Topic Latent Dirichlet Allocation Research Articles

COVID-19 Research Trends in Social Work: LDA Topic Modeling Analysis in South Korea

This study utilizes LDA topic modeling to examine research trends related to COVID-19 within the field of social work, analyzing these trends through peer-reviewed articles from the Korea Citation Index database. Latent Dirichlet Allocation (LDA) topic modeling, a statistical method for discovering abstract topics within a collection of documents, is applied to categorize and summarize the thematic concentration of the literature. Five themes have emerged: healthcare service and digitalized methods, exploring mental health status, qualitative approaches to social service program responses to COVID-19, evaluation of care services, and public service and program responses to COVID-19. Key findings reveal that the Korean social work academia focused on digital-based non-face-to-face services, evaluating the adequacy of public social services, and analyzing mental health and caregiving services during the pandemic. These findings indicate a reassessment of social work practices in response to the pandemic, underscoring the need to explore the challenges and opportunities presented by varied national responses. Additionally, the application of Python and machine learning in this research has shown significant benefits for social work studies, enabling a deeper analysis of complex, big data and facilitating more informed decision-making.

Literature data mining on the current research status of uveitis in China

Objective: To analyze the current research status of uveitis in China. Methods: It was a bibliometric analysis study. Using search formulas covering uveitis and its multiple subtypes, uveitis-related literature in English with publication dates from 2013 to 2022 was retrieved in Web of Science core databases through certain search strategies. This study used the latent Dirichlet allocation (LDA) algorithm to build topic models and analyzed the trends of research topics in recent years. Bibliometric analysis was used to analyze and visualize the bibliometric indicators (e.g., number of publications, citations, and H-index) of the included literature using tools such as VOSviewer software. Results: Over the past decade, China has published 1 657 papers on uveitis, ranking second globally. However, there is still room for improvement in terms of the H-index (58) and citation (12.28 per publication). Countries such as the USA (43.04%) and the United Kingdom (62.54%) were engaged in more international collaboration. We identified ten optimal LDA topics for uveitis literature in China such as immunotherapy, Behçet's disease, and Vogt-Koyanagi-Harada syndrome. Research on uveitis in China was mostly published in Ocular Immunology and Inflammation (92). Conclusions: China has made remarkable progress in uveitis research. Nonetheless, there is still untapped potential to enhance our global academic influence. It is encouraged to promote international collaborations, harness our expertise in areas like Behçet's disease and VKH syndrome, and publish our scientific achievements in high-impact journals.

Research on Mining Talent Demand for E-commerce Majors based on LDA Topic Model

[Purpose] Based on the text mining method, this study analyzes the job demand for domestic e-commerce industry positions in the Internet-oriented recruitment data, promotes the matching of e-commerce positions and talents, and promotes the good construction of the employment environment in the domestic e-commerce industry.[Method] Use the LDA(Latent dirichlet allocation) topic model for job competency requirements to calculate the similarity of the word segmentation results for professional talent requirements, determine the optimal number of topics, and output the visualization results of the LDA topic model.[Conclusion] In eastern China, there is a high demand for e-commerce jobs, and the majority of these positions only call for 1-3 years of experience. The basic requirement for the majority of jobs is a college degree. The Director of Operations and Director of Network Operations focus on e-commerce operations and management, the Network Promotion Specialist focuses on e-commerce advertising and product promotion, the Data Analyst and Market Analyst focus on data analysis and market research, and the Sales Representative focuses on sales. There is a correlation between the salary of the positions and education, experience and region; the higher the education, the more experience and the higher the level of regional development, the higher the salary level.

ChatGPT and computational-based research: benefits, drawbacks, and machine learning applications

Generative artificial intelligence (GenAI) systems are disrupting how research is conducted across a wide range of disciplines. Many journals have decided not to allow these tools to be co-authors for the purposes of publication, but rather they must be acknowledged by authors as having been utilised in the writing process. Furthermore, due to the hallucinations that these models sometimes produce, authors are to review what is generated and recognise that they hold it to be true and accurate. To date, there has been varying research conducted on the accuracy of GenAI systems and their production of written text. However, new functions that allow GenAI systems to produce coding for constructing tools in computer programming languages highlights a new area that warrants investigation. Therefore, this article puts forth an account of using ChatGPT 3.5 to construct coding to be utilised for a Latent Dirichlet Allocation Topic Model (LDA-TM) for use in a Systematic Literature Review. This is hoped to address three elements of using ChatGPT 3.5 for coding: code review, error resolution, and scripting new code. The code will be aimed at designating an appropriate Hyper-parameter for the Random State for use in the LDA-TM. Within this context, this article will discuss the advantages and drawbacks of utilising this new tool and what it means for researchers who wish to augment their work with computer programming-based applications. To the authors knowledge, this is the first time this has been discussed within the context of the research being conducted.

