A spatio-temporal emotional framework for knowledge extraction and mining in digital humanities
| DOI | https://doi.org/10.1108/AJIM-09-2021-0278 |
| Published date | 12 April 2022 |
| Date | 12 April 2022 |
| Pages | 1103-1125 |
| Subject Matter | Library & information science,Information behaviour & retrieval,Information & knowledge management,Information management & governance,Information management |
| Author | Jun Deng,Chuyi Zhong,Shaodan Sun,Ruan Wang |
A spatio-temporal emotional
framework for knowledge
extraction and mining in
digital humanities
Jun Deng and Chuyi Zhong
Jilin University, Changchun, China
Shaodan Sun
Jilin University, Changchun, China and
Nanjing University of Science and Technology, Jiangyin, China, and
Ruan Wang
Jilin University, Changchun, China
Abstract
Purpose –This paper aims to construct a spatio-temporal emotional framework (STEF) for digital humanities
from a quantitative perspective, applying knowledge extraction and mining technology to promote innovation
of humanities research paradigm and method.
Design/methodology/approach –The proposed STEF uses methods of information extraction, sentiment
analysis and geographic information system to achieve knowledge extraction and mining. STEF integrates
time, space and emotional elements to visualize the spatial and temporal evolution of emotions, which thus
enriches the analytical paradigm in digital humanities.
Findings –The case study shows that STEF can effectively extract knowledge fromunstructured texts in the field of
Chinese Qing Dynasty novels. First, STEF introduces the knowledge extraction tools –MARKUS and DocuSky –to
profile character entities and perform plots extraction. Second, STEF extracts the characters’emotional evolutionary
trajectory from the temporal and spatial perspective. Finally, the study draws a spatio-temporal emotional path figure
of the leading characters and integrates the corresponding plots to analyze the causes of emotion fluctuations.
Originality/value–The STEF is constructedbased on the “spatio-temporal narrative theory”and“emotional
narrative theory”. It is the first framework to integrate elements of time, space and emotion to analyze the
emotional evolution trajectories of characters in novels. The execuability and operability of the framework is
also verified with a case novel to suggest a new path for quantitative analysis of other novels.
Keywords Digital humanities, Knowledge extraction, Knowledge mining, Spatio-temporal emotional
analysis, Natural language processing
Paper type Research paper
1. Introduction
Since the concept of humanities computing was proposed, digital humanities (DH) has seen
substantial development in research tools, methods and perspectives. With digital
technology, the traditional paradigm of humanities research is changed, bringing
innovations and vitality to thoughts and perspectives in humanities research (Chen and
Chang, 2019). Consequently, many humanistic texts are digitized, forming a large number of
electronic data, which facilitates knowledge extraction and mining. In this context, the
knowledge demands of humanities scholars are also getting stronger; so, it is necessary to
STEF
application in
the novel The
Scholars
1103
The authors gratefully acknowledge the financial support of National Social Science Fund: “Research on
Knowledge Aggregation and Discovery of Historical Archives Resources from the Perspective of Digital
Humanities”(19BTQ102).
The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/2050-3806.htm
Received 25 September 2021
Revised 30 December 2021
12 March 2022
Accepted 19 March 2022
Aslib Journal of Information
Management
Vol. 74 No. 6, 2022
pp. 1103-1125
© Emerald Publishing Limited
2050-3806
DOI 10.1108/AJIM-09-2021-0278
explore knowledge in humanistic texts to promote the transformation of the traditional
humanistic research analysis paradigm (Hong et al., 2020).
As a peakera in the production anddissemination of Chinesenovels after the MingDynasty,
Qing Dynasty novelsmark the highest achievementof ancient Chinese vernacularnovels and
classicalChinese novels and have a far-reachingimpact on Chinese cultureand even the world
culture. Withthe trend of text digitization, a largenumber of digital versions of Qing Dynasty
novels haveemerged, providing basic conditions for DH research.Despite valuable effortsand
insightful findings (e.g. Wang and Qu,2021), previous studies of Qing Dynastynovels tend to
assume a speculative manner to draw impressionistic conclusions subjectively, and little
research has been conducted to quantify analysis from a macroperspective.
To fill the gap of insufficient quantitative research on novels and utilize the advantages of
knowledge organization and mining from the library and information science, we choose the
famous Qing Dynasty novel –“The Scholars”as the data source and propose the spatio-
temporal emotional framework (STEF). By using natural language processing technology to
extract texts and understand characters’emotional evolution path from time and space
perspective, this study helps to interpret the novel from the macro perspective, provides a
reference for other novels analysis and reflects the trend of cross-disciplinary research.
2. Literature review
2.1 Knowledge extraction and mining related to digital humanities
Knowledge extraction and mining is the most widely used technology in big data environments.
Big data, as Sch€
och (2013) so phrased, was introduced to the humanities research in 2010s, and has
brought revolutionary changes to humanitiesresearch in data magnitude and research paradigms.
With the rise of DH, researchers have worked to expand the research agendas in DH. For example,
Schich et al. (2014), by analyzing the spatial data of 150,000 historical celebrities, plotted a diagram
of luminaries’spatial migration to display the history of western culture. Another case of digital
technology-assisted humanities research is Archer and Jockers (2016), which used deep learning
technology to conduct a quantitative analysis of 20,000 novels over 30 years. Both studies show
that knowledge extraction and mining technology help humanities researchers conduct “distant
reading”(Moretti, 2013) and explore corpus in a manner of “bird’seyeview”,soastoidentifynew
problems, explore fresh perspectives and set new research agendas.
Knowledge mining technology is also effective for demonstrating the changes of literary
forms over time and, thus, useful for constructing a multidimensional literary history by building
large-volume corpus and language norms (Jin and Li, 2014). With the help of computational
techniques, it is possible to explore the overall picture of literature in a large data set, the internal
mechanism of literature history and the macroscopic law of literary style and literature
development. Roque (2012) suggested using computational modeling to enable the analysis of
narrative texts. Jockers (2013) introduced “macroanalysis”,demonst rating the possibility of using
big data techniques in literature research. Specifically, scholars haveused knowledge mining and
statistical models, such as automatic word segmentation (Huang et al., 2015b), automatic place
name recognition (Huang et al., 2015a), automatic classification (Wang et al.,2017), topic modeling
(Wang et al., 2018) and event extraction (Li et al., 2020), to examine some pre-Qin classics and
obtained fruitful results. More recently, new technologies have been explored to assist ancient
documents analysis. For example, Cui et al. (2020) used Long Short-Term Memory (LSTM),
Conditional Random Field (CRF) and Bidirectional Encoder Representations from Transformers
(BERT) models to extract the entities in the classical chrysanthemum poetry. Xu et al. (2020) took
the “local chronicles”as the basic corpus to extract the entities with the deep learning network.
Zhou et al. (2019) used knowledge extraction, knowledge integration and knowledge reasoning
technologies to construct the knowledge graph of Tang poetry. Xia et al. (2016) applied the same
methods to develop a genealogy-linked data service platform.
AJIM
74,6
1104
To continue reading
Request your trial