Topic Words Research Articles

MULTI-LEVEL CITY PORTRAIT RESEARCH BASED ON MULTI-SOURCE DATA

Abstract. City portrait is a social impression generated by the interaction between the public and the city, which can help us better understand and perceive the nature and characteristics of the city, and thus provide strong support for the development and governance of the city. However, most existing studies extract thematic semantic labels globally, but ignore the order of the tags and the degree of their contribution in the topic, which affects the city portrait extraction results. In addition, existing studies also lack the analysis of the impact of grid areas as the study scale on city portraits. In this paper, we propose a new approach to accurately identify city labels based on multi-source data grid fusion using a topic feature word extraction model (Weight-LdaVecNet) with fused topic word embedding and network structure analysis with feature word weight constraints. On this basis, we construct a multi-level city portrait description framework using hierarchical cluster analysis, extract tag clusters, and obtain a similarity matrix by combining topic feature tags and region feature tags using similarity analysis to construct a multi-level city region portrait, with a view to achieving a fine-grained construction of a multi-level city portrait. The experimental results show that, compared with the traditional LDA model, our method indicates that the identified city labels with similar thematic semantics have strong aggregation, thus proving the effectiveness of our proposed method. In addition, in the overall multi-level city portrait, we find that Beijing has a strong attractiveness in terms of cultural features. However, the regional distribution of cultural characteristics dimensions is not uniform in the multilevel city-region portrait, and better rational allocation and planning of cultural resources are needed to better meet people's needs.

Revisiting E-Learning in China: Analysis of TPACK Development in China and Other Countries

The development of TPACK for teachers is crucial for the effective use of technology in teaching and the promotion of teacher professional development. TPACK development has become a hot topic in the field of teacher education and educational technology. In order to gain a deeper understanding of the overall situation of current research on teachers’ TPACK development and predict the future trend of research, this study used "TPACK" as the theme word to search for relevant literature from the core journals of CNKI database (SCI/EI/Peking University Core/CSSCI/CSCD/AMI) and English journals of Scopus database. 306 and 1095 relevant literature were selected respectively. High-frequency keyword and co-word analysis methods were conducted on the obtained literature. The results showed that the current research hot spots in the TPACK field are current situation investigation and analysis, ability improvement, influencing factors. It is predicted that the future research on TPACK development will tend to focus on interdisciplinary and specialized research, applied practice research, and teacher TPACK ability improvement training research. By comparing the commonalities and differences of TPACK research in China and other countries. it is proposed to shift from universal research to specialized research, and carry out action and experimental research on teacher TPACK ability improvement training courses. And it is necessary to closely follow the digital process of education and carry out research on the framework development of I-TPACK and AI-TPACK to enhance teachers’ ability of teaching with technology. Keywords: teachers, TPACK , literature, teaching with technology, improvement

Trends in E-learning Implementation in Tertiary Education: A Review of Blended Learning in China and Other Countries

Blended learning is a combination of online learning and face-to-face learning. It is an important trend in the development of tertiary education in China and foreign countries. In recent years, the discussion of blended learning gradually moves from theory to practice, and more and more college teachers begin to pay attention to blended teaching design and put it into practice. This study attempts to give an overview on the topic of blended learning in tertiary education, with a comparison of Chinese and global perspectives. By using “blended learning” and “higher education” as the theme word, and through the advanced search function in CNKI (Core of Peking University /CSSCI) and Scopus database, this paper retrieved 356 and 2293 relevant journals respectively from 2004 to 2023. Utilizing a comparative approach, this study compared the retrieved data from CNKI and Scopus, sorted out the research context in China and foreign countries, analyzed the current research hot-spots, research topics and research trends of blended teaching in tertiary education using a comparative approach, and provided references for the future research in this field. The result showed that the research hot-spots includes the effectiveness of blended learning, blended learning models, blended learning strategies, and influencing factors of blended learning adoption. Furthermore, the study demonstrated a shift in research focus, from initial discussions to investigations into its pedagogical implications.

