About this Research Topic
With the rise of open access and the development of big data, the data acquisition and processing of academic full text becomes possible and feasible. Compared with the bibliographic information, academic full text is not only large in quantity but also rich in semantics. Based on the massive and multidimensional data of academic full text, how to integrate natural language processing into the typical unstructured data knowledge mining of academic full text will play a decisive role in the research of bibliometrics. Against this background, this topic explores the role of natural language processing in bibliometrics. The specific research points are as follows.
(1) Academic full-text knowledge extraction based on a pre-trained model.
On the one hand, how to combine domain-specific data to build a domain-specific pre-trained model is one of the researches on this topic. On the other hand, this research point mainly explores how to apply the pre-trained language models of natural language processing to the automatic extraction of entities, entity relations, and entity attributes of the academic full text.
(2) Citation extraction and analysis for academic full text.
From the perspective of natural language processing, the citations in academic full text are regarded as the components of sentences. Based on traditional machine learning and deep learning, citations are automatically identified and analyzed in combination with bibliometrics knowledge and theory.
(3) Construction of academic full-text corpus from the perspective of bibliometrics
Combined with the research problem of bibliometrics, based on part-of-speech tagging, shallow automatic syntax, and deep automatic syntax, the construction of academic full-text corpus is explored accordingly. Based on the constructed corpus of academic literature, this topic will explore how to construct knowledge graphs and set up retrieval systems and automatic question-answering platforms.
(1) Academic full-text knowledge extraction based on a pre-trained model.
On the one hand, how to combine domain-specific data to build a domain-specific pre-trained model is one of the researches on this topic. On the other hand, this research point mainly explores how to apply the pre-trained language models of natural language processing to the automatic extraction of entities, entity relations, and entity attributes of the academic full text.
(2) Citation extraction and analysis for academic full text.
From the perspective of natural language processing, the citations in academic full text are regarded as the components of sentences. Based on traditional machine learning and deep learning, citations are automatically identified and analyzed in combination with bibliometrics knowledge and theory.
(3) Construction of academic full-text corpus from the perspective of bibliometrics
Combined with the research problem of bibliometrics, based on part-of-speech tagging, shallow automatic syntax, and deep automatic syntax, the construction of academic full-text corpus is explored accordingly. Based on the constructed corpus of academic literature, this topic will explore how to construct knowledge graphs and set up retrieval systems and automatic question-answering platforms.
Keywords: Knowledge extraction, academic full text, deep learning, Natural Language Processing, big data, open access, bibliometrics, Citation extraction, text mining
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
