Review on the Relationship of Absorptive Capacity with Interorganizational Networks and the Internationalization Process

*ere is evidence about the international competitiveness of Small and Medium Enterprises having a close relationship with their absorptive capacity and internationalization networking, and for that reason, it is relevant to find out the main trends in this field of knowledge. *e objective of this study is to provide a bibliometric analysis of the status of the existing research in the field to recognize main topics and help identify research gaps. *is study was done through a review of 1,710 documents published about this relationship from the Scopus and Web of Science databases (1994–2018), using as processing software application that employs two combinations of terms associated with Boolean operators. *is was taken into account in order to optimize the accuracy of the search and to facilitate large data capture.*e results show that these studies are in a period of high production and concentrated in a few countries and researchers’ networks in the United States, the People’s Republic of China, and some European countries. Moreover, the trend words used by researchers are those which link absorptive capacity with networking, open innovation, and firm performance.


Introduction
During the current century, the business's environment shows CEOs, entrepreneurs, researchers, and even politicians that firm's international competitiveness depends basically on innovation [1][2][3][4][5][6][7][8][9][10]. is is why it is obvious that the focus is placed on how to help firms and in consequence to increase national growth. While SMEs engaged in international markets tend to be more productive and innovative than those which are not, they can further improve their performance through internationalization [11]. In most countries, SMEs account for a significant proportion of employment and they are the ones which need more attention. It is common for the following question to be present into the mind of those actors: which is the best way to bring innovation to firms, especially those which have less resources, SMEs?
ere are many theories about the innovation process and the main situations and activities that could reach that goal. One of these analyzes the positive moderating effect that the interorganizational networks and the absorptive capacity (ACAP) have on the innovation process. For SMEs, these networks are critical in order to give them the chance to be competitive. On the contrary, not all firms have the same ability to process this new data and transform knowledge into business opportunities. e capability of having this ability was described by Cohen and Levinthal [12] years ago and named as absorptive capacity. " is concept exhibits a significant importance to analyze the business ability of seizing the external knowledge, combining it with its own domains and generating a dynamic learning and feedback that favors the innovative process and subsequently the maintenance of competitive advantages" [4]. So, if SMEs want to be competitive, the key factors in the innovation process needed for this are (a) the interorganizational networks and (b) ACAP.
In this vein, the concept of sustainability is deemed as a relevant concept in order to achieve competitive advantage. Some authors, Fichman [13] and Wink [14], analyzed the sustainability concept with interorganizational networks and ACAP although it was not until Van Wijk et al.'s study [15] that examine how knowledge, organization, and network level could impact on organizational knowledge and transform into sustainable competitive advantage. Subsequent research [16] has emphasized the importance of corporate sustainability to achieve competitive advantage. Especially for SMEs, the difficulties in international markets make the analysis of the impact of sustainable goals necessary. In such a way, the report by the World Trade Organization is an attempt to address with sustainable development goals included in Agenda 2030 [17].
Although the research field is deemed as interesting and flourishing due to the existing volume of the literature, there are few bibliometric studies that have analyzed ACAP as a knowledge area. While relatively recent studies based on the results of bibliographic analysis concerning ACAP in the management context exist [18][19][20], some others assess if there is a relationship between open innovation and ACAP [15] or analyze the role of ACAP in the relationship between strategic alliance portfolios and innovation performance [21], and none of them has addressed ACAP's relationship in networking and internationalization. e main motivation of this paper is to cover the current gap regarding ACAP's relationship in networking and internationalization so as to help researchers to identify future research trends.
For the reason stated above, the purpose in this study is to show a qualitative and quantitative examination of the dynamics of global research in the last 25 years (from 1994 to 2018) to determine the current state of scientific production about "ACAP's relationship" with networks and internationalization in the process to access external knowledge. To achieve this goal, bibliometric methods were employed. As Capobianco-Uriarte et al. [22] mentioned, bibliometric techniques allow the identification of the main elements of a research topic. In addition, it enables to detect the most productive agents in the research field, authors, institutions, or countries. In such a way, the main driving force behind a field of research is shown [22]. erefore, the contributions of this study are twofold. First, this paper identifies the main trends in ACAP research, revealing the evolution of the field. Second, the link of the main concept to networking-internationalization issues is highlighted, suggesting future research challenges. In the next section, a theoretical view of the main approaches of the concept of ACAP is presented. After that, the bibliometric methodology used is explained. Subsequently, the main results are shown and analyzed. Finally, conclusion section is presented.

