Search

Hyperspectral data and machine learning offer great potential for identifying valuable open ecosystems. Due to the large volume of data, preprocessing of hyperspectral images must involve dimensionality reduction. The main goal of this study was to test the effectiveness of various types of feature reduction (feature selection and feature extraction) when performing classification using the Random Forest algorithm. A comparison was conducted between two ecosystems - heathlands and mires protected as Natura 2000 habitats. Two transformations of feature extraction were chosen, namely Minimum Noise Fraction (MNF) and Principal Component Analysis (PCA), while Linear Discriminant Analysis (LDA) was used as a feature selection method. It was proven that irrespective of the class type, accuracy is higher with the feature extraction method (mean F1 accuracy of 0.922) than with feature selection (mean F1 accuracy of 0.787). At the same time, no significant differences in accuracies were found between the MNF and PCA methods. Although LDA resulted in lower accuracies (0.816 for heathland and 0.750 for mires), the method could also be used due to relatively high F1 values. The effectiveness of the LDA method for feature reduction in open ecosystem identification was confirmed for the first time for open natural vegetation. [ABSTRACT FROM AUTHOR]

Surface‐enhanced Raman spectroscopy (SERS) was utilized to measure low‐level fentanyl concentrations mixed in common cutting agents, cocaine, 3,4‐methylenedioxymethamphetamine (MDMA), methamphetamine, and caffeine. Mixtures were prepared with a fentanyl concentration range of 0–339 μM. Data was initially analyzed by plotting the area of a diagnostic peak (1026 cm−1) against concentration to generate a calibration model. This method was successful with fentanyl/MDMA samples (LOD 0.04 μM) but not for the other mixtures. A chemometric approach was then employed. The data was evaluated using principal component analysis (PCA), partial least squares (PLS1) regression, and linear discriminant analysis (LDA). The LDA model was used to classify samples into one of three designated concentration ranges, low = 0–0.4 mM, medium = 0.4–14 mM, or high >14 mM, with fentanyl concentrations correctly classified with greater than 85% accuracy. This model was then validated using a series of “blind” fentanyl mixtures and these unknown samples were assigned to the correct concentration range with an accuracy >95%. The PLS1 model failed to provide accurate quantitative assignments for the samples but did provide an accurate prediction for the presence or absence of fentanyl. The combination of the two models enabled accurate quantitative assignment of fentanyl in binary mixtures. This work establishes a proof of concept, indicating a larger sample size could generate a more accurate model. It demonstrates that samples, containing variable, low concentrations of fentanyl, can be accurately quantified, using SERS. [ABSTRACT FROM AUTHOR]

The work focuses on the development of a fluorimetric version of the fingerprint method based on conducting indicator reactions in the presence of a test sample. Observation of these reactions over time provides a more detailed information compared to batch methods, thereby improving sample recognition and enabling quantitative analysis. The proposed indicator reaction utilizes a commercial carbocyanine dye and 4-dimethylaminobenzaldehyde, whose interaction results in a decrease in fluorescence intensity and changes in absorbance over time. Three fluoroquinolones—moxifloxacin, levofloxacin, and ofloxacin—selectively change the signal in concentrations of 1 μM or higher, while other drugs, including different fluoroquinolones, do not interfere with the determination. Ofloxacin was detected in human urine samples at various times post-drug intake. A potential of using the same indicator reaction for sample recognition was demonstrated on examples of apple juices, soil extracts, and meat of varying freshness. Chemometric methods, including linear discriminant analysis, were used for data processing. The method achieved 97% accuracy in discriminating fifteen apple juice samples, 94% accuracy for ten apple juices from 2022 and 2023, 99% accuracy for ten soil samples, and successfully determined the freshness of five meat samples. [ABSTRACT FROM AUTHOR]

Although interdisciplinary research has garnered extensive attention in academia, its core knowledge structure has yet to be systematically explored. To address this gap, this study aims to uncover the underlying core knowledge topics within interdisciplinary research, enabling researchers to gain a deeper understanding of the knowledge framework, improve research efficiency, and offer insights for future inquiries. Based on the Web of Science (WoS) database, this study collected 153 highly cited papers and employed the LDA topic model to identify latent topics and extract the knowledge structure within interdisciplinary research. The findings indicate that the core knowledge topics of interdisciplinary research can be categorized into four major areas: the knowledge framework and social impact of interdisciplinary research, multidisciplinary approaches in cancer treatment and patient care, Covid-19 multidisciplinary care and rehabilitation, and multidisciplinary AI and optimization in industrial applications. Moreover, the study reveals that AI-related interdisciplinary research topics are rapidly emerging. Through an in-depth analysis of these topics, the study discusses potential future directions for interdisciplinary research, including the cultivation and development of interdisciplinary talent, evaluation systems and policy support for interdisciplinary research, international cooperation and interdisciplinary globalization, and AI and interdisciplinary research optimization. This study not only uncovers the core knowledge structure of interdisciplinary research but also demonstrates the effectiveness of the LDA topic model as a data mining tool for revealing key topics and trends, providing practical tools for future research. However, this study has two main limitations: the time lag of highly cited papers and the dynamic evolution of interdisciplinary research. Future research should address these limitations to further enhance the understanding of interdisciplinary research. [ABSTRACT FROM AUTHOR]

Online tourism spot recommendations, as a key component of tourism services, aim to present travel options that align with users' personal preferences. However, current recommendation systems often underperform due to the sparsity of tourism data and the wide variance in user preferences. To address this challenge, we propose a Semantic Analysis-Based Tourism Recommendation framework, abbreviated as SABTR (Semantic Analysis-Based Tourism Recommendation). The framework comprises two stages: Firstly, Latent Dirichlet Allocation (LDA) models are utilized to deeply mine data between users and attractions, constructing two core matrices: the user similarity matrix and the attraction similarity matrix. Secondly, based on the user similarity matrix, similarity calculation methods are applied to predict ratings for tourism spots that users have not yet evaluated. Simultaneously, within the attraction similarity matrix, probability distributions for each attraction across various thematic interests are calculated. When the system identifies a user's interest in specific types of attractions, SABTR can select a series of related attractions from associated interest tags. Then, these candidate attractions are ranked according to both known and predicted user ratings, ultimately forming personalized attraction packages recommended to users. Extensive experiments have demonstrated that compared to existing tourism recommendation solutions, our method significantly improves the quality of attraction recommendations and enhances user satisfaction. [ABSTRACT FROM AUTHOR]

