Search

We apply a dynamic influence model to the opinions of the US federal courts to examine the role of the US Supreme Court in influencing the direction of legal discourse in the federal courts. We propose two mechanisms for how the Court affects innovation in legal language: a selection mechanism where the Court's influence primarily derives from its discretionary jurisdiction, and an authorship mechanism in which the Court's influence derives directly from its own innovations. To test these alternative hypotheses, we develop a novel influence measure based on a dynamic topic model that separates the Court's own language innovations from those of the lower courts. Applying this measure to the US federal courts, we find that the Supreme Court primarily exercises influence through the selection mechanism, with modest additional influence attributable to the authorship mechanism. This article is part of the theme issue 'A complexity science approach to law and governance'. [ABSTRACT FROM AUTHOR]

As one of the oldest continuously publishing journals in statistics (published since 1901), Biometrika provides a unique window onto the history of statistics and its epistemic development throughout the 20th and the beginning of the 21st centuries. While the early history of the discipline, with the works of key figures, such as Karl Pearson, Francis Galton, or Ronald Fisher, is relatively well known, the later (and longer) episodes of its intellectual development remain understudied. By applying digital tools to the full-text corpus of the journal articles (N  =   5,596), the objective of this study is to provide a novel quantitative exploration of the history of the statistical sciences via an all-encompassing view of 120 years of Biometrika. To this aim, topic-modelling analyses are used and provide insights into the epistemic content of the journal and its evolution. Striking changes in the thematic content of the journal are documented and quantified for the first time, from the decline of Pearsonian and Weldonian biometrical research and the journal's tight connection to biology in the 1930s to the rise of modern statistical methods beginning in the 1960s and 1970s. Newly developed approaches are used to infer author networks from publication topics. The resulting network of authors shows the existence of several communities, well-aligned with topic clusters and their evolution through time. It also highlights the role of specific figures over more than a century of publishing history and provides a first window onto the foundation, development, and diverse applications of the statistical sciences. [ABSTRACT FROM AUTHOR]

Between 2015 and 2017, the Internet Research Agency (IRA) – a Kremlin-backed "troll farm" based in St. Petersburg – executed a propaganda campaign on Twitter to target US voters. Scholarship has expended relatively little effort to study the role of Islamophobia in the IRA's propaganda campaign. Following critical disinformation research, this article demonstrates that Islamophobia, affect, and white identity played a crucial role in the IRA's targeting of right-wing US voters. With an official release of tweets and associated visual content from Twitter, we use topic modeling and visual analysis to explore both how, and to what extent, the IRA used Islamophobia in its propaganda. To do so, we develop a multimodal distant reading technique to study how the IRA aligned users with contemporary far right social movements by deploying racial and emotional appeals that center on narrating a transnational white identity under threat from Islam and Muslims. [ABSTRACT FROM AUTHOR]

As an extension to the standard symmetric latent Dirichlet allocation topic model, we implement asymmetric Beta-Liouville as a conjugate prior to the multinomial and therefore propose the maximum a posteriori for latent Beta-Liouville allocation as an alternative to maximum likelihood estimator for models such as probabilistic latent semantic indexing, unigrams, and mixture of unigrams. Since most Bayesian posteriors, for complex models, are intractable in general, we propose a point estimate (the mode) that offers a much tractable solution. The maximum a posteriori hypotheses using point estimates are much easier than full Bayesian analysis that integrates over the entire parameter space. We show that the proposed maximum a posteriori reduces the three-level hierarchical latent Beta-Liouville allocation to two-level topic mixture as we marginalize out the latent variables. In each document, the maximum a posteriori provides a soft assignment and constructs dense expectation–maximization probabilities over each word (responsibilities) for accurate estimates. For simplicity, we present a stochastic at word-level online expectation–maximization algorithm as an optimization method for maximum a posteriori latent Beta-Liouville allocation estimation whose unnormalized reparameterization is equivalent to a stochastic collapsed variational Bayes. This implicit connection between the collapsed space and expectation–maximization-based maximum a posteriori latent Beta-Liouville allocation shows its flexibility and helps in providing alternative to model selection. We characterize efficiency in the proposed approach for its ability to simultaneously stream both large-scale data and parameters seamlessly. The performance of the model using predictive perplexities as evaluation method shows the robustness of the proposed technique with text document datasets. [ABSTRACT FROM AUTHOR]

