Search results

ChatGPT is a versatile technology with practical use cases spanning many professional disciplines including marketing. Being a recent innovation, however, there is a lack of academic insight into its tangible applications in the marketing realm. To address this gap, the current study explores ChatGPT’s application in marketing by mining social media data. Additionally, the study employs the stages-of- growth model to assess the current state of ChatGPT’s adoption in marketing organizations.

The study collected tweets related to ChatGPT and marketing using a web-scraping technique (N = 23,757). A topic model was trained on the tweet corpus using latent Dirichlet allocation to delineate ChatGPT’s major areas of applications in marketing.

The topic model produced seven latent topics that encapsulated ChatGPT’s major areas of applications in marketing including content marketing, digital marketing, search engine optimization, customer strategy, B2B marketing and prompt engineering. Further analyses reveal the popularity of and interest in these topics among marketing practitioners.

The findings contribute to the literature by offering empirical evidence of ChatGPT’s applications in marketing. They demonstrate the core use cases of ChatGPT in marketing. Further, the study applies the stages-of-growth model to situate ChatGPT’s current state of adoption in marketing organizations and anticipate its future trajectory.

During crises such as accidents or disasters, an enormous volume of information is generated on the Web. Both people and decision-makers often need to identify relevant and timely content that can help in understanding what happens and take right decisions, as soon it appears online. However, relevant content can be disseminated in document streams. The available information can also contain redundant content published by different sources. Therefore, the need of automatic construction of summaries that aggregate important, non-redundant and non-outdated pieces of information is becoming critical.

The aim of this paper is to present a new temporal summarization approach based on a popular topic model in the information retrieval field, the Latent Dirichlet Allocation. The approach consists of filtering documents over streams, extracting relevant parts of information and then using topic modeling to reveal their underlying aspects to extract the most relevant and novel pieces of information to be added to the summary.

The performance evaluation of the proposed temporal summarization approach based on Latent Dirichlet Allocation, performed on the TREC Temporal Summarization 2014 framework, clearly demonstrates its effectiveness to provide short and precise summaries of events.

Unlike most of the state of the art approaches, the proposed method determines the importance of the pieces of information to be added to the summaries solely relying on their representation in the topic space provided by Latent Dirichlet Allocation, without the use of any external source of evidence.

It has always been a hot topic for online retailers to obtain consumers’ product evaluations from massive online reviews. In the process of online shopping, there is no face-to-face interaction between online retailers and customers. After collecting online reviews left by customers, online retailers are eager to acquire answers to some questions. For example, which product attributes will attract consumers? Or which step brings a better experience to consumers during the process of shopping? This paper aims to associate the latent Dirichlet allocation (LDA) model with the consumers’ attitude and provides a method to calculate the numerical measure of consumers’ product evaluation expressed in each word.

First, all possible pairs of reviews are organized as a document to build the corpus. After that, latent topics of the traditional LDA model noted as the standard LDA model, are separated into shared and differential topics. Then, the authors associate the model with consumers’ attitudes toward each review which is distinguished as positive review and non-positive review. The product evaluation reflected in consumers’ binary attitude is expanded to each word that appeared in the corpus. Finally, a variational optimization is introduced to calculate parameters mentioned in the expanded LDA model.

The experiment’s result illustrates that the LDA model in the research noted as an expanded LDA model, can successfully assign sufficient probability with words related to products attributes or consumers’ product evaluation. Compared with the standard LDA model, the expanded model intended to assign higher probability with words, which have a higher ranking within each topic. Besides, the expanded model also has higher precision on the prediction set, which shows that breaking down the topics into two categories fits better on the data set than the standard LDA model. The product evaluation of each word is calculated by the expanded model and depicted at the end of the experiment.

This research provides a new method to calculate consumers’ product evaluation from reviews in the level of words. Words may be used to describe product attributes or consumers’ experiences in reviews. Assigning words with numerical measures can analyze consumers’ products evaluation quantitatively. Besides, words are labeled themselves, they can also be ranked if a numerical measure is given. Online retailers can benefit from the result for label choosing, advertising or product recommendation.

Discovering the research topics and trends from a large quantity of library electronic references is essential for scientific research. Current research of this kind mainly depends on human justification. The purpose of this paper is to demonstrate how to identify research topics and evolution in trends from library electronic references efficiently and effectively by employing automatic text analysis algorithms.

The authors used the latent Dirichlet allocation (LDA), a probabilistic generative topic model to extract the latent topic from the large quantity of research abstracts. Then, the authors conducted a regression analysis on the document-topic distributions generated by LDA to identify hot and cold topics.

First, this paper discovers 32 significant research topics from the abstracts of 3,737 articles published in the six top accounting journals during the period of 1992-2014. Second, based on the document-topic distributions generated by LDA, the authors identified seven hot topics and six cold topics from the 32 topics.

The topics discovered by LDA are highly consistent with the topics identified by human experts, indicating the validity and effectiveness of the methodology. Therefore, this paper provides novel knowledge to the accounting literature and demonstrates a methodology and process for topic discovery with lower cost and higher efficiency than the current methods.