Current status and emerging trends of cardiac metabolism from the past 20 years: A bibliometric study

BackgroundAbnormal cardiac metabolism is a key factor in the development of cardiovascular diseases. Consequently, there has been considerable emphasis on researching and developing drugs that regulate metabolism. This study employed bibliometric methods to comprehensively and objectively analyze the relevant literature, offering insights into the knowledge dynamics in this field. MethodsThe data source for this study was the Web of Science Core Collection (WoSCC), from which the collected data were imported into bibliometric software for analysis. ResultsThe United States was the leading contributor, accounting for 38.33 % of publications. The University of Washington and Damian J. Tyler were the most active institution and author, respectively. The American Journal of Physiology-Heart and Circulatory Physiology, Journal of Molecular and Cellular Cardiology, Cardiovascular Research, Circulation Research, and American Journal of Physiology-Endocrinology and Metabolism were highly influential journals that published numerous high-quality articles on cardiac metabolism. Common keywords in this research area included heart failure, insulin resistance, skeletal muscle, mitochondria, as well as topic words such as cardiac metabolism, fatty acid oxidation, glucose metabolism, and myocardial metabolism. Co-citation analysis has shown that research on heart failure and in vitro modeling of cardiovascular disease has gained prominence in recent years and making it a research hotspot. ConclusionResearch on cardiac metabolism is steadily growing, with a specific focus on heart failure and the interplay between mitochondrial dysfunction, insulin resistance, and cardiac metabolism. An emerging trend in this field involves the enhancement of maturation in human induced pluripotent stem cell-derived cardiomyocyte (hiPSC-CM) through the manipulation of cardiac metabolism.

PHONOGRAM TEXTS AS AN OBJECT OF FORENSIC LINGUISTIC EXAMINATION

The present paper is dedicated to solving problems at the intersection of linguistics and the legal sphere. In particular, the study presents the experience of linguistic expertise of phonograms in solving the issue of distinguishing the specific semantics of the oral speech of a suspicious person in connection with homonymic phenomena between linguistic units manifested in the judicial process. The process of using the PRAAT computer program to determine the intonation contours of a subject's word is described and the level of reliability in solving the task is assessed. To make an objective conclusion about the semantics of the subject's speech, which has the same sound composition, i.e. the same form, but can express different meanings depending on the features of the prosodic complex, such as intonation, voice intensity, duration, all his speech acts become the subject of research. In this regard, in the linguistic examination of phonogram texts, the method of linguopragmatic analysis is applied, which is aimed at studying the words of the direct addressee (speaking subject), the addressee (listener) and the nature (style) of their interaction, the speech situation, strategies, and tactics of the speech of communicants (addressee and addressee). The paper emphasizes the importance of scientific conclusions in the theory of language in the field of forensic examination.

Public Perception of Islamic Higher Education Institutions on Twitter in Indonesia: A Social Media Sentiment Analysis

With the advent of social media, particularly Twitter, public opinion has become a dynamic and influential force. This study explores the public's perception of Islamic higher education institutions on Twitter, employing sentiment analysis to decipher the nuances of public sentiment. A comprehensive quantitative research methodology employing sentiment analysis was applied to a dataset of 11,809 tweets collected from 2018 to 2022, all focused on perceptions of Islamic higher education. The preprocessing phase involved cleaning, case folding, punctuation, number removal, stopword elimination, and tokenization. Sentiment analysis techniques were utilized to categorize tweets into positive, negative, or neutral sentiments. The analysis revealed a prevalent positive sentiment, constituting 73.16% of the dataset, 21.37% exhibited negative sentiment, and 5.47% reflected neutral sentiment. Visualizations, including word clouds, highlighted key topics shaping public perception. Words like "university," "religion," and "student" emerged prominently, influencing sentiment across categories. These results underscore a positive public perception of Islamic higher education, indicating the effectiveness of social media in shaping favorable opinions. The research emphasizes the significance of social media, specifically Twitter, as a platform for fostering positive perceptions of Islamic higher education. Strategic communication, proactive content dissemination, and responsive engagement are essential for enhancing institutional image. Future research directions, including advanced methodologies like topic modeling and word association techniques, promise a deeper understanding of public sentiment nuances, guiding educational institutions and policymakers toward effective social media engagement strategies

News of Noam: Unpacking Media Coverage of Chomsky

Media inform and obfuscate. Corporate motives influence how news is neglected, reported, and contextualized. It is informative to examine the extent to which news content varies based on source and period. As a longstanding critic of news media, and a leading public intellectual, Noam Chomsky is a worthy case study. Three author-created, corpora of news articles referencing Noam Chomsky were created. The corpora contained articles from Agence France-Presse (AFP; n = 54), The Associated Press (AP; n = 48), and the Cable News Network Wire (CNNW; n = 6), from the years 2012 to 2022. Number and length analyses, along with hypothesis tests, established the degree of similarity existing among the articles of the three news wire services and periods. Analysis of variance showed that news source was significant in terms of both the number of articles (H2), F(2,6) = 5.916, p = 0.038., and article length (H4), F(2,105) = 23.936, p = 0.000 ( = 0.05). Little commonality in content or framing was established among the three news sources or periods in terms of top words or bigrams of merit. Whereas there were differences in the absolute values and degree of change, each of the three news sources contained a slightly negative average sentiment score when using the AFINN lexicon. The results of this study, while limited to a single case, are illustrative of broader concerns and are potentially useful for those engaged in media studies, politics, rhetoric, organizational management, and the social sciences.