To address the limitations of existing methods of short-text entity disambiguation, specifically in terms of their insufficient feature extraction and reliance on massive training samples, we propose an entity disambiguation model called COLBERT, which fuses LDA-based topic features and BERT-based semantic features, as well as using contrastive learning, to enhance the disambiguation process. Experiments on a publicly available Chinese short-text entity disambiguation dataset show that the proposed model achieves an F1-score of 84.0%, which outperforms the benchmark method by 0.6%. Moreover, our model achieves an F1-score of 74.5% with a limited number of training samples, which is 2.8% higher than the benchmark method. These results demonstrate that our model achieves better effectiveness and robustness and can reduce the burden of data annotation as well as training costs. [ABSTRACT FROM AUTHOR]

Topic models have been widely used by researchers across disciplines to automatically analyze large textual data. However, they often fail to automate content analysis, because the algorithms cannot accurately classify individual sentences into pre-defined topics. Aiming to make topic classification more theoretically grounded and content analysis in general more topic-specific, we have developed Seeded Sequential Latent Dirichlet allocation (LDA), extending the existing LDA algorithm, and implementing it in a widely accessible open-source package. Taking a large corpus of speeches delivered by delegates at the United Nations General Assembly as an example, we explain how our algorithm differs from the original algorithm; why it can classify sentences more accurately; how it accepts pre-defined topics in deductive or semi-deductive analysis; how such ex-ante topic mapping differs from ex-post topic mapping; how it enables topic-specific framing analysis in applied research. We also offer practical guidance on how to determine the optimal number of topics and select seed words for the algorithm. [ABSTRACT FROM AUTHOR]

We review popular unsupervised learning methods for the analysis of high‐dimensional data encountered in, for example, genomics, medical imaging, cohort studies, and biobanks. We show that four commonly used methods, principal component analysis, K‐means clustering, nonnegative matrix factorization, and latent Dirichlet allocation, can be written as probabilistic models underpinned by a low‐rank matrix factorization. In addition to highlighting their similarities, this formulation clarifies the various assumptions and restrictions of each approach, which eases identifying the appropriate method for specific applications for applied medical researchers. We also touch upon the most important aspects of inference and model selection for the application of these methods to health data. [ABSTRACT FROM AUTHOR]

With the rapid development of information science and social networks, the Internet has accumulated various data containing valuable information and topics. The topic model has become one of the primary semantic modeling and classification methods. It has been widely studied in academia and industry. However, most topic models only focus on long texts and often suffer from semantic sparsity problems. The sparse, short text content and irregular data have brought major challenges to the application of topic models in semantic modeling and topic discovery. To overcome these challenges, researchers have explored topic models and achieved excellent results. However, most of the current topic models are applicable to a specific model task. The majority of current reviews ignore the whole-cycle perspective and framework. It brings great challenges for novices to learn topic models. To deal with the above challenges, we investigate more than a hundred papers on topic models and summarize the research progress on the entire topic model process, including theory, method, datasets, and evaluation indicator. In addition, we also analyzed the statistical data results of the topic model through experiments and introduced its applications in different fields. The paper provides a whole-cycle learning path for novices. It encourages researchers to give more attention to the topic model algorithm and the theory itself without paying extra attention to understanding the relevant datasets, evaluation methods and latest progress. [ABSTRACT FROM AUTHOR]

The article focuses on addressing a critical limitation in recommender systems based on knowledge graphs (KGs) by introducing a new model called Topic-aware Intention Network (TIN) to explicitly model user intent for improved recommendations. Topics covered include the shortcomings of existing KG-based recommender systems, the proposed TIN model's architecture, and the benefits of incorporating user intent into recommendations.

Topic modeling is a popular method used to describe biological count data. With topic models, the user must specify the number of topics |$K$|⁠. Since there is no definitive way to choose |$K$| and since a true value might not exist, we develop a method, which we call topic alignment , to study the relationships across models with different |$K$|⁠. In addition, we present three diagnostics based on the alignment. These techniques can show how many topics are consistently present across different models, if a topic is only transiently present, or if a topic splits into more topics when |$K$| increases. This strategy gives more insight into the process of generating the data than choosing a single value of |$K$| would. We design a visual representation of these cross-model relationships, show the effectiveness of these tools for interpreting the topics on simulated and real data, and release an accompanying R package, alto [ABSTRACT FROM AUTHOR]