Infodemics surveillance system to detect and analyze health misinformation using big data and AI

Abstract Background Health misinformation disseminated on social media detrimentally affects the public's reactions toward public health measures. This has led to costly and harmful public health crises. Due to a lack of a comprehensive expert system that collects and analyzes large volumes of social media data, public health officials are incapable of mitigating health misinformation trends online. This study aimed to remedy this by designing and developing a big data pipeline and ecosystem (Health Misinformation Analysis System (HMAS)) capable of identifying and analyzing health misinformation online. Methods The HMAS system relies on Python, Elastic Stack, and the Twitter V2 API as the main technologies forming the backbone of the system. It comprises of five main components: (1) Data Extraction Framework (DEF); (2) Latent Dirichlet Allocation Topic Model; (3) Sentiment Analyzer; (4) Misinformation Classifier; (5) Elastic Cloud Deployment. Components 1-4 are hosted on a virtual machine with low computing and memory requirements. Through the DEF, HMAS extracts data from the Twitter V2 API. The HMAS expert system uses pre-trained models to perform automatic health misinformation analysis. The analyzed data are loaded and visualized in Elastic Cloud through dashboards and analytics. Results The HMAS system performance is accurate and efficient. It has been successfully utilized by independent investigators to extract significant insights, including a fluoride-related health misinformation use case, which analyzed data from 2015 to 2021. Moreover, HMAS has been applied to a vaccine hesitancy use case (2007-2022) and a heat-related illnesses use case (2011-2022). Conclusions The novel HMAS expert system can potentially aid public health officials globally to detect and analyze concerning trends in health misinformation online. This system can also grow to integrate social media data from several platforms for multiplatform analysis and support non-Western language content. Key messages • The novel HMAS pipeline can collect, detect, and analyze the rapidly increasing amounts of misinformation proliferating on social media related to a particular topic or set of related topics. • Identifying and monitoring health misinformation trends prevalent on social media through analytics dashboards and anomaly/risk detection alerts can lead to prompt governmental intervention.

Using LDA Topic Modeling to Understand Regrowth Factors of the Chinese Gaming Industry in the COVID-19 Era: Current Situation, Future and Predicament

The gaming industry, which was among the industries least affected by the COVID-19 outbreak, exhibited positive growth trends during the COVID-19 period. This paper explores the impact of COVID-19 on the gaming industry by analyzing news texts from 2020 to 2022 using text mining and LDA (latent Dirichlet allocation) topic classification, and visualizing charts. The study focuses on three themes, namely the current situation, the future, and possible problems of China’s game industry in the post-epidemic era. The findings of this study suggest that the development of the game industry during the COVID-19 outbreak prompted the government to regulate policies and promote the transformation of game companies, which had a positive impact on the development of China’s game industry. However, this study also found that due to the effects of COVID-19, society and the government have increased their focus on the time management of underage game users, which poses a significant challenge to the games industry. This paper recommends improvements from three perspectives, namely society, policy, and enterprise, with the aim of contributing to the long-term development of China’s game industry.

Latent Dirichlet Allocation modeling of environmental microbiomes.

Interactions between stressed organisms and their microbiome environments may provide new routes for understanding and controlling biological systems. However, microbiomes are a form of high-dimensional data, with thousands of taxa present in any given sample, which makes untangling the interaction between an organism and its microbial environment a challenge. Here we apply Latent Dirichlet Allocation (LDA), a technique for language modeling, which decomposes the microbial communities into a set of topics (non-mutually-exclusive sub-communities) that compactly represent the distribution of full communities. LDA provides a lens into the microbiome at broad and fine-grained taxonomic levels, which we show on two datasets. In the first dataset, from the literature, we show how LDA topics succinctly recapitulate many results from a previous study on diseased coral species. We then apply LDA to a new dataset of maize soil microbiomes under drought, and find a large number of significant associations between the microbiome topics and plant traits as well as associations between the microbiome and the experimental factors, e.g. watering level. This yields new information on the plant-microbial interactions in maize and shows that LDA technique is useful for studying the coupling between microbiomes and stressed organisms.

