How to identify metaknowledge trends and features in a certain research field? Evidences from innovation and entrepreneurial ecosystem

Vec2Dynamics: A Temporal Word Embedding Approach to Exploring the Dynamics of Scientific Keywords—Machine Learning as a Case Study

The study of the dynamics or the progress of science has been widely explored with descriptive and statistical analyses. Also this study has attracted several computational approaches that are labelled together as the Computational History of Science, especially with the rise of data science and the development of increasingly powerful computers. Among these approaches, some works have studied dynamism in scientific literature by employing text analysis techniques that rely on topic models to study the dynamics of research topics. Unlike topic models that do not delve deeper into the content of scientific publications, for the first time, this paper uses temporal word embeddings to automatically track the dynamics of scientific keywords over time. To this end, we propose Vec2Dynamics, a neural-based computational history approach that reports stability of k-nearest neighbors of scientific keywords over time; the stability indicates whether the keywords are taking new neighborhood due to evolution of scientific literature. To evaluate how Vec2Dynamics models such relationships in the domain of Machine Learning (ML), we constructed scientific corpora from the papers published in the Neural Information Processing Systems (NIPS; actually abbreviated NeurIPS) conference between 1987 and 2016. The descriptive analysis that we performed in this paper verify the efficacy of our proposed approach. In fact, we found a generally strong consistency between the obtained results and the Machine Learning timeline.

What do analyses of city size distributions have in common?

In this article, I conduct a textual and contextual meta-analysis of the empirical literature on Zipf's law for cities. Combining citation network analysis and bibliometrics, this meta-analysis explores the link between publication bias and reporting bias in the multidisciplinary field of quantitative urban studies. To complement a set of metadata already available, I collect the full-texts and reference lists of 66 scientific articles published in English and construct similarity networks of the terms they use as well as of the references and disciplines they cite. I use these networks as explanatory variables in a model of the similarity network of the distribution of Zipf estimates reported in the 66 articles. I find that the proximity in words frequently used by authors correlates positively with their tendency to report similar values and dispersion of Zipf estimates. The reference framework of articles also plays a role, as articles which cite similar references tend to report similar average values of Zipf estimates. As a complement to previous meta-analyses, the present approach sheds light on the scientific text and context mobilized to report on city size distributions. It allows to identified gaps in the corpus and potentially overlooked articles. It confirms the relationship between publication and reporting biases.

Innovation Ecosystem Research: Emerging Trends and Future Research

The innovation ecosystem term has increasingly been attracting the interest of scholars and practitioners for fifteen years. Contrary to the flourishing landscape, knowledge in this field is criticized as being fragmented. While past reviews revealed the conceptual and theoretical connections between innovation ecosystem and other related concepts, there is still a lack of comprehensive appreciation of the intellectual structure of state-of-the-art innovation ecosystem studies, hindering future research in this domain. To fill this void, this study utilized a systematic literature review approach combining bibliographic coupling and content analysis methods. Drawing on 136 studies reflecting the core and latest knowledge of innovation ecosystem literature, this study identifies five streams of the current innovation ecosystem research (i.e., technology innovation, platform innovation ecosystem, regional development, innovation ecosystem conceptualization and theorization, and entrepreneurship and innovation). Suggestions for future research are distilled via systematic analysis and discussion of these streams. Contributions of this study lie in decoding the intellectual structure of current innovation ecosystem research and offering targeted recommendations for future research.

WHAT IS GOING ON RECENTLY IN THE INNOVATION ECOSYSTEM FIELD? A BIBLIOMETRIC AND CONTENT-BASED ANALYSIS

Purpose: The popularisation of studies on ecosystem occurred, fundamentally, after the work of Adner (2006) and began to gain theoretical strength after the study by Gomes et al. (2018a), who identified a turning point between the concepts of business ecosystems and innovation ecosystems (IE) analysing studies from 1993 to 2016. The objective of this study is to present the characteristics, trends, and current research avenues of the concept post-transition, especially in the period from 2017 to 2019, during which more publications were produced than in the first 24 years.

Design/methodology/approach: Using data from the Web of Science, we performed a systematic literature review, through bibliometric ([Formula: see text] = 153 studies) and content analysis ([Formula: see text] = 15 studies). Data were analysed through bibliometrix (a R-tool package) and reading of the studies selected.

Findings: Our study revealed some new research suggestions linked to actors, infrastructure and financial capital for innovation; public policy, social capital, and development; IE Governance; and IE limits.

Research limitation/implications: One limitation of our study is the fact that it only used the Web of Science database.

Originality/value: Even though the literature on IE is extensive, no studies have captured the recent trends on this topic. These trends can be useful for ‘new avenues’ in IE research. The findings also can be important to clarify the concept of IE.

Topic analysis of academic disciplines based on prolific and authoritative researchers

Purpose

The paper aims to explore whether topic analysis (identification of the core contents, trends and topic distribution in the target field) can be performed using a more low-cost and easily applicable method that relies on a small dataset, and how we can obtain this small dataset based on the features of the publications.

Design/methodology/approach

The paper proposes a topic analysis method based on prolific and authoritative researchers (PARs). First, the authors identify PARs in a specific discipline by considering the number of publications and citations of authors. Based on the research publications of PARs (small dataset), the authors then construct a keyword co-occurrence network and perform a topic analysis. Finally, the authors compare the method with the traditional method.

Findings

The authors found that using a small dataset (only 6.47% of the complete dataset in our experiment) for topic analysis yields relatively high-quality and reliable results. The comparison analysis reveals that the proposed method is quite similar to the results of traditional large dataset analysis in terms of publication time distribution, research areas, core keywords and keyword network density.

Research limitations/implications

Expert opinions are needed in determining the parameters of PARs identification algorithm. The proposed method may neglect the publications of junior researchers and its biases should be discussed.

Practical implications

This paper gives a practical way on how to implement disciplinary analysis based on a small dataset, and how to identify this dataset by proposing a PARs-based topic analysis method. The proposed method presents a useful view of the data based on PARs that can produce results comparable to traditional method, and thus will improve the effectiveness and cost of interdisciplinary topic analysis.

Originality/value

This paper proposes a PARs-based topic analysis method and verifies that topic analysis can be performed using a small dataset.