In this paper, we present the implementation of the Density Functional Theory (DFT) method using the Geant4-DNA framework in the Single Instruction Single Data (SISD) mode. Furthermore, this implementation is improved in terms of execution time within the GeantV project with vectorization techniques such as Single Instruction Multiple Data (SIMD). Within this framework, a set of SIMD strategies in molecular calculation algorithms such as one-electron operators, two-electron operator, quadrature grids, and functionals, was implemented using the VecCore library. The applications developed in this work implement two DFT functionals, the Local Density Approximation (LDA) and the General Gradient Approximation (GGA), to approximate the molecular ground-state energies of small molecules and amino acids. To assess the performance of the implementations, a standard test simulation was performed in multiple CPU platforms. The SIMD vectorization strategy significantly accelerates DFT calculations, leading to time ratios ranging from 1.6 to 5.4 in either individual steps or entire implementations when compared with the scalar process within Geant4. [ABSTRACT FROM AUTHOR]

AbstractThis study aims to investigate the thematic characteristics and evolutionary patterns of library policies in China from 1949 to 2023. Employing the Latent Dirichlet Allocation (LDA) model, the research analyzed 2,703 Chinese government-issued policies related to libraries during this period, identifying key topics and tracing their developmental trajectories. The study recognized four distinct developmental phases, five major topics, and the underlying connections among feature terms within different topics. It discussed the corresponding topic strengths and evolving trends in each phase. Library policies in China are predominantly centered on five key domains: 1) spatial planning and architectural integration, 2) information management and transparency, 3) legal regulations and supervisory enforcement, 4) heritage preservation and sustainable development, and 5) community inclusion and accessibility. Notably, spatial planning and architectural integration consistently garner substantial attention, highlighting their enduring prominence, while there is relatively less emphasis on community inclusion and accessibility, as well as heritage preservation and sustainable development. The contribution of this study lies in conducting a semantic exploration of the overall transformation in library-related policies in China, validating the effectiveness of the LDA model in identifying topics within policy texts, and expanding the research perspective on library policy evolution. [ABSTRACT FROM AUTHOR]

Mango is a highly diverse tropical fruit with numerous varieties that differ in flavor, texture, and chemical composition. Consequently, identifying fraudulent substitutions of mango varieties poses a significant challenge using traditional techniques. Therefore, there is an increasing need for new methods to discriminate between mango varieties. Hyperspectral imaging coupled with machine learning techniques presents a promising approach for varietal discrimination. In this study, mango samples of eleven varieties were collected from a germplasm bank, with four slices obtained from each sample. Hyperspectral images were acquired in the Vis–NIR and NIR ranges for each slice, and spectral profiles were extracted and pretreated. Three discrimination models, linear discriminant analysis, K-nearest neighbor, and artificial neural networks, were implemented and validated using relevant wavelengths selected through a covering array feature selection algorithm. The performance of these models was evaluated using precision, accuracy, and F-score metrics. The average spectral profiles of the studied varieties exhibited a similar behavior with slight differences, which could be used for classification within the evaluated ranges. The optimal number of variables selected to refine the models was 17 for the UV–Vis–NIR range and 21 for the NIR range, with an accuracy ranging between 0.752 and 0.972. This study concludes that hyperspectral imaging combined with machine learning techniques can effectively discriminate between different varieties of mango. [ABSTRACT FROM AUTHOR]

Background/Objectives: Photoplethysmography (PPG) signals, which measure blood volume changes through light absorption, are increasingly used for non-invasive cardiovascular disease (CVD) detection. Analyzing PPG signals can help identify irregular heart patterns and other indicators of CVD. Methods: This research involves a total of 41 subjects sourced from the CapnoBase database, consisting of 21 normal subjects and 20 CVD cases. In the initial stage, heuristic optimization algorithms, such as ABC-PSO, the Cuckoo Search algorithm (CSA), and the Dragonfly algorithm (DFA), were applied to reduce the dimension of the PPG data. Next, these Dimensionally Reduced (DR) PPG data are then fed into various classifiers such as Linear Regression (LR), Linear Regression with Bayesian Linear Discriminant Classifier (LR-BLDC), K-Nearest Neighbors (KNN), PCA-Firefly, Linear Discriminant Analysis (LDA), Kernel LDA (KLDA), Probabilistic LDA (ProbLDA), SVM-Linear, SVM-Polynomial, and SVM-RBF, to identify CVD. Classifier performance is evaluated using Accuracy, Kappa, MCC, F1 Score, Good Detection Rate (GDR), Error rate, and Jaccard Index (JI). Results: The SVM-RBF classifier for ABC PSO dimensionality reduced values outperforms other classifiers, achieving the highest accuracy of 95.12% along with the minimum error rate of 4.88%. In addition to that, it provides an MCC and kappa value of 0.90, a GDR and F1 score of 95%, and a Jaccard Index of 90.48%. Conclusions: This study demonstrated that heuristic-based optimization and machine learning classification of PPG signals are highly effective for the non-invasive detection of cardiovascular disease. [ABSTRACT FROM AUTHOR]

Comments on both community-based and transaction-based platforms include tourism attractions, tourism services, and tourist emotions, but whether these two types will lead to similarities and differences in the overall image and segmented image is a proposition worth exploring. Based on the three-dimensional model of "cognition-emotion-integrity", Lijiang ancient city was used as the case area, Sina Weibo and Ctrip reviews representing community-based and transactional data respectively were used as the research data, combined with Python programming to call Baidu Sentiment Dictionary sentiment analysis method, used LDA to refine image themes, and constructed a co-occurrence relationship network map combining VOSviewer and Gephi to Analyze the similarities and differences between two types of data-driven overall image, cognitive image and emotional image and their influencing factors. The results show that 1) The overall image of destinations based on Sina Weibo's community-based and Ctrip's transaction-based reviews is mainly positive, with similar changes in annual image changes, and the distribution of emotional polarity is basically the same. 2) Although both community-based and transaction-based reviews can reflect cognitive and emotional image themes such as destination services, attractions, and environmental economy, the cognitive image theme score of the former is significantly higher than that of the latter, while the emotional image is the opposite . 3) There is little difference in the contribution rate of cognitive and emotional factors to the overall image of the destination driven by the two kinds of review data, while the effect of the economic price factor of the transaction type review is significantly higher than that of the community type review. The research provides a certain reference basis for using large-scale multi-source heterogeneous data to analyze destination image and tourism marketing work. [ABSTRACT FROM AUTHOR]