Theoretical Background
Innovation is defined as the "implementation of a new product or service or their significant improvement, a better or improved process or marketing method or a new organizational method in business practices" [23] and has just one cause of origin: knowledge. Knowledge is the key factor of the innovation process, and it is possible to obtain it from two different sources: internally, through the own effort of the firms to research and develop (R&D) to improve the process, which is not easy for SMEs, or externally, when the firm tries to obtain it from markets, competitors, providers, universities, and other actors. A huge investment is not usually necessary to attain it, just having the tools to get it from those actors or "nodes" [24]. One and maybe the most important tool to obtain external knowledge is to create or to be part of strong interorganizational networks [9,[24][25][26].
ese networks are critical for SMEs in order to give them the chance to be competitive. On the contrary, not all firms have the same ability to process these new data and transform knowledge into business opportunities.

Interorganizational Networks.
e possibility of accessing knowledge through interorganizational networks is usually the only way that SMEs have to improve their competitiveness [27] by attaining a competitive advantage [28]. e level of internationalization, the size, and the intensity and heterogeneity of the networks are relevant factors that define if a network could help SMEs more or less in the process to access external knowledge [24,[29][30][31][32][33].
All the ties of the network are positive, and all have their own value in the innovation process, but if firms can extend their "nodes" to other countries, the benefits will be greater. In fact, the positive influence of SMEs' international experience can be verified in studies conducted by Chetty and Blankenburg-Holm [34], and Chetty et al. [9,24] developed a matrix that allows identifying the insertion of companies according to the improvements developed and their level of internationalization. To find out what is happening, developing, creating, consuming, and/or studying in international markets and/or environments is critical, especially if the firm hopes to succeed with their products or services. e size of the network gives firms the possibility of establishing relations with many and different actors where new knowledge may be obtained. In this way, they can improve and/or increase the benefits [24,29,30,[35][36][37][38] in two ways: (1) because more "nodes" means more sources and (2) because with more contacts and links, the firms have the chance to obtain different points of view on the same topic [30,38] and this helps the firm to define what is the real value for it and what is not in order to be more competitive.
Finally, the intensity of the connection of the SMEs with the actors or "nodes" of the network is an important factor that gives them more chances to get, discuss, and/or test the ideas or the new concepts [29,36]. Demirkan and Demirkan [31] and Brink [33] show that the intensity of connections as well as the heterogeneity of knowledge and experiences lead to innovation. Capaldo [39] points out in his study where he addressed the way three furniture intensive design companies managed their networks and that the ability to integrate a wide periphery of weak and heterogeneous ties and a core of strong connections with leading companies provides a fertile ground to obtain and sustain competitive advantages. e positive effect between the innovation process and interorganizational networks, especially in relation with international "nodes," is present in many papers [34,39,40].

Absorptive Capacity.
e concept of ACAP [12] analyzes the firm's business capability to process external knowledge, combining it with its own internal sources in order to develop new business opportunities. It is a process which includes different and successive stages: it begins with the recognition of the value of the external knowledge; the next step is its evaluation with reference to the firm's own skills to use it for its benefit; and finally, the decision about if it is possible to transform that external knowledge into business opportunities [41,42]. In the words of Zahra and George [29], ACAP constitutes "a group of organizational routines and strategic processes by means of which companies acquire, assimilate, transform, and exploit knowledge with the intention of creating a value." As SMEs have few chances of creating innovations by themselves and need external knowledge to do so, ACAP is a critical factor and allows the firms to be more proactive and more competitive. So, this process includes not only the assimilation of external knowledge but the creation of new knowledge: a new business opportunity (innovation).
In the construction of ACAP, the human capital that the firm has is important [43]. It is the "gate" through which external data enter [12], and it is important because of its qualification and its experience in business analysis and because it knows the specific capabilities of the firm to use external knowledge [44,45]. erefore, the individual characteristics of this human capital contribute to the development of ACAP and especially so if it has international experience obtained through interorganizational networks [44].
ere are other key factors that firms can handle to improve ACAP. e main one is of course the capacities of internal creation of knowledge from external data, but the information management systems and the practices for social integration which can place external knowledge close to the "gates" are important too [46][47][48].
ere are many studies which prove that ACAP has a positive moderating role in the ACAP innovation process. For instance, Cassol et al. [49] show empirical results referring to Brazilian industries; Kohlbacher et al. [50] regarding a business cluster in Central Europe; Tsai [51] focusing on Taiwanese companies; and Kotabe et al. [8] concerning the international background and the performance of senior executives in China.
