1 Introduction

Human expertise, perceptions and intelligence are in many cases necessary elements to include when modeling real-world phenomena [1]. The seminal continuous-valued logic introduced in the mid-1960s by Prof. Lofti Zadeh, fuzzy sets, fuzzy rule base, fuzzy logic, and thus fuzzy set research [2] all address the inclusion and treatment of uncertain elements in a wide range of problems. The incorporation of the fuzzy toolbox, first used in engineering applications [3] and later applied to a plethora of areas such as robotics [4], health disciplines [5], managerial sciences [6] and business economics [7], led to the effective management of imprecise, incomplete, vague, fragmentary, subjective or missing information in complex and human-like dynamic systems [8].

Currently, the concept and treatment of uncertainty are commonly accepted. However, when first introduced, fuzzy logic faced criticism from classical standpoints [9]. Vagueness and ambiguity are concepts that have been traditionally avoided by scientists due to their inherent undesired state of imprecision [10]. Probability theory dealt at first with uncertainty, applied mainly at the level of randomness; nonetheless, numerous problems required further developments. Fuzzy sets provided the measurable framework involving true values ranging from zero to one, yielding membership functions, which were required for advancements in the field. A special case of fuzzy sets was then proposed—the possibility theory [11, 12]—and from there a series of extensions in the area were exponentially triggered such as type-2 fuzzy sets [13, 14], intuitionistic fuzzy sets [15], interval-valued intuitionistic fuzzy sets [16, 17], fuzzy multi-sets [18], hesitant fuzzy sets [19], and Pythagorean fuzzy sets [20], among others.

The novel developments in fuzzy set theory are disseminated in diverse scientific journals (see, e.g., [21]). These publications compile original research articles from academics all over the world. The increasing impact of fuzzy developments motivates the analysis of connections, networks and influence that describe the activity of the leaders in the field. The present study proposes the examination of 22 narrowly targeted fuzzy set theory journals (NTFSTJ) to generate a picture of these connections. The selection of these journals is based on the framework proposed by Merigó et al. [9]. More specifically, on their proposed indicator: percentage of fuzzy papers in the journal (%PF). The selected 22 journals, including their %PF retrieved from [9], are as follows: the Iranian Journal of Fuzzy Systems (IrJFS), 98; IEEE Transactions on Fuzzy Systems (TFUZZ), 96.9; Fuzzy Sets and Systems (FSS), 91.7; International Journal of Fuzzy Systems (IJFS), 90.6; Fuzzy Optimization and Decision Making (FODM), 87.4; Journal of Intelligent & Fuzzy Systems (JIFS), 71.3; International Journal of Uncertainty Fuzziness and Knowledge-Based Systems (IJUFKS), 65.4; Soft Computing (SC), 44.1; International Journal of Approximate Reasoning (IJAR), 42.3; International Journal of Intelligent Systems (INT), 40.1; Applied Soft Computing (ASOC), 37; Journal of Multiple-Valued Logic and Soft Computing (JMVLSC), 35.4; International Journal of Computational Intelligence Systems (IJCIS), 30.9; International Journal of General Systems (IJGS), 29.2; Information Sciences (INS), 28.5; Intelligent Automation and Soft Computing (IASC), 22.7; Expert Systems with Applications (ESWA), 22.2; Cybernetics And Systems (C&S), 16.5; International Journal of Information Technology & Decision Making (IJITDM), 16.2; and Knowledge Based Systems (KNOSYS), 11.7. In addition, we have included in the study the journal Kybernetes (Kyb) for its well-known contributions to the field and the journal Information Fusion (INFFUS) for the exceptional performance it has achieved in recent years and for its focus on fuzzy set theory. Note that some top journals that have published many seminal articles connected to fuzzy logic are not included in the analysis because these journals have a broader scope and publish only a small percentage of articles related to fuzzy systems.