Accurately classifying road signs is crucial for autonomous driving due to the high stakes involved in ensuring safety and compliance. As Convolutional Neural Networks (CNNs) have largely replaced traditional Machine Learning models in this domain, the demand for substantial training data has increased. This study aims to compare the performance of classical Machine Learning (ML) models and Deep Learning (DL) models under varying amounts of training data, particularly focusing on altered signs to mimic real-world conditions. We evaluated three classical models: Support Vector Machine (SVM), Random Forest, and Linear Discriminant Analysis (LDA), and one Deep Learning model: Convolutional Neural Network (CNN). Using the German Traffic Sign Recognition Benchmark (GTSRB) dataset, which includes approximately 40,000 German traffic signs, we introduced digital alterations to simulate conditions such as environmental wear or vandalism. Additionally, the Histogram of Oriented Gradients (HOG) descriptor was used to assist classical models. Bayesian optimization and k-fold cross-validation were employed for model fine-tuning and performance assessment. Our findings reveal a threshold in training data beyond which accuracy plateaus. Classical models showed a linear performance decrease under increasing alteration, while CNNs, despite being more robust to alterations, did not significantly outperform classical models in overall accuracy. Ultimately, classical Machine Learning models demonstrated performance comparable to CNNs under certain conditions, suggesting that effective road sign classification can be achieved with less computationally intensive approaches. [ABSTRACT FROM AUTHOR]

Simple Summary: Left displaced abomasum is a common disease among Holstein cattle that is causing significant economic losses. The exact causes of the disease are unknown, but it is assumed that LDA is multifactorial in nature. Predisposition to the displacement has been shown to be inherited. We conducted a genome-wide study to identify statistical associations between certain mutations and left displaced abomasum in Holstein cows from several farms in the Leningrad region. We found suggestive statistical associations with this disease and several mutations located in a set of genes involved in a variety of physiological processes. Some of these processes may be directly related to the development of the disease being studied. In addition, we conducted runs of homozygosity analysis and found that cows with displaced abomasum are more likely than healthy cows to carry a specific state of genomic regions containing genes associated with body shape, including width. This may indicate the role of this trait in the development of the disease. Left displaced abomasum (LDA) is a multifactorial disease of cattle that occurs mainly during the transition postpartum period and is characterized by a decrease in milk production and an increased risk of culling. Several studies have been conducted confirming the hereditary nature of predisposition to this disease. The aim of our study is to identify genetic associations characterizing the genomic variability of susceptibility to LDA in Holstein cattle of the Leningrad region of the Russian Federation. The objects of this study were 360 highly productive dairy cows divided into two groups: animals with LDA, and healthy ones (control). Runs of homozygosity analysis revealed one ROH on BTA13 that was found to be significantly more prevalent in the group of animals with LDA than in the healthy group. Fourteen candidate SNPs were found to be nominally associated with left displacement of the abomasum (p-value < 1 × 10−4). When performing functional annotation of genes containing associated polymorphisms or located close to them, candidate genes presumably associated with the development of LDA were identified: ABCB11, SRP72, RGS18, SOX4, GSG1L, FBXL19, and PNPLA4. [ABSTRACT FROM AUTHOR]

Aspect-based sentiment analysis aims to identify sentiment polarities towards specific aspects in textual data. Despite extensive aspect based sentiment analysis research, accurately capturing nuanced semantics remains challenging. This paper presents a novel transformer-based sequence prediction approach using generative pretrained transformer model that overcomes current limitations. The generative pretrained transformer model is an expert in understanding the context of text and aids in identifying minute nuances in text that are useful for sentiment analysis. A hotel review dataset is leveraged for rigorous analysis. Key technical innovations include a Latent Dirichlet Allocation-based aspect extraction method capable of grouping into representative topics, followed by sophisticated generative pretrained transformer model fine-tuning optimized for the granularity and semantic nuances of sentiments and aspects. Extensive quantitative experiments demonstrate 90% train and 84% test accuracy, outperforming previous state-of-the-art convolutional neural network models. Qualitative evaluations further showcase capabilities in modeling interdependent aspect semantics neglected by existing works. [ABSTRACT FROM AUTHOR]

Purpose: This study aims to provide insights into different aspects of the extant literature on the effects of dividend announcements. Along with other outputs of a bibliometric study, this study provides deeper insights into the concentration of the extant literature and suggest future research agendas. Design/methodology/approach: This study uses the bibliometric, network and content analysis of the dividend announcement literature indexed in Scopus. This study presents the temporal analysis, the network of authors, countries, author citations and the co-occurrence of author keywords. This study provides the concentration of the extant literature in three clusters and unearth some key future research areas. This study uses the latent Dirichlet allocation method for robustness. Findings: A total of 54 documents examining the US sample have received 1,804 citations. Interestingly, the first article on emerging markets was published in 2002, when at least 34 articles on developed markets had already been published from 1982 to 2001. The content analysis of top-cited literature unveils diverse insights into dividend announcements' effects on financial markets. Contagion effects negatively impact non-announcing banks, particularly larger ones. Dividend maintenance affects stock market momentum, influencing loser returns. While current dividend/earnings news may not predict future company performance, information content dominates bond market reactions to post-dividend announcements. Concomitantly, while financially constrained firms exhibit short-term gains but worse long-term performance following dividend increases, larger stock dividends send stronger market signals in China. Originality/value: This study significantly contributes to the bibliometric and content analysis literature by analyzing the sample documents based on the sample examined. To the best of the authors' knowledge, no previous bibliometric study in this domain has been conducted to explore the markets (developed and emerging) to which the samples examined belong and the quality of publications from developed and emerging markets. [ABSTRACT FROM AUTHOR]

Game reviews heavily influence public perception. User feedback is crucial for developers, offering valuable insights to enhance game quality. In this research, Metacritic game reviews for Elden Ring were analyzed for topic modeling using Latent Dirichlet Allocation (LDA), Bidirectional Encoder Representations from Transformers (BERT), and a hybrid model combining both to identify effective methods for extracting underlying themes in player feedback. We analyzed and interpreted these models' outputs to learn the game reviews. We aimed to identify the differences, similarities, and variations between the three to determine which provided more valuable and instructive information. Our findings indicate that each method successfully identified keywords with some similarities in identified words. The LDA model had the highest silhouette score, indicating the most distinct clustering. The LDA-BERT model had a 1% higher coherence score than LDA, indicating more meaningful topics. [ABSTRACT FROM AUTHOR]

