives demonstrates widespread agreement about the significance of holistic cybersecurity measures, collaborative frameworks, and innovative developments in this domain. These shared objectives among G20 countries indicate a unified approach to cybersecurity.

Among the shared objectives, emphasis on international cooperation shows a recognition that cyber threats require collective actions. For example, the Budapest Convention on Cybercrime is referenced in several strategies and most of the G20 countries, except for Russia, China, India, Indonesia, and Saudi Arabia, are among the signatory bodies (Council of Europe, 2024). Nations like the USA, UK, and EU members actively participate in international organizations and adopt common standards such as ISO/IEC 27000 series and the NIST Cybersecurity Framework, and GDPR (General Data Protection Regulation) (European Union, 2024). For collaboration, there are various global cybersecurity initiatives and alliances that G20 nations are among the members, such as ENISA (2024), ASEAN (2024), GFCE (Global Forum on Cyber Expertise) (2024), FIRST (Forum of Incident Response and Security Teams) (2024), UN GGE (United Nations Group of Governmental Experts) (United Nations, 2024b), The Paris Call for Trust and Security in Cyberspace (Paris Call 2024), and APEC (Asia-Pacific Economic Cooperation) Cybersecurity Strategy (APEC, 2005). Despite these efforts, because of the different national interests, legal systems, priorities, and technological advancements, challenges remain in widespread cooperation.

Moderately adopted objectives, adopted by 5–8 countries, are raising awareness, and fostering cybersecurity culture, establishing incident response capacity, addressing cybercrime, establishing legislative and regulatory frameworks, and providing secure products and services. These objectives reflect a focus on specific aspects of cybersecurity, such as legal structures, public education, and incident management, which are crucial but perhaps not as universally prioritized as those in the most adopted category.

Selectively adopted objectives, adopted by 3–4 countries, include respecting individual rights and fundamental values, security of new generation technologies, and encouraging the use of international standards. This suggests a more targeted approach to cybersecurity, possibly addressing unique national challenges or focusing on emerging areas in cybersecurity.

In the cybersecurity strategies of G20 countries, certain objectives, adopted by 1–2 countries, demonstrate unique national focuses. India and Japan prioritize the security of the supply chain. The European Union and Russia, on the other hand, emphasize the security of their military information systems. Russia and the USA demonstrate a cross-national strategy by focusing on expanding cybersecurity protection to include the interests of their respective allies. Exclusively, Russia addresses the contemporary aspect of cyberwarfare, which is the task of blocking disinformation and psychological operations online. Turkey shows a dedication to self-reliance and innovation by focusing its efforts on the development of domestic and national cybersecurity solutions. With a unique positioning in crucial cyber technologies, the UK aims to gain a strategic advantage in this quickly changing industry. Finally, to emphasize its commitment to active warfare in the digital domain, the USA takes a proactive and aggressive posture and focuses on the disruption and dismantling of cyber threat actors. These goals show the variety of approaches that the G20 countries have been using, each customizing their strategy to fit unique national interests and security concerns in the global cyber landscape.

RQ2: Latent themes in the cybersecurity strategies of G20 countries

We created Python code to use LDA topic modeling algorithm to uncover latent themes in G20 cybersecurity strategies. LDA algorithm generated topics for each of the strategy documents. There are sets of keywords in the topics with weights that show the importance of the keywords in the entire strategy document. Then, GPT-4 was used to create meaningful statements from each topic to create coherent and interpretable topics. Since there is a potential that AI-generated content might miss the context-specific meanings, we revised the GPT-4 generated statements manually. The process was iterated for each G20 country. This expert revision ensured that the final topic descriptions adequately captured the thematic meaning of the LDA-generated themes. A sample topic’s process is shown below:

  1. (1) LDA topic from Australia’s strategy document: 0.146*‘critical’ + 0.096*‘national’ + 0.072*‘asset’ + 0.069*‘need’ + 0.044*‘invest’. (Note that, the weights indicate how strongly each word contributes to the topic definition.)

  2. (2) GPT-4 generated statement: “Investing in national assets is critical due to the need for robust infrastructure and security.”

  3. (3) To ensure contextual relevance and accuracy, we reviewed and revised these generated statements manually. Expert revised statement: “Investing in critical national assets is needed for robust infrastructure and security.”