Despite the importance of ACAP in the SMEs context and its connection to interorganizational networks in the internationalization process, there is no analysis of scientific research regarding this relationship. us, our work allows the evaluation of research and the analysis of its impact on the academic world. erefore, a bibliometric analysis of ACAP should take into consideration the terms regarding interorganizational networks.
It is also used in research performance evaluation, especially in university and government labs and also by policymakers, research directors and administrators, information specialists and librarians, and scholars themselves.
e results of a bibliometric analysis allow the analysis of past research and the detection of future trends of research. Bibliometric analyses require a bibliometric data source [69].
As stated above, this research adopts the bibliometric analysis technique as the method to conduct this review. As in bibliometric analysis carried out by Terán-Yépez et al. [70], this paper follows five steps: (1) definition of the field of study, (2) database selection, (3) research criteria adjustment, (4) codification of recovered material, and (5) examination of the information (Figure 1). e results of bibliometric analysis may vary depending on the database used [71]. Web of Science, produced by Clarivate Analytics, and Scopus, created by Elsevier, are the two most used bibliometric data sources, together with Google Scholar by Microsoft. Harzing and Alakangas [72] suggest that all three databases provide sufficient stability of coverage to be used across five major disciplines (Humanities, Social Sciences, Engineering, Sciences, and Life Sciences) although Martín-Martín et al. [73] conclude that, in all areas. Google Scholar citation data are essentially a superset of Web of Science and Scopus, with substantial extra coverage. However, Google Scholar lacks transparency, suffers from data quality problems, and is very difficult to use for large-scale analysis [69]. For these reasons, the two most commonly used bibliometric data sources, Web of Science and Scopus, will be used in this study. In this analysis, a search formula was used with two combinations of terms connected together with Boolean operators included in order to optimize the accuracy of the search and to facilitate large data capture, TITLE-ABS-KEY ("absorpt * capa * ") AND ("internationali * ation" OR "network * ") to extend the concept of ACAP in the context of internationalization. e search was limited in the time period encompassing 1994-2018 years, due to the fact that the first year of the search coincides with the contribution by Cohen and Levinthal [12], generally accepted as the founding paper, where ACAP is defined as "the ability of a firm to recognize the value of new, external information, assimilate it, and apply it to commercial ends." Complexity e search in both databases (Scopus + Web of science) was undertaken in July 2019. e selection criteria for the scientific document sample to be analyzed through the proposed bibliometric analysis follow the guidelines of Capobianco-Uriarte et al. [22]. Only scientific articles and reviews, including open access and nonopen access documents, were taken into account in the search.
Data are retrieved from two major databases, Scopus and Web of Science. Once the data were treated, the analysis was carried out using VOSviewer [69], which is a powerful tool that enables scientific maps to be illustrated, visualized, and discovered. Finally, the results are shown by descriptive analysis and content analysis. First, the descriptive analysis shows the most productive agents in the research field, authors, institutions, or countries [22]. And then, the content analysis in a temporal evolution of the fields considered was studied in terms of timelines for the keywords [74] and keywords networks [75].
VOSviewer software [75] was selected for the creation of bibliometric networks and their visualization as bibliometric maps of the topic. VOSviewer is a freely accessible software for academic nonprofit use, offering an easy and fast analysis tool, specifically for authors, institutions, countries, Step 1 Definition of the field study Step 2 Database selection Step 3 Research criteria adjustment Step 4 Condification of recovered material Step 5 Examination of the information  coauthorship, and keywords cooccurrence networks, and combining visualization and clustering techniques [75]. Although VOSviewer can be used to construct and visualize bibliometric maps of cooccurrence data, the software tool does not allow any cooccurrence matrix from the bibliometric data to be extracted and built [76]. is software builds a similarity matrix from a cooccurrence matrix using a similarity measure known as the association strength [75]. Furthermore, the software tool has no preprocessing modules to prepare the data for later analysis (as duplicate items detection, time slicing, and data and network reduction). us, an external process is needed [77]. Moreover, VOSviewer mapping technique builds only on twodimensional map, in which the elements are located. In such a way, the distance between any pair of items reflects their similarity as accurately as possible, while other software features 3D networks (as Gephi or Cyvision). Despite the limitations presented by VOSviewer, this visualization software has been used to construct scientific maps from network-based data due to its powerful user graphic interface, which allows to examine the generated maps easily, being used for a diverse range of disciplines and scientific fields [77].

