Preface

We heard about bibliometrics 10 years ago for the first time. In 2008 Corrado was writing a monograph on fast growing firms, a niche theme, which he approached for the first time. Scientific literature was fairly limited. Scholars came from different disciplines with a variety of approaches and methods that made it difficult to cumulate the findings.

We talked about this research problem once during a football match among scholars. Our discussion continued for several days on the various techniques of systematic analysis of literature. We enjoyed the exchange and concluded that bibliometrics was an interesting method and that it would have been fun to explore it together.

Our goal became to examine the intellectual structure of fast growing firms research. We analyzed all the scientific production published in academic English-written journals. The analysis was complex because it required several steps and diverse analysis and mapping software tools, which were often available only under commercial licenses. All the process was unwieldly, from data-collection to data-visualization. Massimo greatly contributed with his statistical and coding skills. Our collaboration continued in moments of fun, such as our frequent football matches. While analyzing data, we discovered that we enjoyed working together. In short, our friendship soon turned into a scientific collaboration that still lasts.

Within our departments and academic communities, the reaction to our work was positive. At that time, few people talked about bibliometrics in Italy, even from the point of view of research evaluation. Years later we presented a bibliometric analysis paper on performance management at the Annual Conference of the Academy of Management, the largest international management meeting. Also on that occasion, we got positive feedbacks that pushed us to persist. In the same years, young Italian colleagues asked us for suggestions for their literature reviews and for their research. Massimo opened some statistical analysis laboratories in R and together we presented the bibliometric analysis at some workshops.

We are telling this story because without these feedbacks and stimuli we would not have published the bibliometrix release 0.1 in 2016. A year later we are at version 1.7, thanks to our growing passion for bibliometrics and to the suggestions that today come from scholars from all around the world.

R-bibliometrix is currently a free tool for quantitative research in scientometrics and bibliometrics that includes all the main bibliometric methods of analysis, easy to use even for those who have no coding skills.

Bibliometrix is a unique tool, developed in the statistical computing and graphic R language, according to a logical bibliometric workflow. R is highly extensible because it is an object-oriented and functional programming language, and therefore is pretty easy to automate analyses and create new functions. As it has an open-software nature, it is also easy to get help from the users community, mainly composed by prominent statisticians. Therefore, bibliometrix is flexible and can be rapidly upgraded and can be integrated with other statistical R-packages. That why, it is useful in a constantly changing science such as bibliometrics.

Today bibliometrix is more than just a statistical tool. It is becoming a community of international developers and users who exchange questions, impressions, opinions, and examples within an open source project. For this reason, we are very honored that Dr Jie Liof the Research centerfor Safety and security SCITECH trends at the Department of Safety Science and Engineering, Shanghai Maritime University gave us the opportunity to tell you this story and to write an English preface for his book “Using R for Scientometrics data Visualization” that mainly introduces the BIBLIOMETRIX package to scholars and students.

We said that Bibliometrix includes all the main bibliometric methods of analysis, but we use it especially for science mapping and not for measuring science, scientists, or scientific productivity. Synthesizing past research findings is one of the most important tasks in advancing a line of research. Various methods exist to summarize the amount of scientific activity in a domain, but bibliometrics has the potential to introduce a systematic, transparent and reproducible review process. This is very relevant in an age when the number of academic publications is rising at a very fast pace and it is increasingly unfeasible to keep track of everything that is being published; and when the emphasis on empirical contributions is resulting in voluminous and fragmented research streams, and a contested feld. Literature reviews are increasingly playing a crucial role in synthesizing past research findings to effectively use the existing knowledge base, advance a line of research, and give evidence-based insights into the practice of exercising and sustaining professional judgment and expertise. The overwhelming volume of new information, conceptual developments and data are the milieu in which bibliometrics becomes useful, by providing a structured analysis to a large body of information, to infer trends over time, themes researched, identify shifts in the boundaries of the disciplines, to detect most the prolifc scholars and institutions, and to show the “big picture” of extant research.

Naples, Italy July 2017

Massimo Aria and Corrado Cuccurullo

前言

當前，我們正處于科學文獻大數(shù)據(jù)時代。面對海量的文獻，我們如何快速地了解一個研究領域、研究方向或者主題的整體格局以及未來的趨勢？在此背景下，與該問題直接相關的科學計量理論、方法和技術的適時發(fā)展，成為解決上述科研問題的一種有效的途徑。掌握與科學計量相關的技術和方法也成為科研工作者在新時代進行科學研究活動的基本技能要求。在過去十余年里，科學計量數(shù)據(jù)可視化的理論與方法已經大量地滲透到其他學科的研究實踐中。在國內，這種以科學文本數(shù)據(jù)為研究對象，通過可視化技術來揭示學科結構、演進和互動的研究領域被統(tǒng)稱為“科學知識圖譜”。