Why do we love the high line? A case study of understanding long-term user experiences of urban greenways

The High Line park (HLP) in New York City is one of the most successful contemporary greenway parks, inspiring urban planners, designers, artists, and administrators worldwide. This study provides a comprehensive understanding of user experiences in a long-term time frame (2011–2018) through the lens of online reviews. Using a mixed-methods approach, we conducted Latent Dirichlet Allocation (LDA) topic modeling to quantitatively identify the key topics that represent 34,060 reviews and 30,285 users, followed by qualitative analysis to inductively interpret the LDA topics. The results identified experiential, programmatic and physical elements of the HLP that are meaningful to users. Three primary components were found that make HLP successful according to users: spectacular visual and activity-related experiences, the historical transformation and cultural exploration, and the added value from park services ranging from amenities to on-site living performance. The article helps inform future decision-making and planning & design practices for future greenway projects.

Predicting injury-severity for cyclist crashes using natural language processing and neural network modelling

The use of machine learning techniques in safety research has increased as has the interest in using new data sources. This study’s unique contribution is the application of text mining—focusing on perceived cyclist safety and crash occurrence in an urban environment. We analysed crash data collected by the emergency forces in the Capital Region of Denmark from 2013 to 2017 and self-reported textual data provided by cyclists from 2018 to 2019. The analysis included natural language processing and topic modelling to identify Latent Dirichlet Allocation (LDA) topics from self-reports, representing environment characteristics that cyclists’ perceive as unsafe. A multi-output neural network regression model is applied to predict the injury-severity distribution of cyclists involved in crashes (measured by emergency response level [ERL]) based on the obtained topic distributions together with additional variables like cycle flow. We identified six LDA topics which address buses and cycle paths, conflicts with parked cars, roundabouts and inadequate maintenance, fast-moving cars and lack of cycle path, school zones and heavy traffic, and intersections and interactions with vehicles. Cycle flow was found to be the highest impacter on ERL prediction. However, other factors also impacted ERLs, especially school zones and heavy traffic. The results bring new insights into safety perception and actual safety for cyclists. The results contribute to a novel procedure for the joint correlation analysis using machine learning techniques on self-reported textual data thereby providing a better tool for infrastructure planning. The findings show the importance of including perceived safety in crash modelling and that authorities should focus on safety around schools and in intersections in order to improve safety for cyclists in a urban environment.

Evolution of CCUS Technologies Using LDA Topic Model and Derwent Patent Data

Carbon capture, utilization, and storage (CCUS) technology is considered an effective way to reduce greenhouse gases, such as carbon dioxide (CO2), which is significant for achieving carbon neutrality. Based on Derwent patent data, this paper explored the technology topics in CCUS patents by using the latent Dirichlet allocation (LDA) topic model to analyze technology’s hot topics and content evolution. Furthermore, the logistic model was used to fit the patent volume of the key CCUS technologies and predict the maturity and development trends of the key CCUS technologies to provide a reference for the future development of CCUS technology. We found that CCUS technology patents are gradually transforming to the application level, with increases in emerging fields, such as computer science. The main R&D institutes in the United States, Europe, Japan, Korea, and other countries are enterprises, while in China they are universities and research institutes. Hydride production, biological carbon sequestration, dynamic monitoring, geological utilization, geological storage, and CO2 mineralization are the six key technologies of CCUS. In addition, technologies such as hydride production, biological carbon sequestration, and dynamic monitoring have good development prospects, such as CCUS being coupled with hydrogen production to regenerate synthetic methane and CCUS being coupled with biomass to build a dynamic monitoring and safety system.

Using topic-noise models to generate domain-specific topics across data sources.

Domain-specific document collections, such as data sets about the COVID-19 pandemic, politics, and sports, have become more common as platforms grow and develop better ways to connect people whose interests align. These data sets come from many different sources, ranging from traditional sources like open-ended surveys and newspaper articles to one of the dozens of online social media platforms. Most topic models are equipped to generate topics from one or more of these data sources, but models rarely work well across all types of documents. The main problem that many models face is the varying noise levels inherent in different types of documents. We propose topic-noise models, a new type of topic model that jointly models topic and noise distributions to produce a more accurate, flexible representation of documents regardless of their origin and varying qualities. Our topic-noise model, Topic Noise Discriminator (TND) approximates topic and noise distributions side-by-side with the help of word embedding spaces. While topic-noise models are important for the types of short, noisy documents that often originate on social media platforms, TND can also be used with more traditional data sources like newspapers. TND itself generates a noise distribution that when ensembled with other generative topic models can produce more coherent and diverse topic sets. We show the effectiveness of this approach using Latent Dirichlet Allocation (LDA), and demonstrate the ability of TND to improve the quality of LDA topics in noisy document collections. Finally, researchers are beginning to generate topics using multiple sources and finding that they need a way to identify a core set based on text from different sources. We propose using cross-source topic blending (CSTB), an approach that maps topics sets to an s-partite graph and identifies core topics that blend topics from across s sources by identifying subgraphs with certain linkage properties. We demonstrate the effectiveness of topic-noise models and CSTB empirically on large real-world data sets from multiple domains and data sources.