After discovering latent objectives in the strategies, we performed a comparison process to see which of the implicit objectives were not stated explicitly in the strategies. With this comparison, we found areas of implicit focus or unspoken objectives that influence these countries’ cybersecurity postures and priorities. The comparative study revealed that while all explicitly stated objectives were discovered by LDA analysis, there are latent topics that are not explicitly stated in the strategies. See Table 3 for implicit strategic concerns of G20 strategies.

Table 3. Topics found by LDA but not explicitly stated in the strategy documents

The analysis revealed that G20 countries reflect following perspectives in their strategies indirectly:

  • Dynamics of cybersecurity is global and there is a need for international collaboration.

  • There is an interest in technological innovation and advancing cybersecurity technologies.

  • There is an awareness of legal and regulatory considerations that shows the need for legal frameworks.

  • Public-private partnership is vital to improving national cybersecurity initiatives.

  • Critical infrastructures and services have to be protected against cyber attacks.

  • Privacy and data protection should be addressed.

  • Education and workforce development is vital to have a skilled cybersecurity workforce and raise public awareness of cyber threats.

The findings from the LDA analysis provide understanding of the diverse approaches nations use to maintain the balance between explicit and implicit declarations of cybersecurity strategies.

RQ3: Predominant and common keywords in the G20 cybersecurity strategies

Keywords were found by using LDA technique. We excluded common keywords such as “national ““cyber ““security ““cybersecurity ““cyberspace “and “strategy” from the keywords pool. This enabled us to move our focus away from these commonly used but generic terms and onto more distinctive and specialized components of each country’s approach. These common keywords while important frequently mask the subtle distinctions between countries.

During analysis, the top 50% of keywords by weight from each country’s strategy were involved, while concurrently limiting the number of keywords to a range of 15–20. This approach was chosen to ensure both depth and breadth in our analysis, capturing a vast array of significant terms without overwhelming the focus on the most crucial elements. By implementing a percentage threshold, we maintained consistency across datasets of varied sizes, thus accommodating the inherent variability in the volume and detail of the original strategy documents. The additional constraint of 15–20 keywords was important in preventing over- or under-representation, thereby guaranteeing a focused but encompassing description of each country’s strategic priorities.

Following the extraction of keywords from LDA topics, we arranged them according to their respective weights, ensuring an accurate representation of their significance within the cybersecurity strategies of each country. Top keywords and priorities of the G20 countries are given in Table 4.

Table 4. Top keywords in G20 cybersecurity strategies

Apart from keywords such as “national,” “cyber,” “security,” “cybersecurity,” “cyberspace,” and “strategy,” there are 66 common keywords (again generated by LDA topic modeling) that exist in the strategies of at least two countries as shown in Table 5. To prevent the table from extending across multiple pages, keywords that occurred with the same frequency across different countries were consolidated into a single row. The countries where each keyword appeared were enclosed in curly brackets ({}) to indicate the specific nations associated with that keyword.

Table 5. Common keywords in G20 cybersecurity strategies

The analysis revealed a universal emphasis on ‘Information’, appeared in the strategies of 18 countries, underscoring information as a vital aspect of cybersecurity.

Furthermore, the concurrent prominence of ‘Government’ and ‘Technology’ in 12 countries’ strategies may reflect an integration of state-driven policies and technological advancements in cybersecurity. This trend might show growing recognition of the need for governmental intervention in cybersecurity and investments in the technological domain to support national cyber defenses. Additionally, 11 countries have ‘Development’ and ‘International’ keywords that can show a motivation for developing robust cybersecurity posture and improving international cooperation.

The existence of ‘Sector’, ‘System’, and ‘Threat’ in 10 countries might denote a focus on specific sectors, systemic security and threats. The keyword ‘Service’, in nine countries may indicate a service-oriented approach to cybersecurity. Having ‘Digital’, ‘Private’, and ‘Public’ keywords in seven countries may show the collaboration of both private and public sectors in cybersecurity efforts.

The presence of ‘Action’, ‘Critical’, ‘Organization’, and ‘State’ in six strategies might show a focus on active measures to protect critical infrastructure and organizational assets. Additionally, the presence of ‘Cooperation’, ‘Country’, ‘Implementation’, ‘Infrastructure’, and ‘Support’ in five countries’ strategies may reflect a desire for implementation and support of cybersecurity infrastructure through cooperative efforts.