The ability to differentiate species is not adequate for modern forage breeding programs. The measurement of persistence is currently a bottleneck in the breeding system that limits the throughput of cultivars to the marketplace and prevents it from being selected as a trait. The use of hyperspectral data obtained through remote sensing offers the potential to reduce guesswork by identifying the distribution of pasture species, but only if such data alone can distinguish the subtle differences within species, i.e., cultivars. The implementation of this technology faces many challenges due to the spectral and temporal variability of species. To understand the spectral variability between and within species groups, differentiation using hyperspectral data from monoculture plots of turf species was utilized. Spectral data were collected over a year using an ASD FieldSpec® and canopy pasture probe (CAPP). The plots consisted of monocultures of various species, and cultivars (a total of 10 plots). Linear discriminant analysis (LDA) was conducted on the full spectrum and reduced band data. This technique successfully differentiated the species with high accuracy (>98%). We demonstrate the potential of hyperspectral data and analysis techniques to accurately separate differences down to cultivar level. We also show that diurnal variation is measurable in the spectra but does not preclude differentiation. [ABSTRACT FROM AUTHOR]

Standard-essential patents (SEPs) are an important technological resource for firms in the telecommunication industry. The utilisation of technological topic analysis to reveal the global development dynamics of SEPs has significant theoretical and practical implications. First, this study defines the phrase extraction rules and constructs a phrase importance evaluation model to extract key technical phrases in the patent text. Second, the extracted key phrases are used as input for the Latent Dirichlet Allocation (LDA) model, and the relative independence (RI) model is proposed to determine the optimal number of topics based on two dimensions of coherence and similarity. Finally, the technological topic analysis based on the improved LDA model is performed on 30,154 texts of declared 5G SEPs. The results show that (1) the RI model can better identify the optimal number of topics for the LDA model; (2) 23 key technologies and four hot spots in 5G are identified based on the improved LDA model; (3) different firms have different technological layouts, and the diversification trend of technology development appears; and (4) the forecasting results also reveal the dynamics of emerging and declining technical areas in the 5G industry. [ABSTRACT FROM AUTHOR]

Background: Kratom is a substance that alters one's mental state and is used for pain relief, mood enhancement, and opioid withdrawal, despite potential health risks. In this study, we aim to analyze the social media discourse about kratom to provide more insights about kratom's benefits and adverse effects. Also, we aim to demonstrate how algorithmic machine learning approaches, qualitative methods, and data visualization techniques can complement each other to discern diverse reactions to kratom's effects, thereby complementing traditional quantitative and qualitative methods. Methods: Social media data were analyzed using the latent Dirichlet allocation (LDA) algorithm, PyLDAVis, and t-distributed stochastic neighbor embedding (t-SNE) technique to identify kratom's benefits and adverse effects. Results: The analysis showed that kratom aids in addiction recovery and managing opiate withdrawal, alleviates anxiety, depression, and chronic pain, enhances mood, energy, and overall mental well-being, and improves quality of life. Conversely, it may induce nausea, upset stomach, and constipation, elevate heart risks, affect respiratory function, and threaten liver health. Additional reported side effects include brain damage, weight loss, seizures, dry mouth, itchiness, and impacts on sexual function. Conclusion: This combined approach underscores its effectiveness in providing a comprehensive understanding of diverse reactions to kratom, complementing traditional research methodologies used to study kratom. [ABSTRACT FROM AUTHOR]

Copyright of China Pulp & Paper is the property of China Pulp & Paper Magazines Publisher and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Scientific knowledge of image-based crack detection methods is limited in understanding their performance across diverse crack sizes, types, and environmental conditions. Builders and engineers often face difficulties with image resolution, detecting fine cracks, and differentiating between structural and non-structural issues. Enhanced algorithms and analysis techniques are needed for more accurate assessments. Hence, this research aims to generate an intelligent scheme that can recognize the presence of cracks and visualize the percentage of cracks from an image along with an explanation. The proposed method fuses features from concrete surface images through a ResNet-50 convolutional neural network (CNN) and curvelet transform handcrafted (HC) method, optimized by linear discriminant analysis (LDA), and the eXtreme gradient boosting (XGB) classifier then uses these features to recognize cracks. This study evaluates several CNN models, including VGG-16, VGG-19, Inception-V3, and ResNet-50, and various HC techniques, such as wavelet transform, counterlet transform, and curvelet transform for feature extraction. Principal component analysis (PCA) and LDA are assessed for feature optimization. For classification, XGB, random forest (RF), adaptive boosting (AdaBoost), and category boosting (CatBoost) are tested. To isolate and quantify the crack region, this research combines image thresholding, morphological operations, and contour detection with the convex hulls method and forms a novel algorithm. Two explainable AI (XAI) tools, local interpretable model-agnostic explanations (LIMEs) and gradient-weighted class activation mapping++ (Grad-CAM++) are integrated with the proposed method to enhance result clarity. This research introduces a novel feature fusion approach that enhances crack detection accuracy and interpretability. The method demonstrates superior performance by achieving 99.93% and 99.69% accuracy on two existing datasets, outperforming state-of-the-art methods. Additionally, the development of an algorithm for isolating and quantifying crack regions represents a significant advancement in image processing for structural analysis. The proposed approach provides a robust and reliable tool for real-time crack detection and assessment in concrete structures, facilitating timely maintenance and improving structural safety. By offering detailed explanations of the model's decisions, the research addresses the critical need for transparency in AI applications, thus increasing trust and adoption in engineering practice. [ABSTRACT FROM AUTHOR]

Aspect-Based Sentiment Analysis (ABSA) is essential in industries such as healthcare, automotive, and finance, where understanding customer feedback is crucial for business strategies and product development. Traditional sentiment analysis methods often fail to account for specific aspects, providing only generalized sentiment scores (polarities). This study aims to develop a robust ABSA architecture to accurately classify aspect-based sentiment polarity. We propose a Modified Latent Dirichlet Allocation (M-LDA) model that integrates Term Frequency-Inverse Document Frequency (TF-IDF) and Latent Dirichlet Allocation (LDA) to extract relevant aspects. Additionally, an Optimized Bidirectional Encoder Representations from Transformers (O-BERT) model is used to determine aspect-based sentiment polarity, categorizing sentiments as positive, negative, or neutral. A Long Short-Term Memory (LSTM) network is then employed to enhance classification accuracy. The M-LDA O-BERTLSTM for ABSA (MOL-ABSA) performance, evaluated using accuracy and macro-F-score metrics on the SemEval 2014 and a generalized dataset, demonstrates its effectiveness, achieving accuracies of 81.19%, 85.26%, and 87.53%, respectively. This integrated approach combines traditional NLP techniques with advanced machine learning models to ensure high accuracy. [ABSTRACT FROM AUTHOR]