Analysis of Depression in News Articles Before and After the COVID-19 Pandemic Based on Unsupervised Learning and Latent Dirichlet Allocation Topic Modeling

Analysis of Depression in News Articles Before and After the COVID-19 Pandemic Based on Unsupervised Learning and Latent Dirichlet Allocation Topic Modeling

Why Enroll in a Graduate-level Vocational Education Program?: Exploring Views of Graduate Students Using Latent Dirichlet Allocation Topic Modelling

Why Enroll in a Graduate-level Vocational Education Program?: Exploring Views of Graduate Students Using Latent Dirichlet Allocation Topic Modelling

A Framing Analysis and Regional Comparison of Newspaper Media Reports of COVID-19 in Shanghai: Based on the LDA Topic Model

In this study, we analyzed reports published by local and non-local newspapers on the 2022 COVID-19 outbreak in Shanghai using the Latent Dirichlet Allocation (LDA) topic modeling technique. Framing Theory suggests that both coverage types would typically produce similar media frames, notwithstanding the presence of slight differences in bias. We identified the media frames used by local and non-local newspapers on the Shanghai outbreak and our analysis subsequently informed our discussion on the similarities and differences that were uncovered. The paper offers suggestions for media reporting on how to better cover the outbreak.

Deciphering Latent Health Information in Social Media Using a Mixed-Methods Design.

Natural language processing techniques have increased the volume and variety of text data that can be analyzed. The aim of this study was to identify the positive and negative topical sentiments among diet, diabetes, exercise, and obesity tweets. Using a sequential explanatory mixed-method design for our analytical framework, we analyzed a data corpus of 1.7 million diet, diabetes, exercise, and obesity (DDEO)-related tweets collected over 12 months. Sentiment analysis and topic modeling were used to analyze the data. The results show that overall, 29% of the tweets were positive, and 17% were negative. Using sentiment analysis and latent Dirichlet allocation (LDA) topic modeling, we analyzed 800 positive and negative DDEO topics. From the 800 LDA topics-after the qualitative and computational removal of incoherent topics-473 topics were characterized as coherent. Obesity was the only query health topic with a higher percentage of negative tweets. The use of social media by public health practitioners should focus not only on the dissemination of health information based on the topics discovered but also consider what they can do for the health consumer as a result of the interaction in digital spaces such as social media. Future studies will benefit from using multiclass sentiment analysis methods associated with other novel topic modeling approaches.

Text Mining and Determinants of Sentiments towards the COVID-19 Vaccine Booster of Twitter Users in Malaysia

Vaccination is the primary preventive measure against the COVID-19 infection, and an additional vaccine dosage is crucial to increase the immunity level of the community. However, public bias, as reflected on social media, may have a significant impact on the vaccination program. We aim to investigate the attitudes to the COVID-19 vaccination booster in Malaysia by using sentiment analysis. We retrieved 788 tweets containing COVID-19 vaccine booster keywords and identified the common topics discussed in tweets that related to the booster by using latent Dirichlet allocation (LDA) and performed sentiment analysis to understand the determinants for the sentiments to receiving the vaccination booster in Malaysia. We identified three important LDA topics: (1) type of vaccination booster; (2) effects of vaccination booster; (3) vaccination program operation. The type of vaccination further transformed into attributes of “az”, “pfizer”, “sinovac”, and “mix” for determinants’ assessments. Effect and type of vaccine booster associated stronger than program operation topic for the sentiments, and “pfizer” and “mix” were the strongest determinants of the tweet’s sentiments after the Boruta feature selection and validated from the performance of regression analysis. This study provided a comprehensive workflow to retrieve and identify important healthcare topic from social media.