A Study on the Green Building Trend in China—From 2001 to 2022, Focusing on Research Topic Words

The Chinese government has committed to achieving carbon neutrality by 2060. Consequently, a shift towards green building development in the construction industry is inevitable. The purpose of this study is to summarize and predict the development direction of China’s green buildings by understanding their changing trends over time. Accordingly, this study employed content analysis as the research method to analyze 69 research papers and their topic words published from 2001 to 2022, with further in-depth analysis from four sub-dimensions: technical, economic, policy, and systemic. Overall, since 2000, China’s green building research has mainly focused on energy-saving technology and economic aspects, with particular attention being paid to the role of incentive policies. Research on the energy-saving renovation of existing buildings is insufficient, and research on building energy-saving management systems has been scarce. This may be related to China’s emphasis on technology and its neglect of management functions such as operations and maintenance. Considering the important role of management in the construction industry, it is crucial for the academic community to strengthen research on a comprehensive and systematic green building management system and for the Chinese government to increase its support for green building management and practices.

Integrating Text Classification into Topic Discovery Using Semantic Embedding Models

Topic discovery involves identifying the main ideas within large volumes of textual data. It indicates recurring topics in documents, providing an overview of the text. Current topic discovery models receive the text, with or without pre-processing, including stop word removal, text cleaning, and normalization (lowercase conversion). A topic discovery process that receives general domain text with or without processing generates general topics. General topics do not offer detailed overviews of the input text, and manual text categorization is tedious and time-consuming. Extracting topics from text with an automatic classification task is necessary to generate specific topics enriched with top words that maintain semantic relationships among them. Therefore, this paper presents an approach that integrates text classification for topic discovery from large amounts of English textual data, such as 20-Newsgroups and Reuters Corpora. We rely on integrating automatic text classification before the topic discovery process to obtain specific topics for each class with relevant semantic relationships between top words. Text classification performs a word analysis that makes up a document to decide what class or category to identify; then, the proposed integration provides latent and specific topics depicted by top words with high coherence from each obtained class. Text classification accomplishes this with a convolutional neural network (CNN), incorporating an embedding model based on semantic relationships. Topic discovery over categorized text is realized with latent Dirichlet analysis (LDA), probabilistic latent semantic analysis (PLSA), and latent semantic analysis (LSA) algorithms. An evaluation process for topic discovery over categorized text was performed based on the normalized topic coherence metric. The 20-Newsgroups corpus was classified, and twenty topics with the ten top words were identified for each class. The normalized topic coherence obtained was 0.1723 with LDA, 0.1622 with LSA, and 0.1716 with PLSA. The Reuters Corpus was also classified, and twenty and fifty topics were identified. A normalized topic coherence of 0.1441 was achieved when applying the LDA algorithm, obtaining 20 topics for each class; with LSA, the coherence was 0.1360, and with PLSA, it was 0.1436.

LDA 모형의 모형평가 및 잠재집단 해석 방법론에 대한 고찰

LDA model has been widely used to investigate the subject of documents or words in various fields because it can analyze large amounts of data. Although perplexity is used to compare various LDA models, it only presents the goodness of fit, and it is not possible to consider how well each document is clustered. To resolve this problem, coherence measures have been proposed. After model selection using complexity and coherence measures, LDAvis is widely used to understand the relationship and meaning of each topic. Although LDA model has been introduced in many preceding studies, the introduction of the model evaluation method, coherence measure, and LDAvis for model interpretation is not sufficient. In this paper, we first introduce the LDA model and the mini-batch learning method, and introduce the coherence measure. We also introduce LDAvis including similarity measures and dimension reduction method to investigate the relationship between topic. Relevance is also explained to present the top words for each topic. Finally, after fitting the LDA model at various hyper-parameters to the image annotation data, the models were compared through coherence measures, and each subject was interpreted using LDAvis.