Schools develop mission statements in part to communicate their purposes of schooling to internal and external audiences. The goal of this study is to employ text analytic techniques to analyze school mission statements. Focusing on Arizona and New Jersey's schools, we seek to understand: 1) what themes (topics) emerge from their mission statements, 2) whether local context (politics) and institutional factors (being a charter) mediate topic prevalence, and 3) how similar school mission statements are to one another overall and across select factors. Using structural topic modeling we find that five topics emerge as particularly prevalent, and they address key purposes within the US context related to social, academic, and civic development with an emphasis on a safe environment and a sense of community. Local political leanings have a minor impact on the prevalence of topics but being a charter has a distinct influence. Compared to regular schools, charters are more likely to include themes related to preparing for college, academic approaches, and character and family ties, but they place lower emphasis on community and citizenship themes. School missions display moderate levels of similarity, indicating that a common set vocabulary has permeated schools' missions, though differences remain, including between states. Overall, this study deepens our understanding of the varieties of schools' missions and reveals text analytic techniques are well suited for uncovering them. Plain Language Summary: Schools develop mission statements in part to communicate their purposes of schooling to internal and external audiences. In a novel application, the goal of this study is to employ text analytic techniques to analyze school mission statements. Focusing on Arizona and New Jersey schools, we seek to understand what themes (topics) emerge from their mission statements, whether local context (politics) and institutional factors (being a charter) mediate topic prevalence, and how similar school mission statements are to one another overall and across select factors. Employing structural topic modeling, five themes emerge as most prevalent. The themes broadly address key purposes within the US context related to social, academic, and civic development with an emphasis on a safe environment and a sense of community. Local political leanings have a minor impact on the prevalence of topics but being a charter has a distinct influence. Compared to regular schools, charters are more likely to include themes related to preparing for college, academic approaches, and character and family ties, but they place lower emphasis on community and citizenship themes. School missions display moderate levels of similarity, indicating that a common set vocabulary has permeated, though differences remain, including between states. One limitation of the study is the focus on two states; however, preliminary analyses of a small national sample suggest comparable vocabularies. Overall, this study reveals school mission statements present meaningful variations in the purposes of schools. Schools are communicating interpretable goals and text data mining techniques are well suited for uncovering them. [ABSTRACT FROM AUTHOR]

[Purpose/Significance] The introduction of national poverty alleviation policies and rural revitalization strategies has thrust the issue of education for left-behind children into the spotlight of scholarly attention. Education, far beyond serving as a mere instrument for personal growth and human capital accumulation for left-behind children, emerges as a pivotal measure in consolidating rural poverty alleviation endeavors and breaking the transmission of intergenerational poverty in China. It stands as a vital force propelling the future of rural revitalization. Yet, the existing literature on the education of left-behind children remains sporadic and dispersed. A more profound organizational effort, integrating, synthesizing, and evaluating this scattered literature, is imperative to establish a foundational framework for future research, fostering more cohesive and focused research endeavors. Presently, literature review studies primarily fall into three categories: qualitative review methods, meta-analysis, and bibliometric analysis methods employing tools like Citespace. This study sets out to achieve a systematic and comprehensive understanding of education-related issues for rural left-behind children through text mining methods grounded in topic models. [Method/Process] The advent of artificial intelligence and machine learning technologies has empowered the processing and analysis of vast amounts of textual data. Previous research, employing latent dirichlet allocation (LDA) topic models, successfully mined texts related to teacher team construction reform policies, internationalization in higher education literature, news reports, and online comments. In this study, a corpus was meticulously constructed using abstract texts extracted from 2037 journal articles published between 2002 and 2023. The structural topic model (STM) was chosen for topic modeling, overcoming the limitations associated with LDA, with a specific emphasis on exploring the diversity and dynamism of topics within the existing literature. [Results/Conclusions] The culmination of this research effort identified eight distinct research themes: psychological well-being, factors leading to left-behind children, macro-level coping strategies, types of guardianship, review studies, family education, media literacy, and micro-level coping strategies. By synergizing document metadata information, the study systematically unraveled the evolving trends of these topics over time, providing crucial insights into potential shifts in the focus of left-behind children's education research. It is essential to note that this study, while collecting abstracts instead of full texts, may not capture the entirety of information contained in complete research articles. Future research endeavors should explore left-behind children's education more comprehensively, leveraging full-text mining techniques for a more nuanced understanding of this critical subject. [ABSTRACT FROM AUTHOR]