The study empirically analyzed activism participants' roles drawn from the lens of social media affordance and identified the activism opinion leaders based on the framework of network connectivity, message diffusion, and semantic relevancy through the case of the #Marchforourlives Twitter network, which has been rebranded as X. The study defines the #Marchforourlives Twitter network as a co-created activism network in collaboration with different degrees of contributors, such as the core advocates, the advocates, the supporters, and the amplifiers. The results showed that a very small number of tweets created by the core advocates played significant roles due to their extensive adoption by other participants, while many other original tweets were never mentioned or retweeted in the network. This study disclosed the extensive proportion of amplifiers as 95.13% among the examined participants. The study findings suggest that creating core agenda tweets with high amplifiability might be critical for successful hashtag activism to attract like-minded masses as networked protesters. [ABSTRACT FROM AUTHOR]

Abstract Hyperspectral data and machine learning offer great potential for identifying valuable open ecosystems. Due to the large volume of data, preprocessing of hyperspectral images must involve dimensionality reduction. The main goal of this study was to test the effectiveness of various types of feature reduction (feature selection and feature extraction) when performing classification using the Random Forest algorithm. A comparison was conducted between two ecosystems - heathlands and mires protected as Natura 2000 habitats. Two transformations of feature extraction were chosen, namely Minimum Noise Fraction (MNF) and Principal Component Analysis (PCA), while Linear Discriminant Analysis (LDA) was used as a feature selection method. It was proven that irrespective of the class type, accuracy is higher with the feature extraction method (mean F1 accuracy of 0.922) than with feature selection (mean F1 accuracy of 0.787). At the same time, no significant differences in accuracies were found between the MNF and PCA methods. Although LDA resulted in lower accuracies (0.816 for heathland and 0.750 for mires), the method could also be used due to relatively high F1 values. The effectiveness of the LDA method for feature reduction in open ecosystem identification was confirmed for the first time for open natural vegetation.

Financial distress is a socially and economically significant issue that affects almost every firm across the world. Predicting financial distress in the banking industry can substantially aid in the reduction of losses and can help avoid misallocation of banks' financial resources. Models for financial distress prediction of banks are being increasingly employed as important tools to identify early warning signals for the whole banking system. This study attempts to forecast the financial distress of commercial banks by developing a bankruptcy prediction model for banks. The sample size for the study is 75 Indian banks. Logistic, linear discriminant analysis (LDA) and artificial neural network (ANN) models have been applied on the last 5 years' (2015–2019) data of these banks. Data analysis results reveal the logistic and LDA models exhibiting similar prediction accuracy. The results of the ANN prediction model exhibit better prediction accuracy. It is expected that the results of this study will be useful for managers, depositors, regulatory bodies and shareholders to better manage their interests in the banking sector of the country. [ABSTRACT FROM AUTHOR]

Copyright of Austral Comunicacion is the property of Austral Comunicacion and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Abstract Text classification plays a major role in research such as sentiment analysis, opinion mining, and customer feedback analysis. Text classification using hypergraph algorithms is effective in capturing the intricate relationships between words and phrases in documents. The method entails text preprocessing, keyword extraction, feature selection, text classification, and performance metric evaluation. Here, we proposed a Hypergraph Attention Layer with Logistic Regression (HGATT_LR) for text classification in the Amazon review data set. The essential keywords are extracted by utilizing the Latent Dirichlet Allocation (LDA) technique. To build a hypergraph attention layer, feature selection based on node-level and edge-level attention is assessed. The resultant features are passed as an input of Logistic regression for text classification. Through a comparison analysis of different text classifiers on the Amazon data set, the performance metrics are assessed. Text classification using hypergraph Attention Network has been shown to achieve 88% accuracy which is better compared to other state-of-the-art algorithms. The proposed model is scalable and may be easily enhanced with more training data. The solution highlights the utility of hypergraph approaches for text classification as well as their applicability to real-world datasets with complicated interactions between text parts. This type of analysis will empower the business people will improve the quality of the product.

The study empirically analyzed activism participants’ roles drawn from the lens of social media affordance and identified the activism opinion leaders based on the framework of network connectivity, message diffusion, and semantic relevancy through the case of the #Marchforourlives Twitter network, which has been rebranded as X. The study defines the #Marchforourlives Twitter network as a co-created activism network in collaboration with different degrees of contributors, such as the core advocates, the advocates, the supporters, and the amplifiers. The results showed that a very small number of tweets created by the core advocates played significant roles due to their extensive adoption by other participants, while many other original tweets were never mentioned or retweeted in the network. This study disclosed the extensive proportion of amplifiers as 95.13% among the examined participants. The study findings suggest that creating core agenda tweets with high amplifiability might be critical for successful hashtag activism to attract like-minded masses as networked protesters.

More consumers are choosing pro-consumption for social change, but scholars know little about why and how consumers engage in pro-consumption behaviors. A newly emerged pro-consumption behavior called "Don-jjul,'' which appeared during the COVID-19 pandemic in South Korea, refers to compensating businesses that have engaged in altruistic actions by boosting their sales. This study used Latent Dirichlet Allocation (LDA) of topic modeling, sentiment analysis, and in-depth interviews to investigate the perceptions, motivations, and emotions regarding Don-jjul. As a result, the study revealed pro-consumers' perceptions of Don-jjul as "collective pro-consumption for contributing to social well-being.'' Don-jjul has two main motives: "supporting underdogs with difficulties'' and "compensating good businesses economically.'' We also found two ambivalent emotions evoked by Don-jjul: "respect for good business owners'' and "concerns regarding the misuse of Don-jjul.'' The results contribute to pro-consumption research for social well-being, providing business opportunities for retailers and CSR managers with a deep understanding of pro-consumers.

Keyword extraction and topic modeling in the analysis of Gojek user reviews in Indonesian are very important. By understanding user preferences and needs through keyword extraction, as well as grouping user reviews into different topics through topic modeling, stakeholders can use the information to further improve services. This research uses TF-IDF and LDA approaches to analyze text data from Gojek user reviews and feedback. The data spans from Nov 5, 2021, to Jan 2, 2024, totaling 225,002 rows. Each row includes username, content, time, and app version. The focus is on content reviews. The average length is 8 words, with a maximum of 104 and a minimum of a few words. The variability indicates a non-normal distribution. Preprocessing is conducted to maintain topic analysis accuracy. The TF-IDF method is used to extract relevant keywords, while the LDA approach is used to model the topics in user reviews. The topic analysis reveals patterns in Gojek user reviews. The first topic discusses experience, services, and affordable pricing. The second emphasizes app usability and benefits. The third relates to promos, discounts, and vouchers. The fourth reflects positive evaluations of service quality. However, the fifth topic highlights high costs and app issues. The sixth underscores overall user satisfaction and service convenience. Testing on the topic model yielded a coherence level of 0.509, indicating that the model's topics demonstrate a good level of consistency in finding relevant topics from Gojek user review data. The use of a combination of TF-IDF and LDA in Indonesian text analysis, particularly in the context of Gojek user reviews, is an important step in enhancing understanding and utilization of text data to improve overall user experience.