The mention of ‘Cybercrime’, ‘Agency’, ‘Business’, ‘Data’, ‘Internet’, ‘Level’, ‘Network’, and ‘New’ in four countries might reveal a focus on specific actions such as combating cybercrime, improving business value of cybersecurity, data security, and internet and network infrastructure. The inclusion of ‘New’ might mean mindset to adopt innovative approaches and technologies in cybersecurity measures.

Keywords that appear in three or two countries each can indicate an explicit focus on specific aspects of cybersecurity or context-based priorities in those countries.

Based on the analysis of the LDA topics, the heat-map in Figure 2 depicts the weights of the top 20 keywords in G20. Each cell’s color intensity indicates the relative weight or significance of a specific keyword in a country’s cybersecurity strategy.

Figure 2. Heatmap of Top 20 keywords weights by country.

5. Conclusion

In this study, a mixture of qualitative and quantitative methodologies was performed to explore the NCSSs of G20 countries. Through this rigorous analysis, we successfully discovered both the stated objectives and latent themes that are not stated explicitly within these strategies. Our findings reveal a complex tapestry of shared objectives and distinct national priorities among G20 countries in terms of cybersecurity strategies. Having shared objectives among the G20 nations is an indication of unified approaches in the strategies to strengthen cybersecurity.

The implementation of latent dirichlet allocation (LDA) for topic modeling played a significant role in discovering objectives and themes that were not visibly stated in the NCSS documents. This approach provided valuable insights into the implicit concerns and focus areas of different countries and therefore it enhanced our understanding of the strategic underpinnings of national cybersecurity policies. The analysis of predominant and common keywords further improved the understanding the nature and overarching trends in cybersecurity.

Our study’s findings hold significant implications for cybersecurity policymakers at both national and international levels. By identifying shared objectives and latent themes among G20 countries, the insights gained can guide the formulation of more effective and contextually relevant cybersecurity policies. Policymakers can leverage common ground to strengthen international cooperation, adopt proven strategies against cyber threats and improve information sharing. Recognizing unique national strategies enables the development of customized strategies that address particular national problems while keeping up with worldwide cybersecurity trends. Additionally, the study highlights the critical need for continued and in-depth exploration of cybersecurity strategies to keep pace with the rapidly evolving cyber threat landscape.

However, this research is not without its limitations. The primary limitation is the analysis of NCSSs written in various languages, as translations may not capture the nuances and correct terminology of the original texts. All countries except for Brazil and China published English versions of their strategy documents, which minimizes the language limitation. Another limitation is the inherent constraints of algorithmic topic modeling. Normally, this limitation is inevitable while using topic modeling. On the other hand, to overcome this limitation, we manually analyzed the explicit strategies and manually revised the topics. Despite these challenges, the study contributes substantially to the field of cybersecurity policy analysis and offers a framework for future research endeavors that could expand to include a more diverse array of nations and employ more advanced analytical techniques.

For future research, there are some viable options to consider. Researchers could use multilingual semantic analysis tools to better capture the nuances of different languages. They could also explore other methods, such as dynamic topic modeling or deep learning, to gain deeper insights into how policies change over time. Additionally, future studies could focus on underexplored regions, analyze cybersecurity strategies in the private sector, or investigate how NCSSs are implemented and how effective they are across different countries.

In conclusion, our study underscores the importance of a multi-dimensional approach, containing both qualitative and quantitative methodologies, to understanding and developing national cybersecurity strategies. As cyber threats continue to evolve in complexity and scale, such holistic analyses will be crucial in equipping nations to effectively safeguard their digital infrastructure.

Data availability statement

The data was collected from the G-20 countries’ relevant websites publishing their cybersecurity strategies. Data is public.

Author contribution

Conceptualization: H.Ç.; E.E., Methodology: H.Ç., Literature search: H.Ç.; E.E., Draft preparation: H.Ç., Writing: H.Ç.;E.E All authors wrote the paper and approved the final submitted draft.

Funding statement

This work received no specific grant from any funding agency, commercial or not-for-profit sectors.

Competing interest

Author H.Ç. and Author E.E. declare none.