Bibliometric studies have recently gained popularity. This increased interest has mainly been triggered by the recent developments in information technologies, the immediateness of information transfer, and the rapidly expanding publication of findings in the fuzzy set theory field. Bibliometric analyses generally involve the study of large amounts of scientifically published documents employing statistical tools. The bibliographic material gathered can range from searches in the title, abstract, or embedded keywords to sophisticated advanced searches with combinations of diverse key elements such as sources of information, research and specific science categories. These options depend on the scientific database consulted. For the specific case of fuzzy set theory, a significant amount of bibliometric studies have been proposed, e.g., [9] offering a general overview of fuzzy set theory field by applying a search process in the WoS Core Collection on papers addressing the word “fuzzy” in the topic section, the authors find a total of 115,000 papers and present a detailed study on 56,500 refined core studies, sectioning them by annual publications, citation structure, most cited papers in fuzzy research, most influential fuzzy research journals, books, authors, organizations and countries. Ref. [22] shows the findings on a bibliometric study of fuzzy set theory articles and reviews based on WoS databases for documents published from 1986 to 2015 that contain the word “fuzzy” in the abstract title or keywords. This search is set specifically for the region—China, People’s Republic of China or Chinese—and a total of 12,936 documents are explored using CiteSpace software and sectioned by papers’ geographical distribution, international collaborators, top-cited works, top-ranked categories terms and keywords, and a cocitation analysis. Ref. [23] presents a study on the evolution of the fuzzy set theory field with special insights about the contributions of the Spanish community and a comparison to other leading countries in fuzzy set theory research. Here, the authors use the CoPalRed software for a network and a co-word analysis to generate bibliometric maps. Based on the query included, it can be observed that the search process is carried out in WoS for article-type documents including the word “fuzz*” in the topic section or papers published in the journals FSS or TFUZZ and for the countries of Scotland or Japan or Canada or Spain or England or North Ireland or USA or People’s Rep. of China or Wales or Germany. Authors show that the resulting query generates a total of 16,344 research documents analyzed by periods of time, thematic networks, and a specific analysis for Spanish institutions and most influential articles. Ref. [24] presents a bibliometric study including the keywords “fuzzy” and “decision-making” in the topic section; from the gross 14,525 published documents matching the criteria, the authors filter results obtained a final pool of 8135 documents ranging from 1970 to 2014, with the resulting selection partitioned and analyzed using the VOS viewer software to identify the top influential journals, papers, authors, and universities using diverse bibliometric indicators. Ref. [25] retrieves the findings for the proposed search of “linguistic decision making” as the topic in the WoS database, obtaining a total amount of 2017 documents; these results are displayed using CiteSpace software and divided by highly cited authors, journals and subject categories. Ref. [26] discusses the results from a search using the keywords “fuzzy decision*” in the topic section delimited by the years 1900 to 2015 of the WoS database; the resulting 13,901 records were treated using Vantage Point text-mining software aiming to portray 35 years of fuzzy decision research and a dissection on annual trends, regions and countries analyses, most influential institutions and journals, highly cited papers and WoS categories. Ref. [27] shows a review on fuzzy multi-criteria decision-making applications for energy policymaking problems; the authors use preferred reporting items for systematic reviews and meta-analyses, and the search process yields 150 studies that are further analyzed, such that fuzzy AHP in combination with other methods are the most recurrent papers found. Ref. [28] performed a comprehensive literature review from 1984 to 2018 of fuzzy techniques applied to construction project management; their findings are sectioned by decades, and most of the specialized applications in the field are launched in the early 2000s, coinciding with the emergence of information technologies. Finally, [29], using a systematic review of the field of complex fuzzy sets and logic, a recent subarea of fuzzy set theory, performed a citation-like search to the seminal papers [30, 31]; combining it with results from Google Scholar, their findings present a pool of some of the current papers that constitute the main ideas and foundations of this specific subfield of fuzzy sets theory.

The present work presents a citation analysis of authors and research areas using bibliometric tools on 22 NTFSTJ. The aim is to visualize the trends that connect the main leaders of the field and the top citing research areas of each publication. When contrasted with related bibliometric analyses, the originality of the present research lies in the citation search process, which uses not only the direct results from the WoS but also the citation report feature of the database source, adding an in-depth perspective of who and from where the citations of the NTFST journals are coming. In addition, three aggregated results tables are created to see the accumulated figures of the topics of interest. Along with the natural results of compiling and representing large amounts of data in convenient tables, this study offers additional information, including the visualization of connections between researchers and the top research areas that explore fuzzy techniques along with exploring new opportunities for applications in emerging research fields.

The rest of the paper is structured as follows. Section 2 presents the methodology followed for the gathering and treatment of data. Section 3 presents the retrieved structured information, first introducing results for each of the selected NTFSTJs and offering three tables, including the aggregated results. Finally, Sect. 4 presents the concluding comments of the general analysis.

2 Methodology

The scope of bibliometric studies has grown exponentially in recent decades. The main advantage of employing a bibliometric approach in a specific analysis relies on the methodical quantification of the information concentrated in scientific databases [32, 33]. This representation allows a quick understanding of the connections, relations and impact of the studied phenomena, thus proposing both a clear picture of all the relevant elements that shape the field and the areas of opportunity that allow the possibility of generating novel research and synergies (see, e.g., [34]).