Digital technologies have been used for a vast amount of bibliometric analysis research. Although these technologies have made scientific investigation more accessible and efficient, scholars now face the daunting task of sifting through an overwhelming number of documents. This study aims to identify bibliometric research analysis's primary topics, categories, and latent topics from a global perspective. This study utilized topic modeling techniques to analyze the abstracts of 16,039 eligible papers published between 1977 and 2023 in the Scopus database. Through the use of Latent Dirichlet Allocation (LDA) topic modeling, the study was able to identify four distinct research topics and observe how they have evolved over time. The research topic has shifted its focus from individual concepts and words to relationships between nodes and conceptual, intellectual, and social structures. The study’s findings have significant implications for bibliometric analysis-related research, providing valuable insights into trends and patterns in bibliometric analysis content within large digital article archives. The LDA has proven to be an efficient tool for analyzing these trends and patterns quickly. This study's novel approach considers factors for word embedding usage and optimal topic numbers. It focuses on a full understanding of the LDA results and combines statistical analysis, domain knowledge, and temporal exploration to better understand how data structures work. Doi: 10.28991/HIJ-2024-05-02-07 Full Text: PDF

This paper presents a simulation of the structural and optoelectronic properties of Scandium Arsenide (ScAs) and Aluminum Arsenide (AlAs) compounds. Theoretical modeling was performed using ab initio first-principles calculations, specifically the density functional theory (DFT), and the Mindlab numerical software. The software used two methods: the full-potential muffin-tin orbital method (FP-LMTO) and the full-potential plane-wave method (FP-LAPW). These two methods are employed to solve the Schrödinger equation. The exchange correlation effects have been computed using two different approximations: the generalized gradient approximation (GGA) and the local density approximation (LDA). Our findings indicate that the zinc blende structure (B3) is the stable phase, while the Wurtzite phase (B4) is metastable for the AlAs compound. On the other hand, the ScAs compound crystallizes in the NaCl phase (B1). The AlAs compounds undergo three phase transitions: B3→B4, B3→B2 and B3→B1. In contrast, ScAs does not undergo any transition. The obtained results for equilibrium energies, lattice parameters and gap energies are in closer agreement with the experimental and theoretical data. The AlAs compound exhibits a semiconducting character, while ScAs exhibits a semi-metallic character. Additionally, the refractive index of these two compounds is similar to that of silicon, which is crucial for their application in photovoltaic cells. [ABSTRACT FROM AUTHOR]

Text classification plays a major role in research such as sentiment analysis, opinion mining, and customer feedback analysis. Text classification using hypergraph algorithms is effective in capturing the intricate relationships between words and phrases in documents. The method entails text preprocessing, keyword extraction, feature selection, text classification, and performance metric evaluation. Here, we proposed a Hypergraph Attention Layer with Logistic Regression (HGATT_LR) for text classification in the Amazon review data set. The essential keywords are extracted by utilizing the Latent Dirichlet Allocation (LDA) technique. To build a hypergraph attention layer, feature selection based on node-level and edge-level attention is assessed. The resultant features are passed as an input of Logistic regression for text classification. Through a comparison analysis of different text classifiers on the Amazon data set, the performance metrics are assessed. Text classification using hypergraph Attention Network has been shown to achieve 88% accuracy which is better compared to other state-of-the-art algorithms. The proposed model is scalable and may be easily enhanced with more training data. The solution highlights the utility of hypergraph approaches for text classification as well as their applicability to real-world datasets with complicated interactions between text parts. This type of analysis will empower the business people will improve the quality of the product. [ABSTRACT FROM AUTHOR]

The emergence of audio deepfakes has prompted concerns over reputational integrity and dependability. Deepfakes with audio can now be produced more easily, which makes it harder to spot them. Technologies that can identify audio-level deepfakes must be developed in order to address this issue. As a result, we have recognised the importance of feature extraction for these systems and we have created an improved method for feature extraction. On audio Mel spectrogram, we have employed a modified ResNet50 to extract features. Then, Linear Discriminant Analysis (LDA) dimensionality reduction technique have been used to optimise the feature complexity. The chosen features by LDA are then utilised to train these machine learning (ML) models using the backend classification algorithms Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbour (KNN), and Naive Bayes (NB). The ASVspoof 2019 Logical Access (LA) partition is utilised for training, ASVspoof 2021 deep fake partition are used to evaluate the systems. Also, we have used DECRO dataset for evakuating our proposed model under unseen noisy dataset. We have used 20% audios from training dataset for validation purpose. When compared to other models, our proposed method performs better than traditional feature extraction methods such as Mel Frequency Cepstral Coefficients (MFCC) and Gammatone Cepstral Coefficients (GTCC). It achieves an impressive Equal Error Rate (EER) of only 0.4% and an accuracy of 99.7%. [ABSTRACT FROM AUTHOR]

A speech signal can provide various information about a speaker, such as their gender, age, accent, and emotional state. The gender of the speaker is the most salient piece of information contained in the speech signal and is directly or indirectly used in many applications. In this study, a new approach is proposed for recognizing the gender of the speaker based on the use of hybrid features created by stacking different types of features. For this purpose, four different features, namely Mel frequency cepstral coefficients (MFCC), Mel scaled power spectrogram (Mel Spectrogram), Chroma, Spectral contrast (Contrast), and Tonal Centroid (Tonnetz), and twelve hybrid features created by stacking these features were used. These features were applied to four different classifiers, two of which were based on traditional machine learning (KNN and LDA) while two were based on the deep learning approach (CNN and MLP), and the performance of each was evaluated separately. In the experiments conducted on the Turkish subset of the Common Voice dataset, it was observed that hybrid features, created by stacking different acoustic features, led to improvements in gender recognition accuracy ranging from 0.3 to 1.73%. [ABSTRACT FROM AUTHOR]