Analysis of the Thematic Structure and Discursive Framing in Articles about Trianon and the Holocaust in the Online Hungarian Press Using LDA Topic Modelling

AbstractIn our research, we examined the discursive framing of Trianon and the Holocaust in today’s Hungarian online media. Our corpus contained 26,519 articles connected to these two historical events published between September 2017 and September 2020. We used a mixed-method approach by combining computational text analysis, namely, LDA (Latent Dirichlet Allocation) topic models and qualitative methods. Our aim was not only to map the latent thematic structure and the discourses about Trianon and the Holocaust but also to identify the main differences in the rhetoric of the different sides of the political spectrum. Using Bull and Hansen’s categories, we found that the far-right and government media use the antagonistic, while the left-liberal media use the cosmopolitan mode of remembering. We identified many discursive differences between the rhetoric of the different political sides. Yet, the different relationship to emotions has the most far-reaching consequences. On the one hand, in antagonistic remembering, thus in the far-right and progovernment media, emotions play an essential role. On the other hand, cosmopolitan remembering, thus the rhetoric featured in the nongovernment media, detaches itself from emotions. This difference contributes significantly to the success of antagonistic remembering in Hungary. At the same time, the left-liberal side loses the opportunity to shape memory and identity politics.

Research on the Disturbance Sources of Vegetable Price Fluctuation Based on Grounded Theory and LDA Topic Model

Vegetables are an important element in people’s dietary structure, and the price fluctuation of vegetables has attracted more and more attention. The disturbance sources of vegetable price fluctuations are characterized by uncertain risks, environmental complexity, nonlinearity, self-organization and mutation. Analyzing the disturbance sources that affect vegetable price fluctuation is helpful to the establishment of early warning and regulation mechanisms of vegetable price risk. To address the problem that existing studies have not comprehensively and objectively clarified the disturbance sources of vegetable price fluctuations, this paper proposes a method of combining the LDA (Latent Dirichlet Allocation) topic model with grounded theory, constructs a system of vegetable price volatility disturbance source indicators and relationship matrix by improved conceptual lattice-weighted cluster method, obtains 23 disturbance sources indicators affecting vegetable price fluctuations in four aspects of supply, demand, natural environment and economic policy environment, and identifies six key factors through calculation and analysis. Through the research of this paper, a system of disturbance source indicators affecting vegetable price fluctuations is constructed, the internal connection of many disturbance sources of vegetable price fluctuations in a complex and uncertain environment is clarified, and key influencing factors are selected, thus facilitating the establishment of vegetable price risk warning models and regulation mechanisms.

Reviews Analysis of Korean Clinics Using LDA Topic Modeling

Objectives: In the health care industry, the influence of online reviews is growing. As medical services are provided mainly by providers, those services have been managed by hospitals and clinics. However, direct promotions of medical services by providers are legally forbidden. Due to this reason, consumers, like patients and clients, search a lot of reviews on the Internet to get any information about hospitals, treatments, prices, etc. It can be determined that online reviews indicate the quality of hospitals, and that analysis should be done for sustainable hospital marketing.Method: Using a Python-based crawler, we collected reviews, written by real patients, who had experienced Korean medicine, about more than 14,000 reviews. To extract the most representative words, reviews were divided by positive and negative; after that reviews were pre-processed to get only nouns and adjectives to get TF(Term Frequency), DF(Document Frequency), and TF-IDF(Term Frequency – Inverse Document Frequency). Finally, to get some topics about reviews, aggregations of extracted words were analyzed by using LDA(Latent Dirichlet Allocation) methods. To avoid overlap, the number of topics is set by Davis visualization.Results and Conclusions: 6 and 3 topics extracted in each positive/negative review, analyzed by LDA Topic Model. The main factors, consisting of topics were 1) Response to patients and customers. 2) Customized treatment (consultation) and management. 3) Hospital/Clinic’s environments.

Topic2Labels: A framework to annotate and classify the social media data through LDA topics and deep learning models for crisis response

The abundant use of social media impacts every aspect of life, including crisis management. Disaster management needs real-time data to be used in machine learning and deep learning models to aid their decision making. Mostly the data that is newly generated from social media is unstructured and unlabeled. Current text classification models based on supervised deep learning models heavily rely on human-labeled data that very small size and imbalanced in the context of disasters, ultimately affecting the generalization of models. In this study, we propose Topic2labels (T2L) framework which provides an automated way of labeling the data through LDA (latent dirichlet allocation) topic modeling approach and utilize Bert (the bidirectional encoder representation from transformer) embeddings for construction of feature vector to be employed to classify the data contextually. Our framework consists of three layers. In the first layer, we adopt LDA to generate the topics from the data, and develop a new algorithm to rank the topics, and map the highest ranked dominant topic into label to annotate the data. In the second layer, we transform the labeled text into feature representation through Bert embeddings and in the third layer we leveraged deep learning models as classifiers to classify the textual data into multiple categories. Experimental results on crisis-related datasets show that our framework performs better in terms of classification performance and yields improvement as compared to other baseline approaches.