When building a bibliometric analysis, a comprehensive, structured and replicable methodology is essential [35]. This study retrieves data from the Web of Science (WoS) scientific database, specifically from its Core Collection. The selection of the WoS database arises from the need for clear and reliable information for constructing bibliometric indicators. It is considered one of the most rigorous and complete scientific databases, therefore ensuring robust information (see [36]). The Core Collection of the WoS includes over 1.5 billion cited references starting from 1900, more than 74.8 total million records, over 21,100 unique global journals and 254 disciplines. To improve the specifications and reproducibility of this study [37], the consulted WoS Core Collection includes the following sub-datasets and coverage periods: Science Citation Index Expanded (SCI-EXPANDED), 1900—present; Social Sciences Citation Index (SSCI), 1900—present; Arts & Humanities Citation Index (A&HCI), 1975—present; Conference Proceedings Citation Index-Science (CPCI-S), 1990—present; Conference Proceedings Citation Index-Social Science & Humanities (CPCI-SSH), 1990—present; Book Citation Index-Science (BKCI-S), 2005—present; Book Citation Index-Social Sciences & Humanities (BKCI-SSH), 2005—present and Emerging Sources Citation Index (ESCI), 2015—present.

2.1 Raw Data Retrieval Process

The following general procedure was performed to collect the information needed for the citation analysis presented in this study. Step 1. The names of each of the selected NTFSTJs, e.g., fuzzy sets and systems, were introduced as keywords in the main search function of the WoS database portal and publication names were searched. The timespan of the search was limited from 1900 to 2018; this range included data until December 31, 2018. Step 2. The resulting search was then refined by articles, reviews, letters and notes, thus limiting results to citations from these transparent sources. Step 3. From the filtered results, a Citation Report was created using the homonymous function in the WoS portal. Step 4. Next, a second filtering, again for articles, reviews, letters and notes of the results, was performed. Step 5. The resulting data were by this point the raw material used for the generation of the following bibliometric analyses. Step 6. From here, two basic extractions of information were generated: on the one hand, the complete list of authors’ papers that cited the selected journal and on the other, the complete list of research areas’ papers that cited the selected journals.

The raw data retrieval process was performed from March until September 2019. During these 7 months, most of the time was used in the thorough revision by name, surname and affiliation of the obtained data for authors. The need for this double checking was the fact that the raw retrieved data included in some cases authors that shared the same initials and surname, therefore aggregating their combined results and positioning them in top places. For example, suppose an author had the common initials and surname “J Smith”; this author’s complete name could be “John Smith” or “James Smith”, but the raw data coming from the database combine the results for author J Smith into a single result, which in our case resulted in a critical miscalculation. This double check was not performed for the research areas, as they included consolidated information previously refined in the WoS.

2.2 Aggregated Initial Results

To fully dimensionalize the obtained results, we first proposed to describe the general scope of the initial aggregated information following the previously defined steps.

Following the raw data retrieval process specifications, for Steps 1 and 2, a total of 62,379 total papers (TP) were found for the 22 NTFSTJ. The initial aggregated results for Steps 3 and 4 show a concentrated number of 1,307,070 total citations (TC) in 448,852 citing papers (CP). Table 1 presents the selected narrowly targeted fuzzy set theory journals with the initial aggregated results from the previously mentioned steps and some performance indicators.

Table 1 Selected narrowly targeted fuzzy set theory journals
Full size table

Please observe that the studies published by these 22 journals present an average of 16.68 citations per document, an average impact factor of 3.52 for 2018, and 3.35 for the 5-year impact factor. Table 2 presents the general citation structure of the papers citing the 22 selected NTFSTJ to visualize the current impact of the published research.

Table 2 General citation structure of the papers citing the selected journals
Full size table

The first publications of the selected NTFST journals appeared in 1975. Specifically, 50 publications (TP) were published, of which 38 corresponded to INS and 12 to IJGS. On the other hand, the first citing papers (CP) were published in 1982, and a detailed exploration revealed a total of 220 CP for the selected journals distributed in the following fashion: 3 for C&S, 44 to FSS, 120 for INS, 37 for IJGS and 16 for Kyb. Figure 1 shows the evolution of the TP and CP of the 22 NTFSTJs over time.

Fig. 1
figure 1

Number of annual publications citing the selected fuzzy set theory journals

Full size image

3 Results

The collected and treated data are presented in Tables 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24. Each table compiles the top 40 authors and research areas citing the 22 NTFSTJ. Please note that the authors and research areas are ranked by the sum of total citing papers. The resulting tables show 401 different authors with 56,014 total papers and 76 research areas with 813,295 papers.

Table 3 Applied soft computing
Full size table
Table 4 Cybernetics and systems
Full size table
Table 5 Expert systems with applications
Full size table
Table 6 Fuzzy optimization and decision making
Full size table
Table 7 Fuzzy sets and systems
Full size table
Table 8 IEEE transactions on fuzzy systems
Full size table
Table 9 Information fusion
Full size table
Table 10 Information sciences
Full size table
Table 11 Intelligent automation and soft computing
Full size table
Table 12 International journal of approximate reasoning
Full size table
Table 13 International journal of computational intelligence systems
Full size table
Table 14 International journal of fuzzy systems
Full size table
Table 15 International journal of general systems
Full size table
Table 16 International journal of information technology and decision making
Full size table
Table 17 International journal of intelligent systems
Full size table
Table 18 International journal of uncertainty fuzziness and knowledge-based systems
Full size table
Table 19 Iranian journal of fuzzy systems
Full size table
Table 20 Journal of intelligent and fuzzy systems
Full size table
Table 21 Journal of multiple-valued logic and soft computing
Full size table
Table 22 Knowledge-based systems
Full size table
Table 23 Kybernetes
Full size table
Table 24 Soft computing
Full size table