Clustering Content Types and User Motivation Using DBSCAN on Twitter

We are currently in an era full of information and communication technology. One of the communication media used is Twitter. Twitter is a microblogging service that is used by its users to express their thoughts on a topic called a tweet. Tweets that are posted can be either positive tweets or negative tweets. One of the topics that is currently being discussed by Twitter users is Anies Baswedan as a 2024 Indonesian Presidential Candidate. Many people have tweeted this but it is not known how many users support or reject Anies Baswedan to run as a 2024 Indonesian presidential candidate. To assist the analysis, use the method clustering namely algorithm (Density-Based Spatial Clustering of Application with Noise). DBSCAN has the advantage of being able to detect data that is not included in a cluster and will be considered noise. This can improve the accuracy of the grouping because the data in the cluster will be cleaner. The TF-IDF Vectorizer is used to make it easier for programs to manage data because it can turn sentences into vectors that can be processed by the algorithm. To determine the evaluation of the program, the silhouette score method will be used. The results of calculating the silhouette score show a value of 0.29 with the formation of 3 clusters. Then an analysis is carried out based on the top words from each cluster and it can be identified that cluster 0 has a positive category supporting Anies Baswedan to run for the 2024 Presidential Candidate and cluster 1 has a negative category that does not support Anies Baswedan not advancing for the 2024 Presidential Candidate.

Comparing Natural Language Processing and Structured Medical Data to Develop a Computable Phenotype for Patients Hospitalized Due to COVID-19: Retrospective Analysis.

Throughout the COVID-19 pandemic, many hospitals conducted routine testing of hospitalized patients for SARS-CoV-2 infection upon admission. Some of these patients are admitted for reasons unrelated to COVID-19 and incidentally test positive for the virus. Because COVID-19-related hospitalizations have become a critical public health indicator, it is important to identify patients who are hospitalized because of COVID-19 as opposed to those who are admitted for other indications. We compared the performance of different computable phenotype definitions for COVID-19 hospitalizations that use different types of data from electronic health records (EHRs), including structured EHR data elements, clinical notes, or a combination of both data types. We conducted a retrospective data analysis, using clinician chart review-based validation at a large academic medical center. We reviewed and analyzed the charts of 586 hospitalized individuals who tested positive for SARS-CoV-2 in January 2022. We used LASSO (least absolute shrinkage and selection operator) regression and random forests to fit classification algorithms that incorporated structured EHR data elements, clinical notes, or a combination of structured data and clinical notes. We used natural language processing to incorporate data from clinical notes. The performance of each model was evaluated based on the area under the receiver operator characteristic curve (AUROC) and an associated decision rule based on sensitivity and positive predictive value. We also identified top words and clinical indicators of COVID-19-specific hospitalization and assessed the impact of different phenotyping strategies on estimated hospital outcome metrics. Based on a chart review, 38.2% (224/586) of patients were determined to have been hospitalized for reasons other than COVID-19, despite having tested positive for SARS-CoV-2. A computable phenotype that used clinical notes had significantly better discrimination than one that used structured EHR data elements (AUROC: 0.894 vs 0.841; P<.001) and performed similarly to a model that combined clinical notes with structured data elements (AUROC: 0.894 vs 0.893; P=.91). Assessments of hospital outcome metrics significantly differed based on whether the population included all hospitalized patients who tested positive for SARS-CoV-2 or those who were determined to have been hospitalized due to COVID-19. These findings highlight the importance of cause-specific phenotyping for COVID-19 hospitalizations. More generally, this work demonstrates the utility of natural language processing approaches for deriving information related to patient hospitalizations in cases where there may be multiple conditions that could serve as the primary indication for hospitalization.

Research on brand operation strategy of integrating traditional culture into the modern tourism industry based on big data analysis

Abstract The rich traditional culture is an inexhaustible treasure trove of resources for the brand operation of the modern tourism industry. This paper proposes an improved method for the LDA model of text big data mining, optimizing the topic representation of LDA by introducing the Topic2Vec model, optimizing the probability calculation of generated words by combining the original Dilley distribution and Bernoulli distribution and optimizing the assignment of topic words in the Gibbs adoption process of LDA model with the help of likelihood function. Then, the big data of tourism brand design text, promotion, and cultural products are mined and analyzed based on the improved LFV-LDA model. In the design of tourism brands focusing on traditional culture, the frequency of the keyword “pattern” was the highest, accounting for 8.20%. In brand promotion, the most distributed channel from 2017 to 2021 was a live broadcast, and the growth rate in 2020 and 2021 reached 22.71% and 36.29%. Regarding traditional cultural and creative products, tourism cultural and creative products have maintained a high growth momentum from 2012 to 2021, with an average annual growth rate of 154.25%. Based on the analysis of big data, it can guide the operation of tourism industry brands in all aspects, such as design, promotion, and landing, and give full play to the value connotation of traditional culture.

