Text Mining Applications Research Articles

Exploring the patterns of China's carbon neutrality policies

To achieve the ambitious carbon neutrality goal by 2060, the Chinese government has implemented a series of carbon neutrality policies. These policy documents are pivotal in facilitating the examination of policy substance, the scrutiny of policy evolution, and the dissection of the policy instruments involved. This study develops an analytical framework for assessing carbon neutrality through policy documents, applying text mining and network analysis to probe the intricacies of policy topics, interagency collaboration, and diffusion dynamics. This research aims to delineate and expound upon the strategic paradigms employed by the Chinese government in its quest for carbon neutrality. The findings reveal a constellation of eleven policy topics, with "green" and "low carbon" being key aspects of each. The policy collaboration network has a density of 0.593, and the National Development and Reform Commission (NDRC)'s high average weighted degree of 14.6 highlights its crucial role in leading and coordinating these policies. In terms of diffusion dynamics, the green energy transition topic has a diffusion speed of 0.967 and a strength of 49, indicating its importance to the Chinese government. On a practical level, the findings offer policy-makers concrete, actionable recommendations to refine policy design and enhance implementation effectiveness. Theoretically, this study advances the scientific understanding of policy dynamics by proposing a novel analytical framework that integrates multiple dimensions of policy analysis, contributing to the methodological development of policy research.

텍스트 마이닝을 활용한 국내 기본심리욕구 연구 동향 분석

Objectives The purpose of this study is to examine research trends related to basic psychological needs in Korea and identify major research topics and trends. Methods To this end, we analyzed 327 academic papers related to basic psychological needs, which were more than candidates for KCI registration published on the Academic Research Information Service. The data was preprocessed through an R program, and the results of keyword frequency analysis were visualized in a word cloud, and analyzed by topic modeling. Results First, according to the literature status, 327 journals have been published in 106 societies, including the Learner-Centered Curriculum Education Association and the Korean Youth Association. Second, in keyword frequency analysis, terms such as basic psychology, desire, research, relationship, influence, mediation, analysis, autonomy, effectiveness, satisfaction, and competence appeared as top words. Third, topic modeling results were categorized by effects on individual psychological well-being, research methods, relationships with elementary school students and psychological well-being, the influence of young people's home and school environments, the mental health of college students, and relationships with stressors in modern society. Conclusions This study is significant in that it identified the characteristics of basic psychological needs research trends in Korea by applying text mining techniques to a vast amount of information on basic psychological needs, and it can be used as useful basic data for setting the direction of basic psychological needs research and selecting research topics in the future.

Navigating the blame game: Investigating automated vehicle fault in collisions under mixed traffic conditions

Vehicle being at fault in a crash has extensively been associated with its driver's behaviors and other human errors for human-driven vehicles (HDV). The introduction of automated vehicles (AVs) is expected to eliminate such human errors due to the ability of AVs to communicate with the external environment. However, various reports have documented AVs being at fault in collisions. This study applied text mining and mixed-effects logistic regression (MELR) on crash data involving AVs collected between 2017 and 2022 in California to explore the likelihood of an AV being at fault during a collision. It was found that among 497 crashes, a relatively small percentage (14.29 %) involved AVs being at fault. The text network results revealed patterns of keywords associated with the AVs being at fault. Such patterns include conventional mode of operation, area of impact, and resulting injuries. Furthermore, with about a 93 % prediction accuracy and an 83 % sensitivity score, the MELR results revealed that the likelihood of AVs being at fault increases when they are operated in conventional mode or when disengagement is involved. Moreover, turning, merging, or changing lane movements, unclear weather conditions, and operating on roadways with four or more lanes significantly increased the odds of an AV being at fault during a crash. Conversely, AVs were less likely to be at fault in commercial land use than residential land use, at intersection locations, and when the crash involved a truck. The practical implications of the findings are presented to improve AV operations.