References

Adams, S, Carter, B, Fleming, C and Beling, PA (2018) Selecting system specific cybersecurity attack patterns using topic modeling. In 17th IEEE International Conference on Trust, Security and Privacy in Computing and Communications/12th IEEE International Conference on Big Data Science and Engineering (TrustCom/BigDataSE). IEEE, pp. 490497. 10.1109/TrustCom/BigDataSE.2018.00076.Google Scholar
Ali, SM, Razzaq, A, Abbass, H, Yousaf, M and Shan, R us (2024, October 18) A Hybrid Analytical Framework for Enhancing Cybersecurity in Underdeveloped Countries. In 2024 IEEE 4th International Conference on Electronic Communications, Internet of Things and Big Data (ICEIB). https://doi.org/10.20944/preprints202410.1442.v1.CrossRefGoogle Scholar
APEC (2005) APEC (Asia-Pacific Economic Cooperation) Cybersecurity Strategy. Retrieved from https://www.apec.org/docs/default-source/groups/tel/05_tel_apecstrategy.pdfGoogle Scholar
Argentina Government (2023) Second National Cybersecurity Strategy. Retrieved from https://www.argentina.gob.ar/sites /default/files/anexo_6777529_1.pdfGoogle Scholar
ASEAN (2024) Association of Southeast Asian Nations. https://asean.org/ (accessed 17 November 2024)Google Scholar
Australian Government (2020) Australia’s Cyber Security Strategy. Retrieved from https://www.homeaffairs.gov.au/cyber-security-subsite/files/cyber-security-strategy-2020.pdfGoogle Scholar
Azmi, R, Tibben, W and Win, KT (2018) Review of cybersecurity frameworks: Context and shared concepts. Journal of Cyber Policy 3(2), 258283. https://doi.org/10.1080/23738871.2018.1520271.CrossRefGoogle Scholar
Baezner, M and Cordey, S (2019) National Cybersecurity Strategies in Comparison – Challenges for Switzerland. Zürich: Center for Security Studies (CSS)Google Scholar
Barik, K, Misra, S, Konar, K, Kaushik, M and Ahuja, R (2022) A comparative study on the application of text mining in cybersecurity. Recent Advances in Computer Science and Communications 16(3). https://doi.org/10.2174/2666255816666220601113550.Google Scholar
Bechor, T and Jung, B (2019) Current state and modeling of research topics in cybersecurity and data science. Systemics, Cybernetics and Informatics 17(1), 129156.Google Scholar
Belfer Center for Science and International Affairs (2020) National cyber power ındex 2020 – Methodology and analytical considerations. In Belfer Center for Science and International Affairs - Harvard Kennedy School (September), pp. 171. https://www.belfercenter.org/sites/default/files/2020-09/NCPI_2020.pdfGoogle Scholar
Blei, DM, Ng, AY and Jordan, MI (2003) Latent dirichlet allocation. Journal of Machine Learning Research, 3, 9931022.Google Scholar
Brasil Presidency of the Republic (2020) Brasil National Cybersecurity Strategy. Retrieved from http://www.planalto.gov.br/CCIVIL_03/_Ato2019-2022/2020/Decreto/D10222.htmGoogle Scholar
Council of Europe (2024) The Convention on Cybercrime (Budapest Convention, ETS No. 185) and Its Protocols. https://www.coe.int/en/web/cybercrime/the-budapest-convention (accessed 17 November 2024)Google Scholar
e-Governance Academy (2024) National Cyber Security Index (NCSI). https://ncsi.ega.ee/methodology/ (accessed 16 November 2024)Google Scholar
Enescu, S (2020) A comparative study on European cyber security strategies. Redefining Community in Intercultural Context 9, 277282.Google Scholar
ENISA (2012a) National Cyber Security Strategies. Retrieved from https://www.enisa.europa.eu/publications/cyber-security-strategies-paperGoogle Scholar
ENISA (2012b) National Cyber Security Strategies – Practical Guide on Development and Execution. Retrieved from https://www.enisa.europa.eu/publications/national-cyber-security-strategies-an-implementation-guideGoogle Scholar