3.1 Individual Overview of the Selected Journals

ASOC stands as a leading journal in the scientific fields of computer science, artificial intelligence, and interdisciplinary applications. The official aim of the journal is to publish quality research on fuzzy logic, neural networks, evolutionary computing, rough sets and similar techniques to deal with real-world problems. For applied soft computing, Pedrycz, Xu and Castillo lead the rank as the most productive authors, with 183, 170 and 129 total papers cited the journal, respectively. In similar studies, Pedrycz and Castillo are ranked as the 1st and 2nd most productive and influential authors for this particular journal, respectively, and Xu is listed as the 4th most cited author by ASOC papers during 2004–2016 [38]. The journal is mainly cited by papers focused on computer science, engineering and operations research management. A congruency between the results presented and the official aim and scope of the journal is observable, as the categories match.

C&S aims to publish documents for computer and information scientists, systems managers, and researchers in fuzzy systems. The journal seeks to share contributions in management, robotics, linguistics, artificial intelligence, operations, and political science. It is observable that the journal has an orientation to humanities and a combination of state-of-the-art cybernetics and systems methods linked to benefits to humankind, as many of the papers citing the journal belong to research areas such as psychology, behavioral sciences, education philosophy and political science. The leading authors citing this publication are Chen with 90 papers, whose main focus resides in engineering applications, Merigó, who published a total of 65 papers and whose main areas of interest are computer science and information systems and Szczerbicki with a total of 61 papers. This last author focuses on computer science and cybernetics.

ESWA is focused on the exchange of information related to expert and intelligent systems linked to applications in industry, government and universities. Fuzzy systems are intrinsically linked to expert and intelligent systems, and the connection between these two fields of knowledge can be tracked back to the foundations of fuzzy set theory [39], which are the reasons for the relevance of the journal ESWA and the critical developments published in it. The results show that most of the papers citing the journal have a robust technical approach to computer engineering and operations research management science; however, sciences such as business economics, social sciences and public administration are also influenced by ESWA. For authors, the one with the most papers citing the journal is Zavadskas, whose publications are oriented to economics, business and engineering, followed by Xu, who focuses on computer science and artificial intelligence, and Kumar, whose research interests are in electrical and electronic engineering, materials and chemistry sciences.

FODM is a journal mainly dedicated to practical issues and theoretical developments of fuzzy optimization and decision-making in uncertain conditions [40]. It promotes advances in fuzzy technology and soft computing methodologies aiming to understand, model and optimize economic, engineering, management, and societal problems. Some of the main authors citing this journal are Herrera-Viedma, whose main specialization is computer science information systems; Liao, whose main publications are on computer science artificial intelligence; and Yager, who focuses on computer science artificial intelligence and computer science theory methods. Some of the most citing research areas fall in technical categories such as computer science, engineering and mathematics. Business economics is 5th in the ranking; however, aside from that area and social sciences, the other topics show a more technical orientation.

One of the classic journals in fuzzy sets theory, systems and applications is FSS. The journal was launched in 1978. Its scope covers advances made on a wide range of aspects of fuzzy sets, e.g., fuzzy set mathematics, such as category theory, topology, algebra and statistical methods; information technologies, namely neural networks, clustering, data analysis and data fusion, pattern recognition and computer vision; other fields, such as calculus of fuzzy numbers, measures of information content, aggregation operations, and theory of relations. In essence, the journal publishes relevant developments that build the connection between human representations and the processing capability of computers. For citing authors, the results show Pedrycz in the first position with a total of 608 papers citing the journal, 55% more papers than the following author in the list, Xu with a total of 390 publications, and Mesiar with 324 publications. The results also show computer science, engineering, mathematics and automation control systems as the main research areas citing this journal, which demonstrates a clear technical approach to the papers influenced by FSS.

TFUZZ is also a classic journal in the theory, design and application of fuzzy systems. This journal’s first issue was launched in 1993 with an editorial note by [41] and compiles exploratory and application studies on fuzzy systems. Special emphasis is not only given to engineering applications, but it is also common to find letters, comments and rebuttals connected to published studies. Ref. [42] has recently published a study on the structure and citation landscape of the journal; another interesting compilation is [43] with a celebration article of the 25th anniversary of the journal. For this journal, Pedrycz is also leading the ranking with a citation count of 419 papers, followed by Xu with 365 and Shi with 354 papers citing the journal. For research areas, the specialization of the journal mainly attracts the attention of studies with a focus on computer science, engineering and automation control systems; these top citing areas are followed by several other technical collections.