Enhancing Keyword Extraction from Academic Articles Using Highlights

ABSTRACTKeywords facilitate rapid comprehension of academic papers for scholars, enhancing research efficiency. As some papers lack author‐assigned keywords, automated keyword extraction becomes crucial. Addressing the limited utilization of external paper information beyond titles and abstracts in prior studies, this research proposed leveraging the highlights section, which summarizes contributions and novelties, for unsupervised keyword extraction. We demonstrated performance improvements across three unsupervised keyword extraction models by integrating highlight information into abstracts. Further analysis indicates we can filter out irrelevant abstract sentences based on their semantic similarity to the highlights. Overall, this study pioneers the exploration of utilizing paper highlights for keyword extraction, simultaneously boosting unsupervised keyword extraction performance and providing insights for other abstract‐based text mining applications.

Sentiment Analysis Of Ijen Crater Reviews Using Naïve Bayes Classification And Oversampling Optimization

Sentiment analysis is a method that applies text mining concepts to provide classifications that have polarity that is positive, negative, or neutral from each sentence or document. In this context, the purpose of this research is to analyse the sentiment of user reviews related to the Ijen crater tourist attractions found on the Google Maps platform. This research is conducted in three main stages: first, Data Collection and Preprocessing by taking data samples obtained from Ijen Crater reviews contained on Google Maps; second Optimisation and Classification by changing the minority class samples to be almost equal to the majority class by randomly duplicating the minority class samples, third, classification performance measurement using confusion matrix. The test is conducted by comparing the performance between NBC classification without optimisation and NBC classification with SMOTE and ADASYN optimisation. The performance results show that SMOTE-optimised NBC classification provides the best improvement in accuracy by 6.74% compared to the performance of ordinary NBC and NBC added with ADASYN.

Uncovering airline service quality dimensions from passengers' online reviews: A text mining approach to validate and extend Servqual

This study investigates airline service quality dimensions by employing text-mining techniques to analyse passengers' emotional responses and experiences with airlines as reflected in their online reviews. The study aims to identify the relationship between these and SERVQUAL dimensions to confirm its reliability in the airline industry and reveal new and promising dimensions of airline service quality. The study contributes to academia by enriching literature on airline service quality and text mining applications, identifying areas for enhancement, and suggesting strategic directions for continuous improvement to practitioners based on passengers' quality perceptions of airline service experiences. With a large dataset from 100 airlines obtained from Skytrax, the study offers a deeper understanding of airline service quality. It encourages using big data and text-mining techniques to explore consumer preferences. The study's findings challenge traditional models and bring out new, context-specific service quality characteristics to provide a greater understanding of passenger experiences. These give the airlines insightful guidance on improving service offerings and fulfilling customer expectations, eventually improving customer satisfaction, loyalty, and competitive positioning.

A Semantic Analysis of the User-Generated Content from Social Network Sites of Cashbuild Limited: Application of Text Mining, Machine Learning and Big Data Analytics toward the Development of a Customer-Centric Marketing Strategy

Social Network Sites (SNS) have grown rapidly over the past 15 years, which is in line with the growth of smartmobile phones. A key phenomenon of SNSs is User-Generated Content (UGC). It enables social engagements between users and their followers. UGC shares various perspectives on news, entertainment, and fun, as well as their perspectives on brands like Customer Experience (CX) and Customer Satisfaction (CS). UGC stimulates the Brand Attitudes (BA) of the readers if the user is trusted and credible. BAs influence the Purchase Intentions (PI) of the readers of UGC. Marketers like that UGC influences PI. They learn about CX and CS, which users share on the UGC. This semantic study uses Text Mining (TM) and Big Data Analysis (BDA) to extract keywords from the UGC of a building materials retail company, Cashbuild Limited (CBL) from the Republic of South Africa (RSA). The UGC was analysed using Power BI (PBI) to identify its sources and sentiments. The majority of the UGC in this study was obtained from Facebook (94.51%), followed by X.com (3.95%). The rest of the UGC was obtained from Blogs (0.79%) and Instagram (0.76%). 65.63% of the UGC was neutral, 29.74% positive, and 4.46% negative. UGC influences PI by mediating hedonic and utilitarian Brand Equity. The generation of a high-quality UGC was encouraged by CS - an outcome of CX. CX comes from the touchpoints across the CJ. Marketers want to ensure a good customer experience (CX) to achieve customer satisfaction (CS). Ultimately, CS influences future customer expectations (CE) when shared through UGC. 84 top keywords were extracted out of 16 328 hit sentences (HS) from the UGC of CBL using the MELTWATER program. The keywords were ranked to 51 frequency positions. CONCOR analysis was used to determine the connectivity and centrality of the keywords using Freeman’s coefficient and the Eigenvector value for each keyword. The results of the CONCOR analysis were used to identify 41 significant keywords based on their higher connectivity and centrality. The 41 significant keywords were matched with the 16 328 HSs. 14 keywords that matched less than 100 HSs were excluded. The remaining 27 keywords were subjected to exploratory factor analysis EFA on Statistical Program for Social Sciences (SPSS). 21 keywords were loaded on 8 factors, and the model explained 68.363% of the variance with a Kaizer Meyer Olkin (KMO) of 0.736 above 0.6. The 8 categories were compared to input from a panel of industry experts. 7 categories proposed by experts matched 7 out of 8 categories from the statistical model. 7 key themes proposed to be used in the study to develop Customer-centric Marketing Strategies (CCMS) in the building materials retail industry (BMRI) are DIY Kings, House Proud, Project Master, Renovators, Interior Image, Building Materials Inventory, and Promotions. All 5 objectives of the study were met. The study adds value to the development of CCMS learning from the UGC.