ENISA (2014) An Evaluation Framework for National Cyber Security Strategies European Union Agency for Network and Information Security (ENISA). Retrieved from https://op.europa.eu/en/publication-detail/-/publication/e751c1c2-cf11-449c-9fb9-78e5660d49b1/language-enGoogle Scholar
ENISA (2016) NCSS Good Practice Guide: Designing and Implementing National Cyber Security Strategies European Union Agency for Network and Information Security (ENISA). Retrieved from https://www.enisa.europa.eu/publications/ncss-good-practice-guide/at_download/fullReportGoogle Scholar
ENISA (2019) Good Practices in Innovation Under NCSS European Union Agency for Cybersecurity (ENISA). Retrieved from https://op.europa.eu/en/publication-detail/-/publication/3046c9cf-47cb-11ea-b81b-01aa75ed71a1/language-enGoogle Scholar
ENISA (2020) National Capabilities Assessment Framework (Sarri, A and Thirriot, A (eds.)). The European Union Agency for Cybersecurity (ENISA). Retrieved from https://www.enisa.europa.eu/publications/national-capabilities-assessment-frameworkGoogle Scholar
ENISA (2024) The European Union Agency for Cybersecurity. https://www.enisa.europa.eu/ (accessed 17 November 2024)Google Scholar
EUI & Booz Allen Hamilton (2011) Cyber Power Index – Findings and Methodology. pp. 136.Google Scholar
European Commission (2020) The EU’s Cybersecurity Strategy for the Digital Decade. Retrieved from https://ec.europa.eu/newsroom/dae/redirection/document/72164Google Scholar
European Union (2024) EU General Data Protection Regulation (GDPR). https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A32016R0679&qid=1731850404866 (accessed 17 November 2024)Google Scholar
Falch, M, Olesen, H, Skouby, KE, Tadayoni, R and Williams, I (2023) Cybersecurity strategies for SMEs in the Nordic Baltic region. Journal of Cyber Security and Mobility 11(6), 727754.Google Scholar
FIRST (2024) Forum of Incident Response and Security Teams. https://www.first.org/ (accessed 17 November 2024)Google Scholar
GFCE (2024) The Global Forum on Cyber Expertise (GFCE). https://thegfce.org/ (accessed 17 November 2024)Google Scholar
Gorka, M (2018) The cybersecurity strategy of the visegrad group countries. Politics in Central Europe 14(2), 7598. 10.2478/pce-2018-0010.CrossRefGoogle Scholar
Government of Canada (2018) National Cyber Security Strategy. Retrieved from https://www.publicsafety.gc.ca/cnt/rsrcs/pblctns/ntnl-cbr-scrt-strtg/ntnl-cbr-scrt-strtg-en.pdfGoogle Scholar
Government of Mexico (2017) National Cybersecurity Strategy. Retrieved from https://www.gob.mx/cms/uploads/attachment/file/399655/ENCS.ENG.final.pdfGoogle Scholar
Ignaczak, L, Goldschmidt, G, Da Costa, CA and Righi, RDR (2022) Text mining in cybersecurity. ACM Computing Surveys 54(7), 136. 10.1145/3462477.CrossRefGoogle Scholar
IISS (2021) Cyber Capabilities and National Power: A Net Assessment. Retrieved from https://www.iiss.org/blogs/research-paper/2021/06/cyber-capabilities-national-powerGoogle Scholar
Indian Ministry of Communication and IT (2013) National Cyber Security Policy. Retrieved from https://www.meity.gov.in/sites/upload_files/dit/files/National%20Cyber%20Security%20Policy%20%281%29.pdfGoogle Scholar
Indonesian Government (2023) National Cyber Security Strategy. Retrieved from https://peraturan.bpk.go.id/Download/312229/Perpres%20Nomor%2047%20Tahun%202023.pdfGoogle Scholar
Iova, R and Watashiba, T (2023) NCSS: A global census of national positions on conflict, neutrality and cooperation. European Conference on Cyber Warfare and Security 22, 420428. 10.34190/eccws.22.1.1168.CrossRefGoogle Scholar
Italian Government (2022) National Cybersecurity Strategy 2022–2026. https://www.acn.gov.it/ACN_EN_Strategia.pdfGoogle Scholar
ITU (2018a) Guide to Developing a National Cyber Security Strategy – Strategic Engagement in Cybersecurity. Geneva. Retrieved from https://www.itu.int/dms_pub/itu-d/opb/str/D-STR-CYB_GUIDE.01-2018-PDF-E.pdfGoogle Scholar
