Citation: Zhou, A.; Zhang, Y.; Lu, M.
Multidimensional Domain
Knowledge Framework for Poet
Proling. Electronics 2023, 12, 656.
hps://doi.org/10.3390/
electronics12030656
Academic Editors: Sławomir
Nowaczyk, Rita P. Ribeiro and
Grzegorz Nalepa
Received: 24 November 2022
Revised: 25 January 2023
Accepted: 26 January 2023
Published: 28 January 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Swierland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Aribution (CC BY) license (hps://
creativecommons.org/licenses/by/
4.0/).
Article
Multidimensional Domain Knowledge Framework for
Poet Proling
Ai Zhou *, Yijia Zhang and Mingyu Lu
College of Information Science and Technology, Dalian Maritime University, Dalian 116026, China
* Correspondence: zhouai9070@dlmu.edu.cn
Abstract: Authorship proling is a subtask of authorship identication. This task can be regarded as
an analysis of personal writing styles, which has been widely investigated. However, no previous
studies have aempted to analyze the authorship of classical Chinese poetry. First, we provide an
approach to evaluate the popularity of poets, and we also establish a public corpus containing the
top 20 most popular poets in the Tang Dynasty for authorship proling. Then, a novel poetry author‑
ship proling framework named multidimensional domain knowledge poet proling (M‑DKPP) is
proposed, combining the knowledge of authorship aribution and the text’s stylistic features with
domain knowledge described by experts in traditional poetry studies. A case study for Li Bai is used
to prove the validity and applicability of our framework. Finally, the performance of M‑DKPP frame‑
work is evaluated with four poem datasets. On all datasets, the proposed framework outperforms
several baseline approaches for authorship aribution.
Keywords: authorship aribution; Chinese classical poetry; authorship proling; transformer
1. Introduction
With the application of computer science and technology in humanities, digital hu‑
manities has become a new interdisciplinary research trend. It abandons the approach of
carefully reading classical pieces in humanistic research, which has been conducted for
centuries, and it regards historical materials; documents; the literature; artistic works; and
other texts, images, and even audio and video materials as data. With the help of computer
and visualization measures, people can analyze and calculate large amounts of data using
dierent diagrams, which introduces a new perspective for humanistic research [1].
Authorship proling is a subtask of authorship identication. This unique task is both
related to personal writing styles and text classication, which means the individual’s writ‑
ing activities unconsciously reect their stylistic “ngerprint” and can credibly provide a
deduction of the language structure used in documents [2]. Authorship proling is an im‑
portant application eld of digital humanities. Research is mainly focused on controversial
authorship or literary style analysis [3]. Authorship proling has been widely studied in
many languages. However, in terms of Chinese, except for the Dream of Red Mansion [4],
no previous studies exist for Chinese authorship proling studies.
Researchers who pay aention to classical Chinese poetry, such as literature collation
and translation appreciation, have become an important basis for the modern research sys‑
tem of Tang poetry. The development of information science facilitates the digital preserva‑
tion, retrieval, and analysis of ancient documents. Research on corpus construction, poetry
generation, automatic word segmentation, automatic subject classication, and knowledge
mapping provides new approaches for the intelligent application and knowledge discov‑
ery of Tang poetry. However, the authorship proling of Tang poetry remains in the tradi‑
tional eld. Researchers only qualitatively describe the poetry style of a single poet, period,
or subject, and they fail to comprehensively describe the writing style of poets from multi‑
ple dimensions.
In this research study, we summarized our major contributions as follows.
Electronics 2023, 12, 656. https://doi.org/10.3390/electronics12030656 https://www.mdpi.com/journal/electronics