Application of text mining in analysing notes to financial statements: A Hungarian case

Company stakeholders must have reliable and accurate information about the companies falling into their sphere of interest. In Hungary, one of the key sources of information for company stakeholders is the financial statements and related explanations, which are included in the notes of the financial statements (notes). This study used text mining to analyse the Hungarian annual financial statements notes for 2017, 2019 and 2021. The selection of the notes was based on the proportions of each sector in the national economy. The research analysed 28,700 company notes annually, totalling 86,100 documents for the three years. The text mining and generation of the Term Frequency Matrix have performed 'quanteda' packages of the R statistical system, which incorporate the results of artificial intelligence research to enhance the efficiency of text mining. Based on the results, the contents of the notes to the financial statements appear to be a rather mixed picture in Hungary. Analysing the term frequency matrix for the 67 most common terms has revealed no significant difference between the years. However, considerable differences have been caused by size categories and sectors. The notes are statistically significant using Jaccard similarity analysis, considering the year, corporate size, and sector.

State of the art of the cow-calf systems in beef and dairy cattle (Bos taurus) operations in EU, USA, and Brazil from 1998 to 2023

Cow-calf systems represent a significant research area in animal husbandry, with differences depending on the final product (meat or milk). This study aimed to apply text mining and topic analysis on literature describing cow-calf systems in European, American, and Brazilian beef and dairy sectors between 1998 and 2023. Additionally, cow-calf contact (CCC) literature data was manually extracted. Our findings revealed the presence of 11 research areas among literature on cow-calf systems, with different priorities identified in the beef and dairy sectors. Beef industry mainly focused on animal proficiency and nutrition, while dairy on animal welfare and CCC, which showed a growing trend as emerging research topic, mostly in the EU. Current debates around calf welfare and EU's planned animal welfare legislation revision appeared to be driving the increasing interest in this topic. Studies in the beef sector were mainly localized in Brazil, showing that research in different contexts and species is important for CCC implementation. Manual data extraction showed considerable variation in the retained CCC documents regarding sample size, type of contact, methods and CCC duration. Learning about the varied CCC approaches used in beef and dairy farms in different locations, concentrating on their strengths and weaknesses, will help to develop novel solutions to global challenges. Adopting validated and robust indicators would help scientists and policymakers to monitor the system's quality. To improve CCC feasibility, match consumer demands, and move towards One Welfare and One Health, future research should focus on a variety of situations to overcome the current shortcomings.