INFFUS’s first issue was published in the year 2000; despite being a relatively recent journal, it has managed to publish top-cited articles, enabling the journal to achieve the largest impact factor among the 22 journals presented in this paper. INFFUS is focused on publishing original papers, letters and occasional reviews in multisensor, multisource, and multiprocess information fusion applications, architectures and algorithms. The scope of the journal allows the combination of many advances in fuzzy set theory, such as fuzzy decision-making and group decision-making, fuzzy aggregations, fuzzy clustering, fuzzy multiattribute methods, etc. Xu leads the citation ranking with a total of 132 papers, closely followed by Herrera with 110 and Liu with a total of 82. The highest number of papers citing for this specific journal comes from technical research areas such as computer science, engineering, telecommunications, physics, and optics mathematics.

The main purpose of INS is to publish original research in the areas of information, knowledge engineering and intelligent systems and a wide range of related fields of knowledge, such as mathematics, physics, statistics, management science, and biology [44]. The first issue of INS was launched in 1968, and until 2018, it included approximately 9300 documents in the WoS. The journal promotes state-of-the-art developments in areas such as information theory, artificial intelligence and soft computing, including fuzzy systems. INS also publishes applications in manufacturing, machine–man interfaces, finance and economic modeling and optimization. For this journal, Pedrycz is the most citing author with a total number of 459 papers, Xu is the second most citing author of the journal with 430 papers and Herrera is the third with 324 publications. Computer science, engineering and mathematics are the top citing research areas; managerial sciences such as business economics stand in eighth place; and the next is mathematical methods in the social sciences, which appears in last place in the ranking.

IASC published its first issue in 1995 and collected approximately 1067 documents by 2018. The journal has a wide range of interests, such as engineering, computer science, intelligent automation, control, manufacturing, modeling and systems engineering. The main scope of the journal includes original and survey papers in intelligent automation and soft computing, i.e., fuzzy logic studies. The most citing author for this publication is Lucas with 14 papers, followed by Waxman with 13 and Wang with 12. It is observable that the journal has a more technical influence, as the top citing research areas are computer science, engineering, and automation control systems, among others. It is interesting that health areas such as medical informatics, genetics and oncology are included in the top citing research areas.

The IJAR specific scope is the treatment of imprecision and uncertainty in artificial and computational intelligence in both state-of-the-art developments and innovative applications. This journal launched its first issue in 1987, and by the close of 2018, it compiled 1060 articles in the WoS Core Collection. Some of the topics covered by this journal are probabilistic reasoning, possibility theory, random sets, fuzzy sets, rough sets, decision theory, uncertain reason systems such as risk analysis, information fusion, machine learning, computer vision, and intelligent data analysis. Xu is the top citing author for IJAR with a total of 225 papers, followed by Pedrycz with 165, Herrera with 155 documents, Mesiar citing 102 and Yager with 92. Computer science, engineering, mathematics, operations research and automation and control systems are some of the top citing research areas for this journal.

IJCIS is a journal that launched in 2008 and had 980 documents in the WoS by the end of 2018. This periodical is fully oriented toward publishing original papers in aspects of applied computational intelligence, with special emphasis on papers focused on fuzzy logic, neural networks, evolutionary computation and probabilistic reasoning, and a wide range of applications, including data science, data mining, system control and bioinformatics. The results show that this journal’s top citing authors are Wang with 77 papers, Wei with 70, Liu with 67, Herrera citing 65 and Xu with 64 documents. For research areas, computer science, engineering, mathematics, and operations research management science are some of the top citing fields in IJCIS.

The IJFS journal publishes original research, reviews, surveys and expository essays on theory, design and application of fuzzy systems, soft computing systems, gray systems, and other theory systems [45]. This journal publishes 6 issues per year and compiled approximately 980 documents in the WoS from 2008 until 2018. Some of the top citing authors are Wang with a total of 88 documents, Wei with 87, Merigó with 73 papers, Xu with 56 and Liu with 52. IJFS is the official journal of Taiwan Fuzzy Systems Association (TFSA). The main citing research areas of this IJFS are computer science, engineering, automation and control systems; some other interesting research areas are oceanography, instrument instrumentation and geology.

IJGS primarily publishes original research, book reviews, bibliographies and letters on system science, in this case, concepts’ methods, principles, theories and experimental results on a wide range of systems. The major focus of the journal is on uncertainty, randomness, fuzziness and imprecision, learning systems, complex systems, and inductive and deductive reasoning about systems. Its objective is to publish novel transversal research that is applicable to diverse phenomena. IJGS started publishing in 1974 and had compiled approximately 1600 documents in the WoS. Xu is the most citing author, with a total of 202 citing documents, followed by Yager with 166, Wang with 109, Pedrycz with 95 and Wei with a count of 85. Computer science is the research area with the most influence by this journal, with a total of 5917 citing documents, followed by engineering and mathematics. It is observable that this journal has a wide range of transversal topics, as some other top citing research areas are business economics, psychology, life sciences and public administration.