Development of a travel recommendation algorithm based on multi-modal and multi-vector data mining.

Given the rise of the tourism industry, there is an increasing urgency among tourists to access information about various tourist attractions. To address this challenge, innovative solutions have emerged, utilizing recommendation algorithms to offer customers personalized product recommendations. Nonetheless, existing recommendation algorithms predominantly rely on textual data, which is insufficient to harness the full potential of online tourism data. The most valuable tourism information is often found in the multi-modal data on social media, characterized by its voluminous and content-rich nature. Against this backdrop, our article posits a groundbreaking travel recommendation algorithm that leverages multi-modal data mining techniques. The proposed algorithm uses a travel recommendation platform, designed using multi-vector word sense segmentation and multi-modal data fusion, to improve the recommendation performance by introducing topic words. In our final experimental comparison, we verify the recommendation performance of the proposed algorithm on the real data set of TripAdvisor. Our proposed algorithm has the best degree of confusion with various topics. With a LOP of 20, the Precision and MAP values reach 0.0026 and 0.0089, respectively. It has the potential to better serve the tourism industry in terms of tourist destination recommendations. It can effectively mine the multi-modal data of the tourism industry to generate more excellent economic and social value.

Global development of artificial intelligence in cancer field: a bibliometric analysis range from 1983 to 2022.

Artificial intelligence (AI) is widely applied in cancer field nowadays. The aim of this study is to explore the hotspots and trends of AI in cancer research. The retrieval term includes four topic words ("tumor," "cancer," "carcinoma," and "artificial intelligence"), which were searched in the database of Web of Science from January 1983 to December 2022. Then, we documented and processed all data, including the country, continent, Journal Impact Factor, and so on using the bibliometric software. A total of 6,920 papers were collected and analyzed. We presented the annual publications and citations, most productive countries/regions, most influential scholars, the collaborations of journals and institutions, and research focus and hotspots in AI-based cancer research. This study systematically summarizes the current research overview of AI in cancer research so as to lay the foundation for future research.

Intelligence system for sentiment classification with deep topic embedding using N-gram based topic modeling

Multi-modal information outbreak is consistently increasing in social media. Classification of tweet sentiments using various information modalities will help the recommender systems to achieve success in digital marketing. Moreover, aspect-level sentiment analysis categorizes a target’s sentiment polarity in a specific environment. Using topic modelling in aspect-level sentiment analysis enables the identification of more accurate aspect-based tweet sentiments. The existing sentiment classification techniques used for the development of recommendation systems do not focus on the aspect-based approach modelled using deep learning classifier with temporal analysis on the social media data. Hence, this paper proposes an efficient sentiment classification model that highlights the impact of topic modelling-based word feature embedding for improvising the classification of Twitter sentiments and product reviews based on temporal reasoning and analysis for performing predictive analysis. For tweets context analysis, Latent Dirichlet Allocation based topic modelling is used in this work which generates the topics. For each topic, the sentiment is calculated separately and the topic guided feature expansion is done using Senti-wordnet. Moreover, an extended deep learning classification algorithm called Long Short-Term Memory (LSTM) with word embedding and temporal reasoning(LSTMWTR) is proposed in this paper for improving the classification accuracy. Finally, the labelled data are classified using the existing machine learning algorithms namely Naïve Bayes, Support Vector Machines and also using the deep learning models such as Convolution Neural Network(CNN),LSTM, Recurrent Neural Networks (RNN) and the transformer model namelyBi-directional Encoder Representation from Transformers (BERT),Convolution Bi-directional Recurrent Neural Network (CBRNN) and the proposed deep learning algorithm namelyLSTMWTR. These sentiment classification algorithms have been evaluated with word embedding for tweet sentiment classification and product review classification. The results obtained from this work show that the proposed LSTMWTR algorithm emerges as the highly accurate model for tweet sentiment and product review classification.