Analysis of child development facts and myths using text mining techniques and classification models

The rapid dissemination of misinformation on the internet complicates the decision-making process for individuals seeking reliable information, particularly parents researching child development topics. This misinformation can lead to adverse consequences, such as inappropriate treatment of children based on myths. While previous research has utilized text-mining techniques to predict child abuse cases, there has been a gap in the analysis of child development myths and facts. This study addresses this gap by applying text mining techniques and classification models to distinguish between myths and facts about child development, leveraging newly gathered data from publicly available websites. The research methodology involved several stages. First, text mining techniques were employed to pre-process the data, ensuring enhanced accuracy. Subsequently, the structured data was analysed using six robust Machine Learning (ML) classifiers and one Deep Learning (DL) model, with two feature extraction techniques applied to assess their performance across three different training-testing splits. To ensure the reliability of the results, cross-validation was performed using both k-fold and leave-one-out methods. Among the classification models tested, Logistic Regression (LR) demonstrated the highest accuracy, achieving a 90 % accuracy with the Bag-of-Words (BoW) feature extraction technique. LR stands out for its exceptional speed and efficiency, maintaining low testing time per statement (0.97 μs). These findings suggest that LR, when combined with BoW, is effective in accurately classifying child development information, thus providing a valuable tool for combating misinformation and assisting parents in making informed decisions.

Advancing Sentiment Analysis Through Emotionally-Agnostic Text Mining in Large Language Models (LLMS)

The conventional methodology for sentiment analysis within large language models (LLMs) has predominantly drawn upon human emotional frameworks, incorporating physiological cues that are inherently absent in text-only communication. This research proposes a paradigm shift towards an emotionallyagnostic approach to sentiment analysis in LLMs, which concentrates on purely textual expressions of sentiment, circumventing the confounding effects of human physiological responses. The aim is to refine sentiment analysis algorithms to discern and generate emotionally congruent responses strictly from text-based cues. This study presents a comprehensive framework for an emotionally-agnostic sentiment analysis model that systematically excludes physiological indicators whilst maintaining the analytical depth required for accurate emotion detection. A novel suite of metrics tailored to this approach is developed, facilitating a nuanced interpretation of sentiment within text data, which is paramount for enhancing user interaction across a spectrum of text mining applications, including recommendation systems and interactive AI characters. The research undertakes a critical comparative analysis, juxtaposing the newly proposed model with traditional sentiment analysis techniques, to evaluate efficacy enhancements and to substantiate its application potential. Further, the investigation delves into the short-term memory capabilities of LLMs, exploring the implications for character AI roleplaying interfaces and their ability to recall and respond to user input within a text-driven emotional framework. Findings indicate that this emotionally-agnostic sentiment analysis approach not only simplifies the sentiment assessment process within LLMs but also opens avenues for a more precise and contextually appropriate emotional response generation. This abstracted model, devoid of human physiological constraints, represents a significant advancement in text mining, fostering improved interactions and contributing to the evolution of LLMs as adept analytical tools in various domains where emotional intelligence is crucial.

Empowering healthcare with NLP-driven deep learning unveiling biomedical materials through text mining

This study presents a comprehensive approach to automated biomedical materials discovery in healthcare applications by integrating text mining and deep learning techniques. The research methodology encompasses two main components: exploration of key research questions through graphical representation and evaluation of model performance using precision, recall, and F1-score metrics. The identification of pertinent research questions is visualized using bar charts, offering insights into the distribution of studies across domains such as data harmonization, heterogeneity, industrial textual data, and sequential data performance. The precision comparison chart highlights the strengths and weaknesses of different models, with model 1 demonstrating notable precision. The recall comparison chart emphasizes model 2’s outstanding performance in capturing relevant information, while the f1-score comparison chart showcases the balanced metrics of model 2 and 4. These visual analyses contribute to a nuanced understanding of the research landscape and guide the development of the proposed NLP-ML pipeline. The study’s findings underscore the significance of addressing data harmonization challenges and extracting insights from industrial textual data in advancing biomedical materials discovery. Overall, this research amalgamates exploratory data analysis and quantitative model evaluation to contribute to the evolving field of text mining and deep learning applications in biomedical material discovery for healthcare applications.

Influence of the perceptions of amenities on consumer emotions in urban consumption spaces.

The purpose of this study is to understand how consumers perceive amenities and their impact on promoting positive consumer emotions and comprehend the role of amenities in fostering urban consumption. We collected over 120,000 comments spanning 60 months (2015-2019) from 18 urban shopping centers in Shanghai. By applying text mining methods, we constructed a panel data model on the perception of four types of amenities and consumer emotions. Results indicate that different perceptions of amenities yield significantly different effects on consumer emotions. Specifically, we found that perceptions of cultural and safeguarded amenities significantly influence consumer emotions, albeit with different directions of impact. In contrast, perceptions of commercial and natural amenities did not significantly affect consumer emotions. The findings of this study provide a key reference for how to scientifically plan and reasonably introduce the types of amenities in urban consumption space, so as to reflect the promoting effect of amenities on urban consumption.