Descriptive Analysis.
ough the concept of ACAP was introduced in 1990 by Cohen and Levinthal [12], belonging to the Carnegie Mellon University and the University of Pennsylvania, respectively, Carlsson and Jacobsson [78] published the first article linking ACAP with topics of internationalization and network formation. From 1994 to now, a large number of articles have been published in the Web of Science and Scopus, a total of 1,710. Linking ACAP with internationalization and network formation has been addressed by 3,297 authors, whose works have been published in 553 academic journals indexed in databases considered between 1994 and 2018 (Table 1). Carlsson and Jacobsson [78], who related ACAP with topics of network formation, belong to Case Western Reserve University (United States) and Chalmers University of Technology (Sweden), respectively. In addition to this, these authors analyzed networks of agents interacting in a specific area of technology (the Swedish automotive sector) under a particular institutional infrastructure and confirmed that this constitutes a useful unit of analysis not only for innovation and diffusion studies but also for work related to technology policy. Although the authors did not directly address the issue of internationalization, they supported the early identification of important developments and increasing the economy's ACAP as being important aspects of public policy.
ere were few publications during the first 5 years (1994-1998), but one of them [79] has stood out for the relevant number of citations obtained (Table 2).
Szulansky's study [79] evidences that the major barriers to internal knowledge transfer are knowledge-related factors such as the recipient's lack of absorptive capacity, causal ambiguity, and an arduous relationship between the source and the recipient, contrary to conventional wisdom that primarily blames motivational factors. In the following 5 years (1999)(2000)(2001)(2002)(2003)(2004), the subject took on a greater dimension as to the volume of publications, though this remained at a low level, not surpassing 38 publications annually. At the beginning of this period, in 1999, the first article appeared which related ACAP with topics of firm internationalization [80]. In this study, Meyer-Krahmer and Reger [80] highlighted the changes in the innovation strategies of large multinational companies, included in the agenda of  Complexity technology policy in Europe, a greater emphasis on collaboration and mobility outside Europe, strengthening the attractiveness of the European Union for foreign investment in R&D and the absorption capacity of R&D organizations in Europe. In 2011, there were more than 100 annual publications. In 2016, the production duplicated this with more than 200 publications. e evolution of the scientific production in ACAP related with internationalization and network formation shows an exponential growth from its beginnings in 1994 ( Figure 2). e number of journals interested in these topics increases from 2011 and 2012.
is increase is significant, being an average of 100 different journals per year which publish on this matter. A greater number of authors (681) and of journals (149) interested in it is noted in the last year (2018). Table 1 indicates a significant number of countries and institutions participating in indexed publications on these topics although initially the average for each one of them is low (less than 27 articles per country and slightly more than 2 per institution).
ese figures are notably modified if we only consider the last 5 years of production (2014-2018) as 56% of the total of articles of the 24-year period are concentrated in these five years (Table 2). e number of authors who have addressed the topic of ACAP in internationalization and network formation has increased polynomially since 2004, having a maximum of annual production in 2011, with the participation of more than 400 authors who published at least one article (Figure 3).
ere were 681 authors in 2018. e authorship average in the articles published on this topic is between 2 and 3 authors per article (2.63). ere is an average concentration of the productivity as to the number of authors who publish. e distribution of frequency of authors according to the number of articles published is the following: 15.20% of the articles are produced by a sole author, 34.04% by 2 authors, 32.57% by 3 authors, and the remaining 18.19% by four or more authors. e journals which published scientific works related with the topic analyzed have also constantly grown, surpassing a hundred in 2011 and reaching 149 in 2018. e distribution of the articles published in the journals indexed in Scopus and Web of Science is very atomized, as there are many journals which publish on the subject and most present a low average of articles published. Taking into account that the first 10 journals in the ranking of volume of articles published (Table 3)  e former is a multidisciplinary social science journal. e latter is a specialized journal. International Business Review provides a forum to share the latest developments and advances in the knowledge and practice of international business. However, Journal of Knowledge Management is published by Emerald and is also a specialized journal, dedicated to all aspects of managing knowledge in organizations. It is highlighted that Strategic Management Journal, although it is neither the most productive journal nor the most cited within the top ten journals group, has published 2 of the 5 most cited articles on the topic (Table 4).