The present article's primary purpose is the topic modeling of the global coronavirus publications in the last 50 years.

The present study is applied research that has been conducted using text mining. The statistical population is the coronavirus publications that have been collected from the Web of Science Core Collection (1970–2020). The main keywords were extracted from the Medical Subject Heading browser to design the search strategy. Latent Dirichlet allocation and Python programming language were applied to analyze the data and implement the text mining algorithms of topic modeling.

The findings indicated that the SARS, science, protein, MERS, veterinary, cell, human, RNA, medicine and virology are the most important keywords in the global coronavirus publications. Also, eight important topics were identified in the global coronavirus publications by implementing the topic modeling algorithm. The highest number of publications were respectively on the following topics: “structure and proteomics,” “Cell signaling and immune response,” “clinical presentation and detection,” “Gene sequence and genomics,” “Diagnosis tests,” “vaccine and immune response and outbreak,” “Epidemiology and Transmission” and “gastrointestinal tissue.”

The originality of this article can be considered in three ways. First, text mining and Latent Dirichlet allocation were applied to analyzing coronavirus literature for the first time. Second, coronavirus is mentioned as a hot topic of research. Finally, in addition to the retrospective approaches to 50 years of data collection and analysis, the results can be exploited with prospective approaches to strategic planning and macro-policymaking.

The set of 2,509 documents related to the human-centric aspect of manufacturing were retrieved from Scopus database and systmatically analyzed. Using an unsupervised machine learning approach based on Latent Dirichlet Allocation we were able to identify latent topics related to human-centric aspect of Industry 5.0.

This study aims to create a scientific map of the human-centric aspect of manufacturing and thus provide a systematic framework for further research development of Industry 5.0.

In this study a 140 unique research topics were identified, 19 of which had sufficient research impact and research interest so that we could mark them as the most significant. In addition to the most significant topics, this study contains a detailed analysis of their development and points out their connections.

Industry 5.0 has three pillars – human-centric, sustainable, and resilient. The sustainable and resilient aspect of manufacturing has been the subject of many studies in the past. The human-centric aspect of such a systematic description and deep analysis of latent topics is currently just passing through.

The purpose of this paper is to study tourists’ spatial and psychological involvement reflected through tourism destination image (TDI), TDI is divided into on-site and after-trip groups and the two groups are compared in the frame of three-dimensional continuums.

By conducting latent Dirichlet allocation (LDA) modeling to tourism user-generated content, structural topic models are established. The topics separated out from unstructured raw texts are structural themes and representations of TDI. Social network analysis (SNA) reveals the quantitative and structural differences of three-dimensional continuums of the two TDI groups.

The findings reveal that from the stage of on-site to after-trip, tourist perception of TDI shifts from psychologically to functionally-oriented, from common to unique, and from holistic to more attribute focused. Also, it is suggested that from a postmodernism perspective, TDI is never unique, fixed or universal, but has different image perceptions and feedbacks for different tourists.

With the assistance of social sensing, a panoramic view of TDI could be established. Targeted and precision destination marketing and image promotion could be applied out to each individual tourist.

Combining with the perspectives of the tourist-destination space system and the tourism involvement theory, this research proposes a TDI transformation model and an explanation of the internal mechanism. The originality of research also lies in the methodological innovation of social sensing data and the LDA topic model.

本研究针对旅游目的地形象（TDI）及其体现出的游客空间和心理涉入, 将旅游目的地形象划分为在场形象和游后形象, 并将二者在TDI三维连续体（Three-dimensional continuums）框架下进行比较。

本研究应用内容分析法, 通过对旅游用户生成内容（tourism UGC）进行LDA（Latent Dirichlet Allocation）建模, 从非结构化的原始文本中建立起结构化的语义主题模型, 并且应用社会网络分析（Social Network Analysis）, 从定量和结构化的角度揭示了游中与游后目的地形象的差异。

研究发现, 从游中到游后, 游客的目的地形象感知经历了从心理到功能、从一般到特殊、从整体到属性的转变。同时, 基于后现代主义的视角, 旅游目的地形象并不是唯一的、固定的或放之四海而皆准的, 而是在不同的游客感知中有不同的形象和体现。

应用社会感知（Social Sensing）理论可以全面解析旅游目的地形象。同时可以针对特定游客采取精准定点的旅游目的地营销和形象推广手段。

本研究从旅游目的地空间系统和旅游涉入理论视角出发, 提出了旅游目的地形象转变的模型和其内在机制解释, 在方法上创新性地使用了社会感知数据和LDA主题模型。

关键词 旅游目的地形象, 在场形象, 游后形象, 旅游用户生成内容 (tourism UGC), LDA（Latent Dirichlet Allocation）建模, 社会感知

Para estudiar el grado de participación espacial y psicológica de los turistas reflejado en la imagen del destino turístico (TDI), el TDI se divide en grupo en el sitio y grupo posterior al viaje, y los dos grupos se comparan en el marco del continuo tridimensional.

