Building the Bridge: Topic Modeling for Comparative Research

Abstract

In communication research, topic modeling is primarily used for discovering systematic patterns in monolingual text corpora. To advance the usage, we provide an overview of recently presented strategies to extract topics from multilingual text collections for the purpose of comparative research. Moreover, we discuss, demonstrate, and facilitate the usability of the "Polylingual Topic Model" (PLTM) for such analyses. The appeal of this model is that it derives lists of related clustered words in different languages with little reliance on translation or multilingual dictionaries and without the need for manual post-hoc matching of topics. PLTM bridges the gap between languages by making use of document connections in training documents. As these training documents are the crucial resource for the model, we compare model evaluation metrics for different strategies to build training documents. By discussing the advantages and limitations of the different strategies in respect to different scenarios, our study contributes to the methodological discussion on automated content analysis of multilingual text corpora.

Measuring the diffusion of conspiracy theories in digital information ecologies

“…Languages. The advances in handling multi-language corpora (Chan et al, 2020;Lind et al, 2019Lind et al, , 2021Reber, 2019) enable cross-language analysis, at least for dictionary and topic modeling approaches, notwithstanding the complexity of multilingual dictionary construction and language transformations in topic models. To date, narrative detection based on semantic features is, to our knowledge, of limited applicability in cross-language analyses as its compatibility with current algorithmic and machine translation techniques is untested.…”

“…The different interpretive granularities of topics further hinder direct comparability across cases, platforms and media. That is, the topics in the sub-corpora may represent conceptually distinct constructs (for a broader discussion, see Lind et al, 2021). On the other hand, combining corpora from different platforms poses challenges concerning different text styles, formats, etc.…”

“…Conspiracy-related research on transnational and global diffusion can consider approaches allowing for cross-lingual classifications of conspirational content. The problem of cross-lingual text classification is intensively addressed in current research (Chan et al, 2020;Lind et al, 2019Lind et al, , 2021Reber, 2019). Scholars provide guidance for constructing multilingual term lists for a keyword-driven approach to collecting conspirational content across countries and for multilingual dictionary construction (Maier et al, 2021).…”

Discontentment trumps Euphoria: Interacting with European Politicians’ migration-related messages on social media

“…Wu et al, 2016) and is widely employed for multilingual automated content analysis (e.g. Courtney et al, 2020;Lind et al, 2021). To gain additional certainty for the translation results obtained for our study, we conducted a separate investigation (see Section A2 in the Appendix).…”

Development of Risk-Situation Scenario for Autonomous Vehicles on Expressway Using Topic Modeling

“…In general, topic modeling is applied across various fields, as it allows for handling a vast amount of data and of intuitively and elaborately demonstrating the relationships between topics (Kim et [22][23][24][25][26][27][28][29][30].…”