Simple Summary: Single-cell RNA sequencing has significantly contributed to the discovery of heterogenous cellular programs that are highly expressed by distinct cell subtypes. Yet, the quantification of cell subtype-specific and shared gene co-expressing modules (GEMs) associated with cell differentiation is a challenge. Herein, we have developed scGEM to uncover such hidden GEMs and conducted a systematic evaluation of model performance as well as a comparison with existing methods. We demonstrate that scGEM has the potential to generate a better biological explanation of GEMs using simulated and real-world datasets. The positive impacts of this study can illuminate the interpretation of gene modules in single-cell transcriptome analysis and shed light on the cell–cell communication and regulatory network. Background: Single-cell transcriptome analysis has fundamentally changed biological research by allowing higher-resolution computational analysis of individual cells and subsets of cell types. However, few methods have met the need to recognize and quantify the underlying cellular programs that determine the specialization and differentiation of the cell types. Methods: In this study, we present scGEM, a nested tree-structured nonparametric Bayesian model, to reveal the gene co-expression modules (GEMs) reflecting transcriptome processes in single cells. Results: We show that scGEM can discover shared and specialized transcriptome signals across different cell types using peripheral blood mononuclear single cells and early brain development single cells. scGEM outperformed other methods in perplexity and topic coherence (p < 0.001) on our simulation data. Larger datasets, deeper trees and pre-trained models are shown to be positively associated with better scGEM performance. The GEMs obtained from triple-negative breast cancer single cells exhibited better correlations with lymphocyte infiltration (p = 0.009) and the cell cycle (p < 0.001) than other methods in additional validation on the bulk RNAseq dataset. Conclusions: Altogether, we demonstrate that scGEM can be used to model the hidden cellular functions of single cells, thereby unveiling the specialization and generalization of transcriptomic programs across different types of cells. [ABSTRACT FROM AUTHOR]

Background: Topic models are a class of unsupervised machine learning models, which facilitate summarization, browsing and retrieval from large unstructured document collections. This study reviews several methods for assessing the quality of unsupervised topic models estimated using non-negative matrix factorization. Techniques for topic model validation have been developed across disparate fields. We synthesize this literature, discuss the advantages and disadvantages of different techniques for topic model validation, and illustrate their usefulness for guiding model selection on a large clinical text corpus. Design, setting and data: Using a retrospective cohort design, we curated a text corpus containing 382,666 clinical notes collected between 01/01/2017 through 12/31/2020 from primary care electronic medical records in Toronto Canada. Methods: Several topic model quality metrics have been proposed to assess different aspects of model fit. We explored the following metrics: reconstruction error, topic coherence, rank biased overlap, Kendall's weighted tau, partition coefficient, partition entropy and the Xie-Beni statistic. Depending on context, cross-validation and/or bootstrap stability analysis were used to estimate these metrics on our corpus. Results: Cross-validated reconstruction error favored large topic models (K ≥ 100 topics) on our corpus. Stability analysis using topic coherence and the Xie-Beni statistic also favored large models (K = 100 topics). Rank biased overlap and Kendall's weighted tau favored small models (K = 5 topics). Few model evaluation metrics suggested mid-sized topic models (25 ≤ K ≤ 75) as being optimal. However, human judgement suggested that mid-sized topic models produced expressive low-dimensional summarizations of the corpus. Conclusions: Topic model quality indices are transparent quantitative tools for guiding model selection and evaluation. Our empirical illustration demonstrated that different topic model quality indices favor models of different complexity; and may not select models aligning with human judgment. This suggests that different metrics capture different aspects of model goodness of fit. A combination of topic model quality indices, coupled with human validation, may be useful in appraising unsupervised topic models. [ABSTRACT FROM AUTHOR]

Location-based social networks (LBSNs) have become a popular platform for users to share their activities with friends and families, which provide abundant information for us to study issues of group venue recommendation by utilizing the characteristics of check-in data. Although there are some studies on group recommendation for venues, few studies consider the group’s venue preference in different temporal patterns. In this article, we discover that the group’s activity venue has a temporal effect, that is, the group’s preference for the activity venue is different at different times. For example, a couple of lovers prefer to travel to tropical regions in winter and relax in bars in the evening. Based on this discovery, we present a Time-aware Multi-pattern (TaMp) topic model to capture the group’s interest in the activity venue in multiple temporal patterns (including the daily pattern, the weekly pattern, the monthly pattern, and the quarterly pattern). The TaMp model takes into account the topic, members, temporality, and venue information of group activities and the latent relations among them, especially the strong correlation between the activity time and the corresponding activity venue. Then, we propose a group venue recommendation method based on the TaMp model. In addition, an improved grouping algorithm (iGA) in LBSNs is put forward to enhance the rationality of grouping and the accuracy of group venue recommendation. We conduct comprehensive experiments to evaluate the performance of TaMp on two real-world datasets. The results show that our proposed method outperforms the state-of-the-art group venue recommendation and demonstrates the significance of temporal effects in explaining group activities. [ABSTRACT FROM AUTHOR]

