La investigación bibliométrica en historia de la educaciónsituación actual, desarrollo de bases de datos específicas y propuestas desde la inteligencia artificial

Zusammenfassung

This thesis is based on the question of whether it is possible to quantitatively analyse the complexity of research in History of Education. In order to answer this question, the research work has been based, first of all, on scientometrics, its origins, evolution and problems. Subsequently, we have identified the particularities of research in the field, such as the subjects and periods studied, which are not included in the bibliographical information contained in the records of general databases. This has led to another difficulty: how can this specific data be stored alongside traditional bibliographic information? The answer has been the development of a specific database with the Design-Based Research methodology, called Hecumen, which makes it possible to combine both traditional bibliographic information and qualitative evaluations on these specific data referring to the subjects and periods studied. Its development has ended with another problem related to the complexity of manually cataloguing a large volume of articles. To solve this problem, the research approached Artificial Intelligence applied to the resolution of classification problems. The technology has been conceptualised, as well as the different algorithms, carrying out various experiments in order to identify the most suitable for the classification of texts. In addition, the methodology to be followed has been proposed based on the identification of the n most repeated words in each article, which has been the necessary information to nurture the learning of various connectionist artificial intelligences. With all this methodological and tool base, to conclude, a bibliometric study of 11 specialised journals of History of Education of international relevance has been carried out, covering the time window from 1961 to the present day. To do so, the tools previously developed were used. The sample consisted of 5217 articles, through which an analysis was made of the evolution of production, the study of global citations, authors, their collaboration and the representation by sex among the most productive and productive. The most frequent countries of affiliation, institutions and languages have also been analysed. An in-depth analysis of the topics has been carried out through the study of keywords, both in terms of their occurrence in pairs and the evolution of their interest over the last decade. The aim of this thesis has been to increase the understanding of the field of History of Education from a global perspective, as well as to provide the research community with a set of tools to analyse the past and present of research in order to be able to address future research challenges.