Spanish researchers stand out in the ten most productive authors in ACAP related with internationalization and network formation. e most cited authors are American and Dutch researchers. Molina-Morales F. X., of the University Jaume I, is the author who has the greatest quantity of articles published and Lavie D., of the University Texas, is the most cited author (Table 5).
It is not only interesting to analyze the productivity of the researchers but also the collaboration networks, which are both intrainstitutional and interinstitutional and 6 Complexity intranational and international. For these purposes, the first author of each of the works published is considered and the author Williams C. is excluded for being the only U.S. reference in a Chinese network and simultaneously representing five different institutions around the world; it not being possible to identify if this reference corresponds to the same person who is affiliated to different institutions or to different people with the same name. Additionally, and there not being another North American node in this Chinese network, this reference is excluded given its scant relevance in the total universe. Taking into account 10 or more scientific documents in common, it is possible to detect the following five scientific networks of different collaborations (Table 6). Figure 4 shows the cumulative time-varying collaboration networks between authors and research groups during

Intranational networks International networks (i) Swedish (i) American (USA)-European-Asian (ii) Spanish (ii) American (Canadian)-Asian (iii) Chinese (iv) British
Source: own elaboration. 8 Complexity the period 1994-2018. is information is useful not only for the visualization of the evolution of scientific production of the networks but also for the changes in the configuration of these collaborative networks. e relevance of the study in the time-varying collaboration networks is highlighted in the work by Viana et al. [81], who claimed the importance of the analysis of the moment in which a network emerges and the identification of the characteristics of the collaborative patterns. In this vein, this analysis allows the stability of the activity of networks. Figure 4 e scientific network with the greatest scientific productivity is national and interinstitutional (Table 6) Within the other transnational networks, it is underscored that, unlike the Spanish network (interinstitutional), there are two intrainstitutional networks: the British network and the Chinese network (Table 6). In 2009, the first British network was created in this field, with Love J. H. (Aston University) and Roper S. (University Warwick), still working (Figure 4(b)). e Asian network, with Guan J. C. and Yan Y., was created in 2014 (Figure 4(d)), both authors belonging to University Chinese Acad Sci (People's Republic of China).
On the contrary, the most productive American author, Lyles M. A. of Indiana University (Table 5), leads the most extensive international network with European links, along with Pedersen T. of the Centre of Strategic Management and Globalization (Denmark) and Volberda H. W. and van Wijk R. of the Erasmus University (Netherlands) in 2008 (Figure 4(b)). In 2016, this network continues working and linked the American-Denmark-Dutch network to the Asian one (   Republic of China), with a Chinese intranational collaboration, Wu X. and Du J. of the Zhejiang University, Liu X. of Xiamen University, and Wang C. and Zhang G. of the Shandong University, and with a European collaboration, Vanhaverbeke W. of the Hasselt University (Belgium) and Duysters G. of the Tilburg University (Netherlands). e Canadian-Chinese (Table 6) network is made up of the Canadian scientists Clercq D. and Zhou L. of the Brock University in 2012 (Figure 4(c)), and then, in 2016, Wu A. Q. of Zhejiang University (Figure 3(d)) joined this Canadian network.
To sum up, 3 collaborative networks worked until 2005 (G 1 � 3) in the field of interorganizational networks and the internationalization process.
ereafter, in 2008, the first international network American-Denmark-Dutch (GJ 2 � 2) was created becoming the first international network. e British network was born in 2009. In 2010, the Spanish network appeared, highlighting for being intrainstitutional and the most productive up until this time (GN 3 � 2). Later, in 2012, the Canadian network emerged. In 2014, the Chinese group was born, and the Canadian and Spanish networks (GA 4 � 2) were extended. Finally, 2014 highlights for the extension of the international network American-Denmark-Dutch-Chinese (GJ 4 � 5) that continues being the unique international network.
Finally, the Chinese scientists Zhao X., Yan Y., and Guan J., along with the Spanish scientists Belso-Martinez J. A. and Diez-Vial I. present the most recent scientific activity ( Figure 5).