Copyright of Techno.com is the property of Universitas Dian Nuswantoro, Fakultas Ilmu Komputer and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Cultural heritage is a fundamental part of the society's identity, and its conservation is of great relevance as it transcends time and memory. To minimize its deterioration, cultural heritage has traditionally undergone diverse preservation and maintenance treatments, and the attention of researchers to developing new and innovative methods for state diagnosis and protection treatments has been increasing in the recent decades. Despite extensive efforts in preservation, there remains a need for comprehensive and systematic mapping of scientific research to identify emerging trends and innovations in the field. To address this gap, in this study, a literature review using a bibliometric analysis and LDA methodology was conducted to systematically map scientific research outputs on cultural heritage conservation and protection. Data were retrieved from the Scopus database, and the annual publications, countries, most-cited publications, authors, institutions, and keywords have been comprehensively analyzed, leading to the detection of research trends and contributing to the existing knowledge in the field. The findings show an increasing number of studies in this field in the last decades, particularly since 2010. Italy, home to the largest number of UNESCO heritage sites, is the most prolific country. Most of the studies are related to metal, paper, and stone as substrates to be protected. Significant progress has been made in understanding the deterioration processes through precise diagnosis and the development of innovative treatments for protection. In this sense, the latest trends have been detected, such as the use of non-invasive techniques for diagnosis and the use of nanotechnology and nature-based treatments for conservation treatments. [ABSTRACT FROM AUTHOR]

Concentrations of metal(loid)s, Ag, Al, As, Cd, Co, Cr, Cu, Fe, Mn, Ni, Se, Sr, V and Zn, were determined in rice on sale in Las Vegas. The rice samples were grown in five different countries, the USA, Thailand, India, Pakistan, and Bangladesh. The elemental concentrations in rice grain were determined using inductively coupled plasma mass spectrometry (ICP-MS) following hot block-assisted digestion. The accuracy of the laboratory procedure was verified by the analysis of rice flour standard reference material (NIST SRM 1568b). The mean metal(loid) contents in rice of various geographic origins were 3.18–5.91 mg kg−1 for Al, 0.05–0.12 mg kg−1 for As, 3.64–41 μg kg−1 for Cd, 5.11–12 μg kg−1 for Co, 0.12–0.14 mg kg−1 for Cr, 1.5–1.91 mg kg−1 for Cu, 3.04–4.98 mg kg−1 for Fe, 4.2–10.4 mg kg−1 for Mn, 0.21–0.41 mg kg−1 for Ni, 0.02–0.07 mg kg−1 for Se, 0.68–0.88 mg kg−1 for Sr, 3.64–5.26 μg kg−1 for V, and 16.6–19.9 mg kg−1 for Zn. respectively. The mean concentration of As in US rice was significantly higher than in Indian, Pakistani, and Bangladeshi rice. On the other hand, it was found a significantly low mean level of Cd in US-grown rice. It was also found that the concentrations of metal(loid)s in black and brown rice on sale in Las Vegas were statistically similar, except for Mn and Se. The geographic origin traceability of rice grain involved the use of ICP-MS analysis coupled with chemometrics that allowed their differentiation based on the rice metal(loid) profile, thus confirming their origins. Data were processed by linear discriminant analysis, and US and Thai rice samples were cross-validated with higher accuracy (100%). This authentication quickly discriminates US rice from the other regions and adds verifiable food safety measures for consumers. [ABSTRACT FROM AUTHOR]

The environmental impact of the golf industry has garnered increasing attention, emphasizing the need for sustainable practices in golf course management. This study aims to develop strategic frameworks that enable the golf industry to address global environmental challenges and integrate eco-friendly principles. Utilizing topic modeling, the research examines key areas such as pesticide use, waste management, and energy consumption within golf courses, and explores strategies for achieving carbon neutrality and developing sustainable golf wear and branding. The findings reveal specific management practices that significantly reduce environmental harm, including integrated pest management, waste reduction techniques, and the adoption of renewable energy sources. Case studies of leading golf clubs demonstrate successful implementation of carbon neutrality strategies. Furthermore, the research highlights the potential of eco-friendly golf wear to enhance brand credibility and meet consumer demands. These insights hold valuable implications for golf course managers, club operators, policymakers, and golfers, promoting a comprehensive approach to sustainability in the golf industry. The study's contributions extend beyond the immediate environmental challenges, offering a framework for long-term sustainability and competitive advantage. [ABSTRACT FROM AUTHOR]

Mobile Money (MM) and Mobile Payments (MP) have become levers of growth. They drive economic, social, and Sustainable Development (SD). Despite the rich literature on MM and MP, we need more synthesis that cover their contributions as forms of Information and Communications Technologies for Development (ICT4D). We explored MM/MP as ICT4D through a systematic literature review coupled with a Latent Dirichlet Allocation (LDA) approach. The data analyses revealed that publications have increased significantly since 2018. Financial inclusion is the most studied area. It's studied for its contribution to SD. We also need to study these technologies’ effects on SD. This includes their effects on society and the environment. We propose a research agenda regarding these technologies at individual, organizational, country, regional, and global levels of analysis. It can help to find how these ICT4D improve people’s lives and foster sustainable development. [ABSTRACT FROM AUTHOR]

The nation's explosive growth in Massive Open Online Courses (MOOCs) is likely to lead to low effectiveness of MOOCs, therefore, it is necessary to promote the high-quality and long-term development of MOOCs through understanding learner satisfaction. The present research adopted the Latent Dirichlet Allocation (LDA) to derive factors affecting the satisfaction of the learners and conducted a fuzzy-set qualitative comparative analysis (QCA) to analyze configurations of the high-level and low high level of learning satisfaction with computer science MOOCs. This study identified how course design, teaching style, teaching content, communication, exercise, and platform impact learner satisfaction based on 27,316 students' reviews of MOOCs. It has been found that five sufficient configurations are demonstrated to achieve high satisfaction while the other three configurations do not result in high satisfaction. In addition, although the six factors cannot alone constitute the necessary conditions for configurations of learner satisfaction, course design, teaching styles, and exercise play an indispensable role in improving learner satisfaction. Overall, this study adds to the literature by examining specific learner-level factors and extending configural understanding of the factors that affect MOOC learner satisfaction. [ABSTRACT FROM AUTHOR]

China proposed the Belt and Road Initiative to strengthen regional connectivity so as to embrace a brighter future together. Since the Initiative was put forward, it has brought many challenges to China's English education policy. By employing Latent Dirichlet Allocation (LDA) and Word2Vec, this study analyzes the evolution of topics and challenges in China's English education policy under the Belt and Road Initiative. The results indicate that after the initiative, the policy focus has changed. English education has shifted from testing abilities to cultivating students' intercultural communication skills in order to meet the needs with countries alongside the "Belt and Road". Moreover, teaching strategies that were examination-oriented have also changed to emphasizing teaching methods and feedback. The focus and teaching strategies have also undergone great changes. China's English education policy has shifted from focusing on improving students' writing skills, English proficiency, and creativity to conducting in-depth research and addressing specific issues, including challenges in linguistics, media influence, educational institutions and programs, online courses, attitudes and self-efficacy, use of multiple languages and globalization, teaching issues, and curriculum design. These findings shed light on how the Belt and Road Initiative changed China's English education policy and provide further directions for future research. [ABSTRACT FROM AUTHOR]