Narrative Education Combined With Experiential Teaching in the Development of Empathic Competence of Undergraduate Nursing Students: Pre-Test Post-Test Design

Empathy is a fundamental quality that nursing staff should possess, but the empathy ability of nursing students in China is currently at an intermediate level and requires improvement. This study aimed to explore teaching methods for nursing humanistic education, that can improve the empathy ability of nursing students and cultivate high-quality nursing talents. This study employed narrative education and experiential teaching methods to develop empathic competence in teaching pediatric nursing courses. The empathic ability of nursing students was measured using the JSPE-NS scale before and after teaching and was analyzed to refine the teaching reflection theme words. After teaching, nursing students scored higher on the total empathy scale, perspective-taking dimension, and emotional care dimension than before teaching, with statistically significant differences ( p &lt; 0.05). Five themes were extracted from the teaching reflection, including emotional empathy, pediatric nursing skills, gratitude to parents, reverence for life, and course experience. The results show that the combination of narrative education and experiential teaching methods applied to pediatric nursing courses effectively improved nursing students’ empathy, strengthened their identification with the nursing profession, and promoted the teaching reform of humanistic education in nursing courses. The combination of narrative education and experiential teaching methods can improve the empathic ability of nursing students. This approach is suitable for integrating humanistic education into nursing professional courses and provides a feasible idea for developing humanistic qualities in nurses.

Text-mining metadata: What can titles tell us of the history of modern and contemporary art?

The use of statistical text-mining to investigate the linguistic structure of textual resources has received limited attention in digital art history. In this paper the question I address is that of what text-mining titles as given in metadata can tell us about the history of modern and contemporary art. To investigate this question I constructed a dataset from the metadata for over 170,000 artworks, drawing on the online collections of 133 art museums in 30 countries. The use of topic modelling, parts-of-speech tagging and word counting allows me to identify large-scale and long-run patterns in the language used in the titles of those artworks. I set out an art historical reading of those patterns in which artistic interests signalled by the language used in titles come and go and are re-inflected, epistemic perspectives on the kinds of knowledge art can or should engender change, and artists engage with the ways that the title functions. My ‘distant’ reading is consistent with the canonical history of modern and contemporary art and cuts across the particularities of artist, period or movement which often feature in such accounts, providing a fresh perspective on that history. It also complements and extends the scholarship on the history of titles in the visual arts. The analytical framework and the dataset I have developed are not limited to answering the question addressed in this paper, and I consider some of the possibilities for future work.

A Survey on Medical Health Records and AI

As the population increases, so do the number of patients getting admitted in hospitals. This generates an overwhelming amount of data within the electronic health records (EHRs) that is impossible to manage manually. This is where machine learning concepts come in handy. The ML algorithms for regression-classification have become increasingly popular within the healthcare sector. Especially for cardiovascular diseases (CVD). Estimation states that by 2030, over 23 million people will die from CVD each year. But, it is estimated that 90% of CVD is preventable. The on time recognition and diagnosis of heart failure from the pre-existing medical records is a way to do so. However, the EHRs are not particularly reliable when it comes to the comparison of structured and unstructured data. This is due to the use of colloquial language and possible existence of sparse content. To tackle this issue, the proposed system uses the KTI-RNN model for recognition of unstructured data and removal of sparse content, TF-IWF model to extract the keyword set, the LDA model to extract the topic word set, and finally, the GA-BiRNN model to identify heart failure from extensive medical texts. The GA-BiRNN model is made up of a bidirectional recurrent neural network model and its output layer embedded with global attention mechanism and gating mechanism.

Incorporating structural topic modeling into short text analysis

Abstract The past few decades have seen the rapid development of topic modeling. So far, research has been more concerned with determining the ideal number of topics or meaningful topic clustering words than with applying topic modeling techniques to evaluate linguistic theories. This study proposes the Structural Topic Model (STM)-led framework to facilitate the interpretation of topic modeling results and standardize text analysis. STM encompasses various model training mechanisms, thereby requiring systematic designs to properly combine language studies. “Structural” in STM refers to the inclusion of metadata structure. Unlike the corpus-based keyness approach, STM can capture contextual cues and meta-information for the interpretation of topical results. Besides, STM can make cross-corpora comparisons via topical contrast, a challenging task for corpus-driven related models such as the Biterm Topic Model (BTM). Stylistic variations in song lyrics are taken as an illustration to show how to use the suggested framework to delve into the linguistic theory proposed by Pennebaker (2013). The topical model and iterable model in the proposed paradigm can clarify how pronouns affect style distinction. We believe the proposed STM-led framework can shed light on text analysis by conducting a reproducible cross-corpora comparison on short texts.