Al modelar la posible asignación de Dirichlet (LDA) del contenido generado por el usuario turístico (UGC), se estableció un modelo de tema estructural. El tema que está separado del texto original no estructurado es el tema estructurado y la representación de TDI. El análisis de redes sociales reveló diferencias en el número y la estructura de los continuos tridimensionales de los dos grupos de TDI.

Los resultados de la encuesta muestran que, desde la escena hasta los viajes, la percepción de los turistas de TDI cambia de orientación psicológica a funcional, de lo ordinario a lo único, y de una atención general a más. Además, se sugiere que desde una perspectiva posmoderna, TDI nunca es único, fijo o universal, sino que tiene diferentes percepciones de imagen y comentarios para diferentes visitantes.

Con la ayuda de la detección social, se podría establecer una vista panorámica de TDI. El marketing de destino y la promoción de imágenes dirigidos y precisos podrían aplicarse a cada turista individual.

Combinando con las perspectivas del sistema espacial de destino turístico y la teoría de la participación turística, esta investigación propone un modelo de transformación TDI y la explicación del mecanismo interno. La originalidad de la investigación también radica en la innovación metodológica de los datos de detección social y el modelo de tema LDA.

This study furthers one’s understanding of the motivations of the crowdfunding crowd by empirically examining critical factors that influence the crowd's decision to support a crowdfunding project.

Backer's comments from a sample of the top 100 most funded technology product projects on KickStarter were collected. A latent Dirichlet allocation (LDA) analysis strategy was adopted to investigate critical motivational factors. Three experts mapped those factors to the known theoretical constructs of social exchange theory (SET).

Although backers are motivated by value, they are also motivated by far less tangible social factors including trust and a feeling of psychological ownership. Findings suggest that the crowd is far more than a passive group of investors or customers and should be viewed as participatory stakeholders. This study serves as guidance for project owners hoping to motivate the crowd and for future investigators examining backer motivations in other types of crowdsourcing projects.

Online chatter in the form of user-generated comments is an excellent data source for researchers to mine for value and meaning.

Given strong feelings of psychological ownership, project owners should actively engage the crowd and solicit the crowd for advice and help in order to motivate them.

The study presents the first empirical exploration of backer motivations using LDA guided by theory and the knowledge of experts. A framework of latent motivational factors is proposed.

Crowdfunding platforms are important innovations that allow nascent entrepreneurs to gain access to financial resources and crowd inputs to better refine and develop their business idea. The purpose of this paper is to investigate user-generated content (UGC) from both reward-based and equity-based crowdfunding platforms, in order to determine its implications for open and user innovation.

A total sample of 200 most funded technology products was extracted from four distinct crowdfunding platforms. A latent Dirichlet allocation (LDA) analysis was performed in an attempt to identify critical latent factors. The analysis was carried out through the theoretical lens of innovation literature, in an attempt to uncover the implications for open and user innovation.

The authors were able to highlight the implications of crowd inputs for open and user innovation, as backers provided nascent entrepreneurs with several types of feedback, ranging from product co-development to strategy and marketing. Furthermore, the study provided an overview of the key differences emerging between reward-based and equity-based crowdfunding platforms in terms of crowd inputs.

The present study features intrinsic limitations of the LDA approach being adopted. More specifically, it only provides a “snapshot” in time of the current sample, rather than investigating its development over time.

The present study solidifies the value of UGC as a resource to mine for trends and feedback.

The study contributes to both the innovation literature and the crowdfunding literature. It bridges several gaps found in both literature streams, by providing empirical evidence to test and verify pre-existing exploratory research.

It has been eight months into the global pandemic health crises COVID-19, yet the severity of the crises is just getting worse in many parts of the world. At this stage, it is essential to understand and observe the general attitude of the public toward COVID crises and the major concerns the public has voiced out and how it varies across months. Understanding the impact that the COVID-19 crises have created also helps policymakers and health-care organizations access the primary steps that need to be taken for the welfare of the community. The purpose of this study is to understand the general public's response towards COVID-19 crises and the major issues that concerns them.

For the analysis, data were collected from Twitter. Tweets regarding COVID-19 crises were collected from February 1, 2020, to June 27, 2020. In all, 433,195 tweets were used for this study. Natural language processing (NLP), which is a part of Machine learning, was used for this study. NLP was used to track the changes in the general public's sentiment toward COVID-19 crises and LDA was used to understand the issues that shape the general public's sentiments the crises time. Using Python library Wordcloud, the authors further derived how the primary concerns regarding COVID crises various from February to June of the year 2020.

This study was conducted in two parts. Study 1 results showed that the attitude of the general public toward COVID crises was reasonably neutral at the beginning of the crises (Month of February). As the crises become severe, the sentiments toward COVID increasingly become negative yet a considerable percentage of neutral sentiments existed even at the peak time of the crises. Study 2 finds out that issues including the severity of the disease, Precautionary measures need to be taken, and Personal issues like unemployment and traveling during the pandemic time were identified as the public's primary concerns.

The research adds value to the literature on understanding the major issues and concerns, the public voices out about the current ongoing pandemic. To the best of the authors’ knowledge, this is the first study with an extended period of timeframe (Five months). In this research, the authors have collected data till June for analysis that makes the results and findings more relevant to the current time.