The IJITDM periodically focuses on publishing theoretical and empirical original research on information technology and topics on decision-making techniques [46]. This journal provides a forum for researchers on a wide range of topics, such as artificial intelligence and decision-making, data mining, data clustering, decision support systems, fuzzy logic and the internet, group decision-making and software, information overload, and multiple criteria decision-making in information technology, optimization systems and web-based language, among other topics. The IJITDM first issue was launched in 2010, and it had 786 documents cataloged in the WoS, information through 2018. The top citing authors are Xu with 103, followed by Zavadskas with 100 documents, Herrera-Viedma with 69 and Kou and Wei with 53 documents each. Research areas with the most influence are computer science, operations research management science, engineering, business economics and mathematics. The transversal applications of this journal’s publications are evidenced by the top citing areas, such as healthcare sciences, geology, forestry and oceanography.

INT is a specialized publication that aims to distribute new developments in the field of intelligent systems such as analysis creation, information retrieval, and man–computer interactions, among others [47]. The journal clearly identifies its readership as computer scientists, engineers, cognitive scientists, knowledge engineers, logicians and information scientists. Its first issue was introduced in 1986 with an editorial of Ronald R. Yager. By 2018, this journal exhibited 1818 indexed documents in the WoS. The top citing authors for this journal are Xu with 368 documents citing, Yager with a total of 202 documents, Herrera with 174 publications, Herrera-Viedma with 141 total articles and Merigó with 136 documents. The journal has clearly identified their audience as the top citing research areas are computer science, engineering and mathematics. Interesting top citing research fields appear, such as neurology and nuclear science.

IJUFKS is a specialized journal for imprecise, vague, uncertain or incomplete information studies [48]. The scope of this publication is theoretical and methodological research on diverse kinds of methods for the manipulation of imperfect pieces of knowledge. The journal excludes results of pure mathematics such as fuzzy topology and fuzzy algebra or simple applications of published theoretical results. Its published applications cover intelligent systems, decision-making systems, information fusion and retrieval, data science, and robotics, among others. Some of the methods covered by this journal are nonstandard logics, deep learning, fuzzy data analysis and Bayesian systems. The top citing authors for this journal include Xu with 173 documents, Mesiar with 135 citing papers, Merigó with 116 published works, Wei with 112 and Herrera-Viedma with 95. The top citing research areas include classic computer science, engineering and mathematics, along with other fields such as operations research management science, business economics and telecommunications.

The IrJFS is a specialized journal in fuzzy sets, systems theories and applications. Its main scope is original research in finance and management, mathematics, robotics, uncertainty modeling, soft computing, decision-making, and deep learning, such as artificial intelligence, pattern recognition and image processing. This journal’s first issue was launched in March 2004. By the end of 2018, this journal’s publications in the WoS reached 549. Among the top citing authors for this journal, we can find Wei with 71 documents, Davvaz with 56, Shi with a total of 36, Pang with 28 and Zhan with 20 papers. For top citing documents in the research areas, we can see a clear preference from technical fields such as computer science, mathematics engineering, science technology and automation control systems.

The IJFS disseminates results on cross-section, broad and novel current and potential applications and case studies in the field of fuzzy and intelligent systems. The journal aims to publish works in technical areas such as engineering, civil engineering, manufacturing, industrial, chemical, electrical, biomedical, management and electronic engineering along with soft sciences, social and environmental developments. The journal published its first issue in 1993 with a foreword from L. Zadeh. By the end of 2018, the number of documents indexed in the WoS for this journal was 4128. The top five citing authors for this journal are Xu with 138 papers, Liu with 122, Davvaz with 115, Wang with 115 and Wei with a total of 92 citing publications. The top citing research areas for this specific publication are technically oriented fields such as computer science, engineering, mathematics and automation control systems.

JMVLSC is mainly focused on the publication of multiple-valued logic and soft computing works. The related fields this journal publishes are those that are tolerant of imprecision and uncertainty, and deal with partial or incomplete information. Some of the specific issues that this journal has as its scope are logic, circuits, deep learning, fuzzy logic applications and theoretical aspects, soft computing and data and machine learning. The journal records start in 2003 with the ninth volume, and by the end of 2018, a total of 691 documents were listed in the WoS scientific database. The top citing authors are led by Davvaz with 92 published publications citing the selected journal, followed by Herrera with 79, Kahraman with 31, Garcia with 29 and Akram with 27 papers. The research areas most influenced by this publication are those falling in computer science, mathematics and engineering; some other interesting top citing areas are neurology, linguistics and psychology.

