Using structural topic modeling to gain insight into challenges faced by leaders

Sociologists frequently make use of language as data in their research using methodologies including open-ended surveys, in-depth interviews, and content analyses. Unfortunately, the ability of researchers to analyze the growing amount of these data declines as the costs and time associated with the research process increases. Topic modeling is a computer-assisted technique that can help social scientists to address these data challenges. Despite the central role of language in sociological research, to date, the field has largely overlooked the promise of automated text analysis in favor of more familiar and more traditional methods. This article provides an overview of a topic modeling framework especially suited for social scientific research. By way of a case study using abstracts from social movement studies literature, a short tutorial from data preparation through data analysis is given for the method of structural topic modeling. This example demonstrates how text analytics can be applied to research in sociology and encourages academics to consider such methods not merely as novel tools, but as useful supplements that can work beside and enhance existing methodologies.

Download Full-text

Content analysis of newspaper coverage of wolf recolonization in France using structural topic modeling

Biological Conservation ◽

10.1016/j.biocon.2018.01.029 ◽

2018 ◽

Vol 220 ◽

pp. 254-261 ◽

Cited By ~ 24

Author(s):

Marie Chandelier ◽

Agnès Steuckardt ◽

Raphaël Mathevet ◽

Sascha Diwersy ◽

Olivier Gimenez

Keyword(s):

Content Analysis ◽

Topic Modeling ◽

Newspaper Coverage ◽

Structural Topic Modeling

Download Full-text

A Structural Topic Modeling-Based Bibliometric Study of Sentiment Analysis Literature

Cognitive Computation ◽

10.1007/s12559-020-09745-1 ◽

2020 ◽

Vol 12 (6) ◽

pp. 1097-1129 ◽

Cited By ~ 1

Author(s):

Xieling Chen ◽

Haoran Xie

Keyword(s):

Sentiment Analysis ◽

Topic Modeling ◽

Bibliometric Study ◽

Structural Topic Modeling

Download Full-text

Taiwan as ghost island? Ambivalent articulation of marginalized identities in computer-mediated discourses

Discourse & Society ◽

10.1177/0957926519889124 ◽

2019 ◽

Vol 31 (3) ◽

pp. 285-306

Author(s):

Yao-Tai Li ◽

Yunya Song

Keyword(s):

Critical Discourse Analysis ◽

Topic Modeling ◽

Mainland China ◽

Critical Discourse ◽

Bulletin Board ◽

Social Critique ◽

Computer Mediated ◽

Discourse Historical Approach ◽

Self Presentations ◽

Structural Topic Modeling

This study examines the conflicting self-presentations when using the term ‘ghost island’ in Taiwan, a self-mocking way to belittle the homeland. While some view this term as a form of social critique, others consider it to be suggestive of a social malaise affecting contemporary Taiwanese. Drawing on online posts and comments from the most popular bulletin board system in Taiwan, this study combines topic modeling with a discourse-historical approach (DHA) to critical discourse analysis (CDA) to examine the constructions of ‘ghost island’ by Taiwanese netizens. A computer-aided content analysis was implemented using Structural Topic Modeling (STM) to identify discourse topics associated with netizens’ discourses on ghost island. Our findings suggest that the images of ‘us’ (the ordinary people) are presented as victims as against powerful ‘others’ (e.g. mainland China and local elites). Specifically, self-mockery was often invoked to project a loser image and marginalized status living on the island, whereas self-assertive narratives were invoked to affirm Taiwanese society’s democracy and freedom. The conflicting narratives – with a mixture of grudge, helplessness, pessimism, hope and pride – point to Taiwanese netizens’ ambivalent articulation of marginalized identities that operates to strengthen affective connectedness and virtual bonding.

Download Full-text

Research Trend Analysis on Smart City based on Structural Topic Modeling(STM)

Journal of Digital Contents Society ◽

10.9728/dcs.2019.20.9.1839 ◽

2019 ◽

Vol 20 (8) ◽

pp. 1839-1846 ◽

Cited By ~ 2

Author(s):

Hansaem Park ◽

Dong-Hyun Kim ◽

Seongju Chang

Keyword(s):

Trend Analysis ◽

Smart City ◽

Topic Modeling ◽

Research Trend ◽

Structural Topic Modeling

Download Full-text

Study of the Yahoo-yahoo Hash-tag Tweets Using Sentiment Analysis and Opinion Mining Algorithms

10.21203/rs.3.rs-354801/v1 ◽

2021 ◽

Author(s):

Adebayo Abayomi-Alli ◽

Olusola Abayomi-Alli ◽

Sanjay Misra ◽

Luis Fernandez-Sanz

Keyword(s):

Sentiment Analysis ◽

Topic Modeling ◽

Latent Dirichlet Allocation ◽

Opinion Mining ◽

Latent Semantic Indexing ◽

Workflow Engine ◽

Semantic Indexing ◽

Short Period ◽

Mining Algorithms ◽

Insight Into

Abstract BackgroundSocial media opinion has become a medium to quickly access large, valuable, and rich details of information on any subject matter within a short period. Twitter being a social microblog site, generate over 330 million tweets monthly across different countries. Analyzing trending topics on Twitter presents opportunities to extract meaningful insight into different opinions on various issues.AimThis study aims to gain insights into the trending yahoo-yahoo topic on Twitter using content analysis of selected historical tweets.MethodologyThe widgets and workflow engine in the Orange Data mining toolbox were employed for all the text mining tasks. 5500 tweets were collected from Twitter using the 'yahoo yahoo' hashtag. The corpus was pre-processed using a pre-trained tweet tokenizer, Valence Aware Dictionary for Sentiment Reasoning (VADER) was used for the sentiment and opinion mining, Latent Dirichlet Allocation (LDA) and Latent Semantic Indexing (LSI) was used for topic modeling. In contrast, Multidimensional scaling (MDS) was used to visualize the modeled topics. ResultsResults showed that "yahoo" appeared in the corpus 9555 times, 175 unique tweets were returned after duplicate removal. Contrary to expectation, Spain had the highest number of participants tweeting on the 'yahoo yahoo' topic within the period. The result of Vader sentiment analysis returned 35.85%, 24.53%, 15.09%, and 24.53%, negative, neutral, no-zone, and positive sentiment tweets, respectively. The word yahoo was highly representative of the LDA topics 1, 3, 4, 6, and LSI topic 1.ConclusionIt can be concluded that emojis are even more representative of the sentiments in tweets faster than the textual contents. Also, despite popular belief, a significant number of youths regard cybercrime as a detriment to society.

Download Full-text