Within the 10 most productive institutions regarding publications about the reference topic ( Table 7), nine of them are European and one is Asian. ough the number of articles published by each institution is relatively low (maximum 19, minimum 10, with an average of 14 articles per institution), the Spanish production stands out as it includes four institutions within the 10 main ones (Table 7), which in turn concentrates 40% (64 of 162) of the total of this country (Table 8).
e same situation is presented with respect to the most productive authors ( Table 5), three of them belonging to the first two most productive academic institutions and marking a significant presence from the moment that a growing interest began to be shown toward the topic under analysis (2010-2012), all of them being maintained active at the last record (2018). e institutional networks are based on networks formed among the scientists. e interinstitutional network formed at the international level, which studies the relation of ACAP with internationalization and network formation, presents six main clusters, taking into account 10 or more scientific works in common ( Figure 6). e blue cluster is the most productive inasmuch, as it includes five Spanish academic institutions. As can be seen from Table 7, they concentrate the most productive authors and it shows a relevant network between them. Surprisingly, other two Spanish universities, namely, Granada University and Seville University, are also productive regarding number of articles (Table 7), but there is a lack of linkage between them despite being geographically proximate. e blue cluster, through the University Bologna and Uppsala University, is linked with the green cluster. e green cluster is led by the Copenhagen Business School of the Centre of Strategic Management and Globalization of Denmark, the third research center of highest production at the global level (Table 3). is center is directly linked with British universities (University Manchester, University Nottingham, and University Sheffield), the Dutch universities of the University Groningen (pale blue cluster) and the Tilburg University (violet cluster), and intercontinentally with the Asian university Zhejiang University, fourth at the global level for its productivity (Table 3). is Danish research center and the Chinese university are the most interrelated academic institutions at a global level.
e Zhejiang University is linked with the yellow cluster made up of a British university (University Lancaster), a Canadian university (University Ottawa), U.S. universities (University Indiana, University George Washington, and University Minnesota), and Finnish universities (Aalto University and Lappeenranta University Tech.). Furthermore, Zhejiang University works in collaboration with the red cluster, formed by other Asian institutions (Hong Kong University, University Chinese Acad Sci, and Xi An Jiao Ton University of the People's Republic of China) and European institutions (Erasmus University and Oxford University, among others).
Finally, Zhejiang University is interrelated with the violet cluster made up mainly of European universities, such as Hasselt University, Tilburg University, Maastricht University, and the Natl. University Singapore of south-east Asia. Lastly, the smallest cluster is formed of two European academic institutions, University Groningen and University Granada.
Timewise, the institutions which present the most recent scientific activity (Figure 7) are University Chinese Acad Sci and University Groningen, and within the most productive cluster is the University Castilla La Mancha and the University Politech. Valencia.    is behavior is reproduced in some way in the origin of the researchers and/or institutions involved (Table 5). Almost two-thirds of them (1,095 articles published by researchers and institutions of these countries out of a total of 1,711) have their origin in only 10 countries (Table 8). In addition, the five first countries of the ranking have been responsible for 45% of the total production (773 out of 1,711) and their composition within this universe of 5 is distributed among countries belonging to the European Union (53%), the United States (29%), and the People's Republic of China (18%), which are in turn the three main actors of the global economy and trade.
If we analyze the structure of the research at the country level, taking into account 15 or more works in common, five clusters are noted (Figure 8). Two of the most productive countries are intensely related in the green cluster, the United States and the United Kingdom, where the United States is directly related with Canada. On the contrary, Spain and its intranational and interinstitutional network are located in the yellow cluster, along with Belgium and the Netherlands. e red cluster is the most numerous, made up mostly of European Union countries, such as Denmark, France, Finland, Germany, Sweden, and Italy. e blue cluster is formed by south-east Asian countries and is led by the People's Republic of China, which presents a direct and intense relation with the United States. e countries which show scientific activity in the most recent network (Figure 9) are the People's Republic of China, Portugal, and India.