ITU (2018b) Guide to Developing a National Cybersecurity Strategy – Strategic Engagement in Cybersecurity. International Telecommunication Union.Google Scholar
ITU (2024) Global Cybersecurity Index (GCI). https://www.itu.int/en/ITU-D/Cybersecurity/pages/global-cybersecurity-index.aspx (accessed 16 November 2024)Google Scholar
Jacuch, A (2021) Comparative Analysis of Cybersecurity Strategies. European Union Strategy and Policies. Polish and Selected Countries Strategies. On-Line Journal Modelling the New Europe 37(6), 102120. https://doi.org/10.24193/OJMNE.2021.37.06.CrossRefGoogle Scholar
Japanese Government (2021) Outline of Japan’s Next Cybersecurity Strategy. Retrieved from https://www.nisc.go.jp/eng/pdf/txt_next_CS_strategy_outline.pdfGoogle Scholar
Kingdom of Saudi Arabia (2020) National Cybersecurity Strategy. Retrieved from https://nca.gov.sa/national_cybersecurity_strategy-en.pdfGoogle Scholar
Kolini, F and Janczewski, L (2017) Clustering and Topic Modelling: A New Approach for Analysis of National Cyber Security Strategies.Google Scholar
Kovacs, L (2019) National Cybersecurity Strategy Framework. Academic and Applied Research in Military and Public Management Science 18(2), 6576. https://doi.org/10.32565/aarms.2019.2.9.CrossRefGoogle Scholar
Luiijf, E, Besseling, K and Graaf, P (2013) Nineteen National Cyber Security Strategies. International Journal of Critical Infrastructure Protection 9, 3. https://doi.org/10.1504/IJCIS.2013.051608.CrossRefGoogle Scholar
Montasari, R (2023) Cyber threats and the security risks they pose to national security: An assessment of cybersecurity policy in the United Kingdom. In Montasari, R (ed.), Countering Cyberterrorism: The Confluence of Artificial Intelligence, Cyber Forensics and Digital Policing in US and UK National Cybersecurity. Cham: Springer International Publishing, pp. 725. https://doi.org/10.1007/978-3-031-21920-7_2.CrossRefGoogle Scholar
NATO CCD COE (2012) National Cyber Security Framework Manual, Klimburg, A (ed.), 1st Edn. NATO CCD COE Publications.Google Scholar
Newmeyer, KP (2015) Elements of National Cybersecurity Strategy for Developing Nations. National Cybersecurity Institute Journal 1(3), 919.Google Scholar
Odebade, A and Benkhelifa, E (2023) A Comparative Study of National Cyber Security Strategies of Ten Nations. https://doi.org/10.48550/arXiv.2303.13938CrossRefGoogle Scholar
OECD (2012) Cybersecurity Policy Making at a Turning Point – Analysing a New Generation of National Cybersecurity Strategies for the Internet Economy Organisation for Economic Co-operation and Development. Retrieved from http://www.oecd.org/sti/ieconomy/cybersecuritypolicy making.pdfGoogle Scholar
Organization of American States (2022) National Cybersecurity Strategies: Lessons Learned and Reflections from the Americas and Other Regions.Google Scholar
Ovchinnikova, O and Upadhyay, NK (2023) The level of cybersecurity of the BRICS member countries in ınternational ratings: Prospects for cooperation. BRICS Law Journal 10(1), 734. https://doi.org/10.21684/2412-2343-2023-10-1-7-34.CrossRefGoogle Scholar
Oxford GCSCC (2021) Cybersecurity Capacity Maturity Model for Nations (CMM). Retrieved from https://gcscc.ox.ac.uk/the-cmmCrossRefGoogle Scholar
Paris Call (2024) The Paris Call for Trust and Security in Cyberspace. https://pariscall.international/en/ (accessed 17 November 2024)Google Scholar
Potomac Institute for Policy Studies (2015) Cyber Readiness Index 2.0. Retrieved from https://potomacinstitute.org/images/CRIndex2.0.pdfGoogle Scholar
Russian Federation (2016) Doctrine of Information Security. Retrieved from http://kremlin.ru/acts/bank/41460Google Scholar
Sabillon, R (2021) National Cybersecurity Strategies. pp. 84102. https://doi.org/10.4018/978-1-7998-4162-3.ch005.CrossRefGoogle Scholar