An Ensemble Keyword Extraction Model for News Texts with Statistical and Graphical Features

Keyword extraction is an essential tool for many text mining applications such as automatic indexing, summarizing, classification, clustering and automatic filtering. Automated keyword extraction is essential as the daily text data to be reached and processed have increased tremendously over the Internet, e.g. millions of news articles are published daily online. In this paper, a novel ensemble model for automatic extraction of keywords from news articles is proposed. The proposed model handles keyword extraction as a sequence labeling task. Two sub-modules representing the statistical and graphical features by their calculated scores for each input token were combined in the token classification module. The Ensemble Token Classification module was trained and tested separately with the ensemble algorithms Random Forest, XgBoost, Decision Tree and Voting Classification. For training, we collected two news datasets from Kazakh and Russian news sites published in Cyrillic alphabet. We also collected an Arabic news dataset, ArabianNews. The performance of the model was also compared with the widely used 500N-KPCrowd dataset in the literature, which consists of English news content in Latin alphabet. The proposed model achieved the best performance with an [Formula: see text]-score of 0.71 and 0.86 on the 500N-KPCrowd and Russian datasets, respectively. We attained the best [Formula: see text]-score (0.97) with the KazakhNews and ArabianNews datasets.

Contribution to the harm assessment of darknet markets: topic modelling drug reviews on Dark0de Reborn

Amid the global opioid crisis, the volume of drug trade via darknet markets has risen to an all-time high. The steady increase can be explained by the reliable operation of darknet markets, affected by community-building trust factors reducing the risks during the process of the darknet drug trade. This study was designed to explore the risk reduction efforts of the community of a selected darknet market and therefore contribute to the harm assessment of darknet markets. We performed Latent Dirichlet Allocation topic modelling on customer reviews of drug products (n = 25,107) scraped from the darknet market Dark0de Reborn in 2021. We obtained a model resulting in 4 topics (coherence score = 0.57): (1) feedback on satisfaction with the transaction; (2) report on order not received; (3) information on the quality of the product; and (4) feedback on vendor reliability. These topics identified in the customer reviews suggest that the community of the selected darknet market implemented a safer form of drug supply, reducing risks at the payment and delivery stages and the potential harms of drug use. However, the pitfalls of this form of community-initiated safer supply support the need for universally available and professional harm reduction and drug checking services. These findings, and our methodological remarks on applying text mining, can enhance future research to further examine risk and harm reduction efforts across darknet markets.

Comorbidity-Guided Text Mining and Omics Pipeline to Identify Candidate Genes and Drugs for Alzheimer's Disease.

Alzheimer's disease (AD), a multifactorial neurodegenerative disorder, is prevalent among the elderly population. It is a complex trait with mutations in multiple genes. Although the US Food and Drug Administration (FDA) has approved a few drugs for AD treatment, a definitive cure remains elusive. Research efforts persist in seeking improved treatment options for AD. Here, a hybrid pipeline is proposed to apply text mining to identify comorbid diseases for AD and an omics approach to identify the common genes between AD and five comorbid diseases-dementia, type 2 diabetes, hypertension, Parkinson's disease, and Down syndrome. We further identified the pathways and drugs for common genes. The rationale behind this approach is rooted in the fact that elderly individuals often receive multiple medications for various comorbid diseases, and an insight into the genes that are common to comorbid diseases may enhance treatment strategies. We identified seven common genes-PSEN1, PSEN2, MAPT, APP, APOE, NOTCH, and HFE-for AD and five comorbid diseases. We investigated the drugs interacting with these common genes using LINCS gene-drug perturbation. Our analysis unveiled several promising candidates, including MG-132 and Masitinib, which exhibit potential efficacy for both AD and its comorbid diseases. The pipeline can be extended to other diseases.

