This concept prompted the academics to re-examine
translation activities and understand that translation is
essentially a cross-cultural exchange and collision, in
which cultural factors profoundly influence the
choice of translation strategies and the final
presentation of the translated text. Since its inception,
many scholars have conducted in-depth
investigations around the "cultural turn", promoting
the flourishing development of cultural translation
theory and increasing its influence in the field of
translation research.
On the basis of Bassnett's research, many scholars
have further enriched and developed the theory of
cultural translation. Among them, Nida, E. A., in his
book Language, culture, and translating, analyses the
differences between the English and Chinese
nationalities in terms of cultural psychology, thinking
concepts and customary characteristics from the
perspectives of linguistics, literature and culture as
well as translation, Nida believes that translators
should start from the three dimensions of language,
culture and translation, and improve their cultural
literacy by observing the language translations in the
context(Nida,1993). Venuti, L. further advanced
cultural translation theory by outlining two
strategies—"naturalisation" and "alienation"—in The
Translator's Invisibility: A History of
Translation(Venuti,1994).
In the study of intercultural communication, Homi
takes cultural communication as the point of view,
regards the handling of differences in the original text
as the key to the translation process, and for the first
time puts forward the theory of the "third space" in
translation, in which he believes that the differences
between cultures play a role(Bhabha,1994). The
product of this space is the cultural hybrid, which has
the nature of two cultures. Scholar HERMANS T.
explored the specific issue of complicity of
translation in intercultural understanding and
proposed the concept of "thick translation" by
considering the practical ways in which intercultural
translation research may be carried
out(Hermans,2003). Through reading the literature
on cultural translation theory, the researcher found
that there is relatively limited research on cultural
translation theory and its derivatives, such as the
"third space", in the field of social media and
interculturalism. Therefore, this study hopes to
combine cultural translation theory with social media
and cross-cultural fields, so as to provide a reference
for the application of cultural translation theory in the
field of social media.
3 RESEARCH DESIGN
3.1 Research Method
This study integrates two research methods, corpus
text analysis and sentiment analysis, aiming to
conduct an in-depth analysis of user feedback content
under the "TikTok refugee" Memes on the
Xiaohongshu platform.
Corpus linguistics is a research approach that
performs quantitative analysis based on large-scale
textual data, capable of revealing key linguistic
phenomena within specific discourses. This method
identifies and describes overarching textual features
through word frequencies, keywords, word clusters,
and phrases, providing empirical support and deeper
insights for discourse analysis(Baker et al.2008). In
this study, to uncover the key linguistic patterns in
user feedback content, we focus on processing user
comments related to the "TikTok refugee" Memes.
Specifically, for English comments, we employ the
UU online tool to conduct word frequency analysis,
followed by generating a word frequency dendrogram
using Tableau software, which visually represents the
frequency of English words and their
interrelationships. For Chinese comments, the LZL
word cloud tool is used to perform word frequency
analysis and visualization, making vocabulary
distributions immediately apparent. These tools were
selected for their user-friendliness and robust
capacity to efficiently process large-scale text data
while accurately extracting critical information.
Sentiment analysis, or opinion mining, is a
computational discipline examining opinions,
attitudes, and emotions toward entities such as
individuals, events, or topics(Medhat et al.,2014).
This study leverages crawled comment data to
explore users' sentiment orientations in depth. We use
Python's VADER library to analyze Chinese
comment sentiment and the TextBlob library for
English comment sentiment. By integrating
TextBlob's lexicon matching mechanism with
VADER's contextual semantic rules and applying a
weighted algorithm, we classify cleaned text into
three categories: positive, neutral, and negative.
Finally, Tableau is used to create a sentiment polarity
pie chart, clearly and intuitively illustrating the
proportion of user attitudes. This approach was
adopted because different processing libraries offer
distinct advantages in sentiment analysis. By
combining these strengths and applying a weighted
algorithm, we achieve more accurate sentiment
classification, providing a robust foundation for