The KNOSYS journal publishes original developments on systems using knowledge-based processes, models, techniques and methods to support decision-making, learning and human action [49]. The journal emphasizes machine and computer developments and manlike interactions. This journal’s main scope encompasses big data methods, intelligent human interfaces, recommender systems, support decision systems, artificial and deep learning systems and knowledge engineering. The journal also publishes multidisciplinary research on related human endeavor fields ranging from financial technology to health science. KNOSYS published its first issue in 1987, and until the end of 2018, the journal indexed over 3550 articles in the WoS database. The top citing authors for this journal are Xu with 261 papers, Herrera with 145, Acharya with 142, Pedrycz with 132 and Chen with 127 articles. A wide range of technical research areas cite this journal, including computer science, engineering, operations and research management science, as well as medical informatics, biomedicine and public environmental occupational health.

Kybernetes is a multidisciplinary journal that shares original research on systems thinking, more specifically on cybernetics and systems thinking. The aim of Kyb is to close the gap of the understanding on human, social and technological issues, its impact on society and its interdependence. Its wide-ranging publications include studies in diverse domains of the human, social and ecological sciences with a novel focus on sustainability studies and works related to the impact of technology on humanity and society. The first issue of this journal was published in 1972, and by the end of 2018, 3356 documents were included in the WoS. Being a journal with a wider scope, the top citing authors are dissimilar to other specialized journals. The top citing authors for Kyb are Cherruault with 105 citing documents, Lin with 60, Navimipour with 54, Rach with 52 and Wei with 50 articles. The top citing research areas for this journal are similar to the rest of the journals presented, with computer science, mathematics and engineering being the top 3 citing research fields.

SC is a specialized journal on systems solutions and applications based on soft computing techniques, more specifically on evolutionary processes, genetic programming, neural and network science, fuzzy sets, fuzzy systems and chaos systems [50]. The main purpose of the journal is to be the link between diverse disciplines, fostering comparison, extensions and new applications of soft computing techniques. SC’s first issue was published in April 1997. By the end of 2018, the WoS included 3342 documents from the journal. The top five citing authors are Herrera with 172 documents, Davvaz with 115, Zhan with 103, Pedrycz with 88 and Herrera-Viedma with 87 citing articles. Computer science, engineering, mathematics, operations research management science and telecommunications are some of the top citing research fields for this publication.

3.2 Aggregated Analyses of the Selected Journals

Tables 25, 26, 27 present aggregated overviews of the individual results. Table 25 is constructed considering the information of authors presented in Tables 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24. The 40 resultant authors are ranked by the sum of their citations of the selected journals, i.e., total citations. Moreover, the table presents the number of times each author appears in the 22 individual tables and depending on the case, the number of times they occupy the first, second or third position. Additional information including the maximum position reached by the author for any of the NTFSTJ, the maximum number of citations of any of the selected NTFSTJ and the most cited journal of the author is also included.

Table 25 Top 40 in-depth authors analysis
Full size table
Table 26 Aggregated results by author and journal
Full size table
Table 27 Aggregated results by research area and journal
Full size table

In general, Xu leads the ranking with a total of 3563 citations, which represents 35% more documents than the next author, Pedrycz, with a total sum of 2646 citations, followed closely by Herrera with 2162 documents and Merigó with a total of 1394 studies citing the journals. Xu is in the first place of the individual results 5 times more than Pedrycz with a total of 4 first places and Wang with 2. For the rest of the top citing authors, only 5 achieve a first position in the individual results. Merigó leads the “times appearing” count by appearing in 18 of the 22 individual tables, followed by Xu, Herrera and Liu with 17 appearances each. For the individual maximum number of citations of a journal, Pedrycz leads the count with 608 total documents citing FSS, followed by Xu with a sum of 430 documents citing INS, Shi with 354 citations of TFUZZ and Herrera and Mesiar, both with 324 total publications citing INS and FSS, respectively.

Tables 26 and 27 present aggregated results ranked by top citing authors and top citing research areas for each of the 22 NTFSTJs, respectively. As observed before, Xu, Pedrycz, Herrera, Merigó and Herrera-Viedma are the leading citing authors, while computer science, engineering, mathematics, operations research and management sciences are the leading research areas citing the selected journals. Both tables show the influence that certain academic journals have on the ideas presented in articles, letters, reviews and notes published by the discussed authors and research areas.

4 Conclusions

This study presents a citation analysis of 22 narrowly targeted fuzzy set theory journals using bibliometric tools for the quantification and presentation of the retrieved data. The aims and scope of the selected journals clearly follow a fuzzy systems alignment, as demonstrated in [9, 24]. The analysis is focused on top citing authors and research areas. The aim is to advance the knowledge on networks, connections and influence that the selected journals present on both academics and research fields. These connections assemble a general picture that allows the visualization of possible synergies and emerging scientific opportunities for scholars and practitioners.