Signature size and individual preferences

In this digital economy, customers' handwriting, such as signatures and related characteristics, can be easily collected and quantitatively analyzed. From the perspective of decision-making, obtaining information from signature characteristics is an important data-driven business leverage. In this regard, can we infer individual preferences from signatures? Through a large-scale laboratory experiment, this study measured the participants’ risk and time preferences and collected their signature information in order to determine the influence of behavioral preferences on their signatures. Based on the empirical results, the higher the risk and ambiguity aversion of the participants, the smaller were their signature sizes. Moreover, the greater their level of patience, the more likely their signatures showed a trend of increasing size. Overall, this study fills a research gap in signature-related behavior by highlighting the application of text mining in digital handwriting, especially in broad business settings such as the insurance industry.

A Thematic Analysis of English and American Literature Works Based on Text Mining and Sentiment Analysis

A theme analysis model integrating text mining and sentiment analysis has emerged as a powerful tool for understanding English and American literary works. By employing techniques such as topic modeling, keyword extraction, and sentiment analysis, this model can identify recurring themes, motifs, and emotional tones within texts. Through text mining, it extracts key concepts and topics, while sentiment analysis discerns the underlying emotions conveyed by the authors. By combining these approaches, researchers can uncover deeper insights into the thematic elements and cultural contexts of English and American literature. This paper explores the application of text mining and sentiment analysis techniques to analyze a dataset comprising American literary works. With computational methods such as bi-gram analysis, multimodal feature extraction, and sentiment analysis using the Bi-gram Multimodal Sentimental Analysis (Bi-gramMSA) approach. With the proposed Bi-gramMSA model the multimodal features in the American Literature are examined to investigate the thematic, emotional, and multimodal aspects of the literature. Through our analysis, we uncover significant bi-grams, extract multimodal features, and assess sentiment distribution across the texts. The results highlight the effectiveness of these computational methodologies in uncovering patterns, sentiments, and features within the literary corpus. The proposed Bi-gramMSA model achives a higher score for the different scores in the Chinese Literature.

A Survey of the Applications of Text Mining for the Food Domain

In the food domain, text mining techniques are extensively employed to derive valuable insights from large volumes of text data, facilitating applications such as aiding food recalls, offering personalized recipes, and reinforcing food safety regulation. To provide researchers and practitioners with a comprehensive understanding of the latest technology and application scenarios of text mining in the food domain, the pertinent literature is reviewed and analyzed. Initially, the fundamental concepts, principles, and primary tasks of text mining, encompassing text categorization, sentiment analysis, and entity recognition, are elucidated. Subsequently, an analysis of diverse types of data sources within the food domain and the characteristics of text data mining is conducted, spanning social media, reviews, recipe websites, and food safety reports. Furthermore, the applications of text mining in the food domain are scrutinized from the perspective of various scenarios, including leveraging consumer food reviews and feedback to enhance product quality, providing personalized recipe recommendations based on user preferences and dietary requirements, and employing text mining for food safety and fraud monitoring. Lastly, the opportunities and challenges associated with the adoption of text mining techniques in the food domain are summarized and evaluated. In conclusion, text mining holds considerable potential for application in the food domain, thereby propelling the advancement of the food industry and upholding food safety standards.

Unveiling the tapestry of mother–child interactions through text mining and sentiment analysis

The interaction between a mother and child stands as one of the most profound and intricate human connections, weaving a rich tapestry of behavioral and emotional bonds during the formative years. Although mother–child interactions have received substantial attention in the developmental science literature, few studies have tapped into the extensive corpus of speech data available to uncover the nuances of these interactions across developmental stages. This study applied text mining and sentiment analysis on narratives extracted from mother–child conversations to identify the developmental trend of mother–child interactions from early to middle childhood. The results, based on three key areas of dyadic interactions, demonstrated a shift toward more balanced turn-taking dynamics and linguistic congruence as children age. Also, there was a significant interdependence of mother and child expressed emotions across time. Further investigation of the dyadic emotionality revealed a nonlinear effect of mother-expressed emotion on child-expressed emotion: mother-expressed negative emotions followed a cubic-like pattern, while positive emotions followed a mild quadratic trend. Taken together, the findings of this study present a picture of progressive augmentation of mother–child synchrony over time.