The crucial part of the software development cycle is software maintenance. The demands included in the software management are fault fixes and request to change or bring a new feature. If priority is not given to these demands, then it may lead to customer dissatisfaction, inefficient planning, and software failure as well. Therefore, it is important to study the severity of the bug reports to maintain the efficiency of the software. Various research has been conducted in the past to predict the severity of the paper using text mining focusing only on the content of the bug reports. The sentiment of the user while reporting a bug also plays a vital role. In this study, we will be focusing on two aspects, that is, sentiment and content to improve the prediction. We propose a prediction model based on LDA to study the content aspect and emotion analysis to study the sentiment aspect. The model is validated on the datasets collected from the Eclipse project using Convolutional Neural Network (CNN). The results show that the CNN model effectively utilizes the content and sentiment aspect of the data to handle the severity prediction. CNN has weight sharing feature that decreases the number of parameters used for training. It also improves generalization and overfitting is avoided The Accuracy, Precision, Recall, and F‐measure are improved when both aspects are taken into account rather than considering only content. [ABSTRACT FROM AUTHOR]

Supplemental Digital Content is Available in the Text. Understanding, measuring, and mitigating pain-related suffering is a key challenge for both clinical care and pain research. However, there is no consensus on what exactly the concept of pain-related suffering includes, and it is often not precisely operationalized in empirical studies. Here, we (1) systematically review the conceptualization of pain-related suffering in the existing literature, (2) develop a definition and a conceptual framework, and (3) use machine learning to cross-validate the results. We identified 111 articles in a systematic search of Web of Science, PubMed, PsychINFO, and PhilPapers for peer-reviewed articles containing conceptual contributions about the experience of pain-related suffering. We developed a new procedure for extracting and synthesizing study information based on the cross-validation of qualitative analysis with an artificial intelligence–based approach grounded in large language models and topic modeling. We derived a definition from the literature that is representative of current theoretical views and describes pain-related suffering as a severely negative, complex, and dynamic experience in response to a perceived threat to an individual's integrity as a self and identity as a person. We also offer a conceptual framework of pain-related suffering distinguishing 8 dimensions: social, physical, personal, spiritual, existential, cultural, cognitive, and affective. Our data show that pain-related suffering is a multidimensional phenomenon that is closely related to but distinct from pain itself. The present analysis provides a roadmap for further theoretical and empirical development. [ABSTRACT FROM AUTHOR]

Online product reviews (OPR) are a commonly used medium for consumers to communicate their experiences with products during online shopping. Previous studies have investigated the helpfulness of OPRs using frequency-based, linguistic, meta-data, readability, and reviewer attributes. In this study, we explored the impact of robust contextual word embeddings, topic, and language models in predicting the helpfulness of OPRs. In addition, the wrapper-based feature selection technique is employed to select effective subsets from each type of features. Five feature generation techniques including word2vec, FastText, Global Vectors for Word Representation (GloVe), Latent Dirichlet Allocation (LDA), and Embeddings from Language Models (ELMo), were employed. The proposed framework is evaluated on two Amazon datasets (Video games and Health & personal care). The results showed that the ELMo model outperformed the six standard baselines, including the fine-tuned Bidirectional Encoder Representations from Transformers (BERT) model. In addition, ELMo achieved Mean Square Error (MSE) of 0.0887 and 0.0786 respectively on two datasets and MSE of 0.0791 and 0.0708 with the wrapper method. This results in the reduction of 1.43% and 1.63% in MSE as compared to the fine-tuned BERT model on respective datasets. However, the LDA model has a comparable performance with the fine-tuned BERT model but outperforms the other five baselines. The proposed framework demonstrated good generalization abilities by uncovering important factors of product reviews and can be evaluated on other voting platforms. [ABSTRACT FROM AUTHOR]

The study utilized Latent Dirichlet Allocation (LDA) Topic modeling to identify prevalent latent topics within Open Access (OA) Library and Information Science (LIS) journals from 2013 to 2022. Eight core OA Scopus indexed journals were selected based on their SJR scores and DOAJ listing. Titles, Abstracts and keywords of 2589 articles were extracted from the Scopus database. R software packages were used to perform data analysis and LDA topic modeling. The optimal value of k was determined to be 9. The analysis revealed that 53.89% of documents comprise over 50% of a certain topic (θ > 0.50). Notably, 'Scholarly Communication' and 'Information Systems, Models and Frameworks' emerged as dominant topics with the highest proportions of research literature in the corpus. The topic 'Scholarly Communication' experienced significant growth with an average annual growth rate (AAGR) of 4.37%, while 'Collection development and E-resources' exhibited the lowest research proportion and a negative AAGR of − 4.22%. Additionally, topics such as 'Information Seeking Behaviour and User Studies', 'User Education and Information Literacy', and 'Information Retrieval and Systematic Review' remained stable and persistent topics. Conversely, research on traditional topics like 'Librarianship and Profession', 'Bibliometrics' and 'Medical Library and Health Information' showed a gradual decline. The LDA topic modeling approach unveiled previously unknown or unexplored topics in open access LIS research literature, enhancing our understanding of emerging trends. [ABSTRACT FROM AUTHOR]

Tourist's cognitive image network unveils the relationship among the touring cognitive experiences. The cognitive image network evolution reflects the tourists' destination image change, which further implicates the subtle changes in behavior. The tourists' online reviews and experience sharing could effectively help management build the cognitive image network of tourism destinations. Based on crawled travelogues in Guizhou, China from Ctrip.com, with LDA topic mining techniques, this study first constructs the image of a tourism destination, then constructs the tourist's cognitive image network that explores the evolution of the network before and after the COVID-19 pandemic. Research results show that the cognitive image network in Guizhou mainly consists of 10 dimensions: Accommodation, diet, natural landscapes, social environment, and culture dimensions, followed by entertainment, natural environment, history, infrastructure, and economic environment. The cognitive image network evolution shows that tourists have different preferences before and after the COVID-19 pandemic. The study provides a new perspective for the study of destination image theory, which can help destination tourism enterprises to identify the tourist's behavior choices and develop unique marketing strategies. [ABSTRACT FROM AUTHOR]