In the previous section about the analysis of the evolution of the volume of scientific production on the concept of ACAP linked with network formation and the process of firm internationalization, the article of Szulanski [79] titled "Exploring internal stickiness: impediments to the transfer of best practice within the firm" mainly stood out. As well as being one of the first articles which specifically links ACAP with network formation, it was published in the Strategic Management Journal, which also figures in the top ten group of the most published journals (Table 3). It is as well the most cited article since 1994, so it can be considered as an article which marks a trend in the topic analyzed (Table 4). e article titled "Clusters and knowledge: local buzz, global pipelines and the process of knowledge creation" [82] is the second most cited. It deals with the spatial clustering of economic activity and its relation to the spatiality of knowledge creation in interactive learning processes. It questions the view that tacit knowledge transfer is confined to local milieus, whereas codified knowledge may roam the globe almost frictionlessly. eir hypothesis therefore was that the more developed the pipelines between the cluster and distant sites of knowledge, the higher the quality (and value) of local buzz benefiting all firms in the local cluster. is is why a firm will learn more if its neighbouring firms in the cluster are globally well connected rather than being more inwardlooking and insular in their orientation. Finally, some policy implications, stemming from this argument, are identified, especially in the development of global pipelines, which requires institutional and infrastructure support. In the third most cited article, titled "Knowledge transfer in intraorganizational networks: effects of network position and absorptive capacity on business unit innovation and performance," Tsai [51] carried out an empirical analysis about 24 business units in a petrochemical company and 36 business units in a food-manufacturing company and concluded that organizational units can produce more innovations and enjoy better performance if they occupy central network positions that provide access to new knowledge developed by other units and depend on the units' absorptive capacity or ability to successfully replicate new knowledge. In Reagans and McEvily's [83] article titled "Network structure and knowledge transfer: the effects of cohesion and range," the research considered how different features of informal networks, such as network structure influences, affect knowledge transfer.
eir results indicated that both social cohesion and network range ease knowledge transfer, over and above the effect of the strength of the tie between two people. e last of the five most cited articles is the article of Yli-Renko et al. [84], titled "Social capital, knowledge acquisition, and knowledge exploitation in young technology-based firms," in which they examined the effects of social capital in key customer relationships on knowledge acquisition and knowledge exploitation. ey carried on an empirical study of 180 entrepreneurial high-technology ventures based in the United Kingdom. e most cited articles mentioned of this search are focused on the study of ACAP and different types of networks, but they do not deal with the link of ACAP with internationalization. Meyer-Krahmer and Reger's article [80], although it is the first one which relates ACAP with firm internationalization topics, is not the most cited paper. Lavie and Miller's article [85], titled "Alliance portfolio internationalization and firm performance," published in the journal Organizational Science, has 184 citations.
is study introduced the notion of alliance portfolio internationalization, which refers to the degree of foreignness of partners in a firm's collection of immediate alliance relationships. ey test the framework using data on the alliance portfolios of U.S.-based software firms (1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001), and their results provide support for the sigmoid relationship as well as for our predictions that firms, which have gained experience with foreign partners and maintained wholly owned subsidiaries in their partners' countries of origin and can overcome some of the liabilities of alliance portfolio internationalization and better leverage its benefits.

Content Analysis.
e number of articles which have ACAP as a topic axis has grown notably since 2008 (Figure 2) and have conserved, over the 25 years of the period analyzed, a close relation with three key concepts: innovation, networks, and their performance or innovative result. Firstly, the close relationship of innovation concept with ACAP construct stands out (Tables 9 and 10 and Figure 10).
is is highlighted even more if it is taken into account that the word "innovation" is not in the search formula. e second term emphasizes the topic of networks, a concept which, the same as innovation, is not a fashion in the scientific production as it remains invariably and significantly linked during all the periods. On the contrary, the studies concerning the result (performance) of ACAP in relation with the innovative result and its greater competitiveness have appeared more intensely in the last 10 years and stand out even more so if they are added to those which specifically address the performance of SMEs (Figure 10, period 2014-2018).
ough numerous other links exist which relate ACAP with diverse and varied topics, it is also important to note that there is a greater relevance of the association of the topic with SMEs than with large firms, as well as the association with other firm dynamic capacities, in particular those linked to the management of knowledge. e previously revealed verification (the greater association of the ACAP construct with SMEs) is logical and reasonable given the possibilities of innovating SMEs being much more related with the possibility of accessing external knowledge than that of generating new knowledge. is is particularly due to the scarcity of resources available for this purpose, at least if we compare them with large firms.
In a similar vein to what was stated before, SMEs should find advantages in their internationalization process from ACAP and the innovations derived from it. However, evidence does not exist, at least until now, that the studies in this sense are relevant. On the contrary, Table 4 shows that "internationalization" only appears in Numerous works published in recent years, and which we have already provided an account of, show the positive influence of networks in the internationalization process of SMEs as do the previous experience, which they have had in external markets and the favorable environment to internationalization in which they participate [4,9]. e temporal evolution was examined by considering the keywords timelines [74]. e frequencies were normalized by the total number of keywords in each time period. e resulting timelines are shown in Figure 10. e timelines of ACAP show an extraordinary growth, but its growth rate decreased in the last period (2014-2018). is type of visualization allows detecting keywords that can belong to different research fields. For example, the first subperiod (1994)(1995)(1996)(1997)(1998) shows that the keyword "firm" is the one that has increased. e keyword "performance" emerged in the second subperiod (1999)(2000)(2001)(2002)(2003) until the fourth subperiod.
us, in the third subperiod, the keyword "firm performance" emerged and its use has increased until the last subperiod.