科學計量數(shù)據(jù)可視化背后涉及大量的科學計量學（還包含文獻計量學、網絡計量學以及信息計量學）方面的基礎理論，比如論文的作者生產率分布、論文的共被引、耦合、主題共現(xiàn)以及作者合作等。還包含了統(tǒng)計學和網絡科學等方面的技術和方法，比如多維尺度分析、聚類分析、復雜網絡分析、自然語言處理和文本挖掘等分析方法。上述理論和方法構成了進行科學計量數(shù)據(jù)可視化分析的知識基礎，是進行知識圖譜分析的前提。在理論和方法的支持下，當前國內外的相關學者已經開發(fā)了數(shù)十種科技文本挖掘方面的軟件或者工具包，這些知名的工具包含了HistCite、BibExcel、CiteSpace、SCI2以及VOSviewer等。這些工具為有意借助領域文獻分析以獲取學科研究格局和動態(tài)的學者提供了可能。

筆者在過去5年從事科學計量和知識圖譜的實踐研究中，相繼撰寫了關于CiteSpace、VOSviewer以及BibExcel等方面的書籍，主要目的在于幫助非科學計量學領域的學者快速應用該方法輔助科學研究。從2016年開始，已經相繼組織了4次與科學計量和知識圖譜相關的活動，與來自國內的數(shù)百名知識圖譜愛好者有過交流。在交流中，最為常見和令我反思的一個問題是：“我得到的圖譜結果應該怎樣解釋呢？”我認為，科學計量及知識圖譜的方法僅僅給我們提供了一種認識知識世界的新方式，但這種認識方式更需要知識圖譜實踐者結合自身的專業(yè)背景和知識圖譜的理論與方法去思考。在進行科學計量和知識圖譜分析的時候，讀者一定要明確自己要解決的問題是什么,以及為什么知識圖譜能夠解決提出的問題，它與其他方法相比優(yōu)勢在哪里,等等。即在進行科學計量和知識圖譜分析之前，一定要確定自己所要研究的問題，然后來選擇使用何種知識圖譜呈現(xiàn)方式解決問題。

本書是《CiteSpace：科技文本挖掘及可視化》《科學計量與知識網絡分析——基于BibExcel等軟件的實踐》《科學知識圖譜原理及應用——VOSviewer與CiteNetExplorer初學者指南》的姊妹篇。與前面這些應用程序不同的是，該書詳細介紹了意大利那不勒斯菲里德里克第二大學（University of Naples Federico II）經濟與統(tǒng)計系Massimo Aria和Corrado Cuccurullo基于R語言開發(fā)的BIBLIOMETRIX工具包。建議讀者在應用時通過提供的鏈接來檢查是否為最新版的BIBLIOMETRIX，在實際的研究中盡可能使用最新版來對數(shù)據(jù)進行分析（BIBLIOMETRIX-R Package for Bibliometric and Co-Citation Analysis，http://www.bibliometrix.org/）。該R工具包基本上涵蓋了進行科學計量和知識可視化的功能（圖0. 1），可以滿足愛好R軟件，并試圖使用R進行科學計量和知識圖譜分析的讀者。在此基礎上，對于科學計量與知識圖譜相關的一些R工具包，如rAltmetric、wordcloud2、gender以及tidytext等工具包進行了介紹。本書對使用R進行英文全文本挖掘的介紹很少，對中文全文本挖掘尚未涉及。在今后的更新中將對使用R進行全文本挖掘進行適當?shù)耐晟啤?/p>

圖0. 1bibliometrix功能概覽

為了便于讀者熟悉bibliometrix工具包，本書大多數(shù)的案例運行采用了工具包自帶的數(shù)據(jù)，一些案例專門下載了Web of Science和Scopus數(shù)據(jù)集并進行了分析。案例中呈現(xiàn)了所分析的結果，但并未就結果進行描述性或者帶有特定研究目的的解讀。讀者通過對這些結果的學習，自己去思考可以做些什么，或者至少可以通過這種方法了解自己所關注領域的基本情況。

本書在撰寫中有如下約定：

>后為代碼

#為代碼的說明

##為代碼運行的結果

感謝Massimo Aria和Corrado Cuccurullo，他們在本書寫作過程中給予了大力幫助，并為本書撰寫了英文序言。感謝首都經濟貿易大學出版社楊玲社長對科學計量與知識圖譜系列叢書出版的極力支持，感謝中國科學院李彬彬博士在提取子矩陣問題上的幫助，感謝滑鐵盧大學博士后于淼對文稿提出的修改建議，感謝本書的責任編輯薛曉紅以及研究生李平對本書的編輯和詳細校對。

回首自己在科學計量和知識圖譜研究與實踐上的經歷，感受五味雜陳。衷心地期望本書及相關系列叢書能進一步促進科學計量與知識圖譜實踐研究在國內的發(fā)展和普及，并使每一位讀者受益。

李杰

2018年5月于北京