The data were retrieved from the WoS scientific database. This source was chosen because of the robust and rigorous indexing of scientific material along with its vast collection of specialized materials. The retrieved data were collected by searching for the publication name in the main search engine of the WoS database. Then a series of filters were applied to identify documents relevant for this analysis. From there, a citation analysis report was retrieved focusing on both top citing authors and most citing research areas. The retrieved raw data were then collected and treated. For the research area information, the process was straightforward; however, deeper insight into the author results was required, mainly because of the overlap in common surnames. After the treatment of the raw data, the results were presented accordingly.

Aggregated results for top citing authors of the 22 NTFSTJ placed Xu in the first position of the ranking, with a total of 3563 citing documents and INS being the most cited journal for this author, followed by Pedrycz with a total of 2646 documents and FSS being his most cited journal, Herrera with 2162 papers and INS as his most cited journal, Merigó with 1394 and INS as the top citing periodical, Herrera-Viedma with 1378 articles and INS his most cited publication, Yager in sixth place with 1340 and FSS as his top cited journal, Chen with 1314 documents and ESWA as his top cited journal, Liu with 1303 and FSS as his main influencing periodical, Wei with 1216 and ESWA being his most influential journal and Mesiar with 1151 studies and FSS as the top citing journal. These aggregated results added the number of documents citing a specific journal; therefore, 1 document may be citing 2 or more of the selected publications; hence, the total sum of the documents presented might be larger than the number of papers produced by these academics. It is interesting that these 10 authors represent 50% of the total citing documents for the 34,599 top 40 aggregated results. In addition, INS is the most cited journal from the top 40 citing authors with 5781 total documents and 33 of the 40 top citing authors, followed by FSS with 4928 citations with 24 top citing authors, ESWA with 3222 and 20 top citing authors, TFUZZ with 3117 and 16 top citing authors, and INT with a total of 2478 citations with a total of 25 top citing authors.

The results obtained for research areas clearly identify computer science as the top citing research field, with a total of 283,438 documents citing ESWA as the most cited journal for this specific field. Engineering is the next top citing area, with a total of 163,183 documents citing this field of knowledge. ESWA is also the top citing journal for this area. The research area of mathematics has 63,198 papers cited, and FSS is the most cited journal. Operations research & management science is the next top citing field, with 47,763 citations, and ESWA is the most cited journal of this research field. Automation control systems is the fifth top citing research area, with 37,145 studies, and TFUZZ is the top citing publication. Business economics follows with 23,511 studies, and ESWA is its top cited reference. In general, 59% of the total studies citing the selected journals fall in two main disciplines, computer science and engineering. From these results, 38% of the top 40 studies belong to the computer science field. Of the 22 selected journals, ESWA, INS, FSS, ASOC and TFUZZ are the 5 most cited journals from the top 40 aggregated results of the research fields.

This study tries to shed light on the use of citation analysis for the in-depth characterization of bibliometric studies. The motivation of this study lies in the proposition of [9], where fuzzy set theory journals are clearly identified by bibliometric indicators; more specifically, the percentage of fuzzy set theory papers published in a journal. The aim is to provide a picture of the citing authors and research areas of the selected NTFSTJ based on a characteristic search process involving the citation report feature of the selected database and a careful manual revision of the obtained data. These characteristics allow us to first identify which sources influence the leading authors of the field the most, therefore providing a broad picture of the extension of the ideas presented in the journal. Second, these characteristics allow us to view which research areas are citing which journal; this result is interesting, as the reader can observe which kind of public readership the journal is attracting and identify, beyond the three main positions, which journals show a citation attraction for areas that might be involved in new opportunities and fields of knowledge.

The present study has some limitations. First, the extracted raw data pool of documents is mainly extracted from the WoS. Second, to achieve the clear identification of authors with similar surnames in this analysis, a manual detailed scrutiny was required; however, some documents might have been omitted due to the validation process. Third, the current counting method for CP is one on one; i.e., a citing paper is added to the accumulated sum if there is one or more citations for any of the NTFSTJ, which may hinder some data and possibly affect some rankings. Further research is required to tackle these limitations, especially regarding the selection on the sources of information, as other scientific catalogs should be included [51] to expand the analysis, e.g., SCOPUS and Google Scholar, to expand the number of documents to be analyzed in the research, adding other NTFSTJs that this study has not covered. Also, the need for a deeper content analysis is evidenced, i.e., further developments are encouraged to analyze the proposals in terms of most cited papers or most influential ideas and for extracting the citations of a document to eliminate the limitation of the one-for-one current method of counting.

Fuzzy set theory and its applications and extensions present exponential growth in both their publications and impact [52]. The quantification and analysis of this expansion is interesting because it focuses on the academic leaders and research fields that are being influenced by the documents disseminated by the journals. The connections, networks and research paths are additional results of this study that allow the visualization of possible synergies and opportunities for research in emerging fields. This study intends to shed light on the scope of fuzzy set theory and its evolution, its influence on diverse minds and fields, and its prospects for future developments.