As can be seen in the cluster analysis of Figure 11, eight heterogeneous keywork groups are shown in different colours. ere are two clusters that stand out. First, the red node labelled "absorptive-capacity" is related to other nodes such as "dynamic capacity" y "knowledge acquisition," which are related conceptually. Second, the green node labelled "innovation" is linked to "networks" keyword together with "collaboration," "research and development," and "technology". Finally, the word "internationalization" is related directly to the red cluster but in a peripheral place. Figure 12 shows the keywords tendency. As can be seen, the current keywords most used appear in yellow, which share the core concept of innovation, such as open innovation, innovation performance, product innovation, innovation capability, and technological innovation. In a similar way, other trendy keywords are firm performance and collaborative networks.
As can be seen in Figure 12, the sustainability topic has gained relevance and interest. Attention is paid to sustainable development and sustainability oriented innovation (SOI). In such a vein, the sustainability concept, by sustainable development of Brundtland report [86], is highlighted in the context of SMEs, as a challenge to achieve competitive advantage by innovation [16]. Moreover, Melane-Lavado andÁlvarez-Herranz [87] argue that the achievement of sustainability is mainly due to innovation and they presented a comparative study to try to shed light on the possible differences in the paths taken to obtain SOI by firms with and without Foreign Direct Investment, considering their different forms of knowledge management.
However, the evidence gathered in this bibliographic study shows that there still exists a practically unexplored field referring precisely to the relationships between networks and the internationalization of SMEs, and the positive moderator effect of ACAP in this process. Keywords, such as innovation capability, and SMEs give an idea of the current trends and close relationship with networking, but much less significant with internationalization, and end up defining that opportunity of exploration.

Conclusions
ACAP is a key factor in the greater competitiveness of firms, particularly of SMEs, as has been demonstrated in various studies referenced in this work as it enables innovating from the access to external knowledge. is greater competitiveness is indispensable when developing strategies of accessing international markets, and in this sense, interorganizational networks favor the process of internationalization, and ACAP acts as a factor which positively moderates this relation.
For this reason, the keywords "ACAP" and "Internationalization-Networking" were selected, meaning to understand how the study of these topics has been developed and how this study is currently and if there exists a gap which a research field can work on.
To that effect, it is verified that the scientific production reviewed in this study with the methodology described and which links the concepts of ACAP and internationalizationnetworking is found to be closely linked to their concepts of innovation and of performance. In the last five-year period, a growing trend has been perceived toward studies which include the topic of SMEs more intensely, but there is a still scant interest in internationalization. Hence, a field without enough exploration has emerged. As a consequence, scientific investigation of this may be used by researchers, governments, and entrepreneurs to improve their international competitiveness. e evidence of the survey carried out shows only 5 networks or clusters of countries and institutions where these studies are relevant, and of them, only 2 have a purely international profile, the North American-European-Asian network and the Canadian-Chinese one. Among the local networks (Spanish, Chinese, and British), the Spanish network stands out for its production and for being the only one with an interinstitutional profile. e other 2 lack this, concentrating themselves mainly on their own study and research centers. e keywords tendency analysis shows that the current keywords most used share the core concept of innovation, such as open innovation, innovation performance, product innovation, innovation capability, and technological innovation. In a similar way, other trendy keywords are firm performance and collaborative networks. Finally, the sustainability topic has gained relevance and interest. Moreover, attention is paid to sustainable development and sustainability oriented innovation (SOI). Although bibliometrics offers valuable information to support research evaluation, Waltman and Noyons [69] hold the position that bibliometric data sources such as Web of Science and Scopus offer only a limited coverage of the scientific and scholarly literature. International journals are typically well covered, but the coverage of national journals, journals in social sciences and humanities, conference proceedings, and books is much more limited. Finally, there is a bias of databases in favor of the literature in English, using a certain set of keywords for the search [88]. Another point to bear in mind about significant disciplinary differences is that different scientific fields have distinct publication, authorship, and citation practices.
Future research in the field of bibliometric studies could be the analysis of bibliometric indicators by 3D visualization.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that there are no conflicts of interest.