Sabillon, R, Cavaller, V and Cano, J (2016) National cyber security strategies: Global trends in cyberspace. International Journal of Computer Science and Software Engineering 5(5), 6781.Google Scholar
Samtani, S, Chinn, K, Larson, C and Chen, H (2016) Azsecure hacker assets portal: Cyber threat intelligence and malware analysis. In 2016 IEEE Conference on Intelligence and Security Informatics (ISI). pp. 1924.CrossRefGoogle Scholar
Santisteban, AS, Ocares, L and Andrade-Arenas, L (2020) Analysis of National Cybersecurity Strategies. International Journal of Advanced Computer Science and Applications 11(12). https://doi.org/10.14569/IJACSA.2020.0111288.CrossRefGoogle Scholar
Shafqat, N and Masood, A (2016) Comparative analysis of various national cybersecurity strategies. International Journal of Computer Science and Information Security 14(1), 129136.Google Scholar
Song, M, Kim, DH, Bae, S and Kim, S-J (2021) Comparative analysis of national cyber security strategies using topic modelling. International Journal of Advanced Computer Science and Applications 12(12). https://doi.org/10.14569/IJACSA.2021.0121209.CrossRefGoogle Scholar
South Africa State Security Agency (2012) The National Cybersecurity Policy Framework. Retrieved from https://www.gov.za/sites/default/files/gcis_document/201512/39475gon609.pdfGoogle Scholar
South Korea National Security Office (2019) National Cybersecurity Strategy. Retrieved from https://ccdcoe.org/uploads/2018/10/South-Korea_English-National-Cybersecurity-Strategy-03-April-2019_English-1.pdfGoogle Scholar
Štitilis, D, Pakutinskas, P and Malinauskaitė, I (2017) EU and NATO cybersecurity strategies and national cyber security strategies: A comparative analysis. Security Journal 30(4), 11511168. https://doi.org/10.1057/s41284-016-0083-9.CrossRefGoogle Scholar
Sunkph, J, Ramjan, S and Ottamakorn, C (2018) Cybersecurity policy in ASEAN countries. In Information Institute Conferences. Las Vegas, pp. 17. https://www.researchgate.net/profile/Jirapon-Sunkpho-2/publication/324106226_Cybersecurity_Policy_in_ASEAN_Countries/links/5abdc2ea45851584fa6fca37/Cybersecurity-Policy-in-ASEAN-Countries.pdf (accessed 16 November 2024)Google Scholar
The Cyberspace Administration of China (2016) National Cyberspace Security Strategy. Retrieved from http://www.cac.gov.cn/2016-12/27/c_1120195926.htmGoogle Scholar
Tikk, E and Kerttunen, M (2020) Routledge Handbook of International Cybersecurity (Tikk, E and Kerttunen, M (eds.)). Routledge. https://doi.org/10.4324/9781351038904.CrossRefGoogle Scholar
UNIDIR (2013) The Cyber Index-International Security Trends and Realities United Nations Publications. Retrieved from http://www.unidir.org/files/publications/pdfs/cyber-index-2013-en-463.pdfGoogle Scholar
United Nations (2024a) Cyber Security Portal. https://cyberpolicyportal.org/Google Scholar
United Nations (2024b) United Nations Group of Governmental Experts. https://disarmament.unoda.org/group-of-governmental-experts/ (accessed 17 November 2024)Google Scholar
Wamala, F (2011) ITU National Cybersecurity Strategy Guide. Retrieved from https://www.itu.int/ITU-D/cyb/cybersecurity/docs/itu-national-cybersecurity-guide.pdfGoogle Scholar
Figure 0

Table 1. Common National Level Cybersecurity indices and frameworks

Figure 1

Figure 1. Study steps.

Figure 2

Table 2. Objectives in the NCSSs of G20 countries

Figure 3

Table 3. Topics found by LDA but not explicitly stated in the strategy documents

Figure 4

Table 4. Top keywords in G20 cybersecurity strategies

Figure 5

Table 5. Common keywords in G20 cybersecurity strategies

Figure 6

Figure 2. Heatmap of Top 20 keywords weights by country.

Submit a response

Comments

No Comments have been published for this article.