Dedicated applications(apps) for child-rearing question and answer(QA) system have become common tools widely used by many parents in Japan. A user can post his/her various questions about child-rearing on the systems and exchange information with other parents online. Since the value of such dedicated applications increases through being used by a large number of users, it is considered important to continuously provide attractive functions to increase user satisfaction. Here, concerns of parents should be changed depending on their children’s life stages. In addition, it can be assumed that users always have concerns of child-rearing even when they do not post any questions on the application. Therefore, analysis and comprehension of the problems of parents rearing small children by taking into account these properties will lead to the improving the users’ satisfaction of QA services. For instance, the app can provide users with proper information with appropriate timing based on the analysis result. Although topic models are widely known to analyze the content of text data, this is not a sufficient method for visualizing changes in topics over time from question text data, which are relatively short sentences posted on dedicated apps for child-rearing QA systems. In this study, the authors propose an analytical model that can extract the topics of users’ concerns from question data and compile topic transitions throughout the various stages of child-rearing. Furthermore, by taking a probabilistic view of the topic transitions, a method for estimating the topic that is most likely to shift from a previous topic at a given time to the next non-question period is constructed. Finally, the authors show the results of applying the proposed method to an actual data set and clarify the usefulness of the estimation result using the proposed method. The proposed method makes it possible to analyze how the worries and concerns of parents raising children evolve over time, and to gain various new insights into the perspectives of support for these users through apps. [ABSTRACT FROM AUTHOR]

Purpose: The purpose of this study is to comprehensively review the human resource management (HRM) and employment relations (ERs) field and explore the knowledge map, knowledge evolution trends and paths and paradigm shifts within this field. Design/methodology/approach: The Structural Topic Model in combination with Word2vec is proposed and applied in this work. First, this paper detects and interprets the research topics by reviewing 23,786 papers from 29 important journals in this field from 1990 to 2021. Then, this research explores popularity trends by aggregating topic proportions from a temporal perspective. Finally, this work explores the research topic evolution from the semantic perspective. Findings: This paper obtains the following findings: (1) Sixteen research topics are identified, which provide the basic research overview of the whole field. (2) The changes in topic popularity over time map the tendency for employee benefits to be valued. (3) The evolutionary trajectories of temporal local topics are provided, which reflect the mechanisms of the paradigm and ideological migration and fusion. Originality/value: This work adopts state-of-the-art textual as well as semantic mining techniques to establish a comprehensive knowledge map for HRM and ER research. Furthermore, these results uniquely demonstrate the pluralistic ideological orientation at the social level is gradually integrated into more micro levels, such as enterprises and individuals. These are the contents that were mentioned from previous studies by scholars, but not meticulously verified and interpreted. [ABSTRACT FROM AUTHOR]

Multidimensional scoring evaluates each constructed‐response answer from more than one rating dimension and/or trait such as lexicon, organization, and supporting ideas instead of only one holistic score, to help students distinguish between various dimensions of writing quality. In this work, we present a bilevel learning model for combining two objectives, the multidimensional automated scoring, and the students' writing structure analysis and interpretation. The dual objectives are enabled by a supervised model, called Latent Dirichlet Allocation Multitask Learning (LDAMTL), integrating a topic model and a multitask learning model with an attention mechanism. Two empirical data sets were employed to indicate LDAMTL model performance. On one hand, results suggested that LDAMTL owns better scoring and QW‐κ values than two other competitor models, the supervised latent Dirichlet allocation, and Bidirectional Encoder Representations from Transformers at the 5% significance level. On the other hand, extracted topic structures revealed that students with a higher language score tended to employ more compelling words to support the argument in their answers. This study suggested that LDAMTL not only demonstrates the model performance by conjugating the underlying shared representation of each topic and learned representation from the neural networks but also helps understand students' writing. [ABSTRACT FROM AUTHOR]

Depression raises a double challenge: besides the negative mood and the intrusive thoughts, the relation to the self also becomes difficult. Online forums are analysed as communicative platforms enabling the interactive reconstruction of the self. The discourses of online depression forums are explored. Firstly, narrative patterns are identified according to their thematic focus (e.g. dysfunctional body, challenges of intimacy) and discursive logic (e.g. information exchange, support). Secondly, narratives are analysed in order to describe various ways of grounding a depressed self. ∼70.000 depression-related posts from the biggest English-speaking online forums (e.g. , ) were analysed. Quantitative (LDA topic modelling) and qualitative (deep reading) approaches were used simultaneously to determine the optimal number of topics and their interpretation. 13 topics were identified and interpreted according to their content and communicative function. Based on the inter-topic distances four clusters were identified (medicalized, intimacy-oriented, critical and uninhabitable self-narratives). The clusters of the 13 topics highlight various ways of narrating depression and the depressed self. Based on a comparison with a systematic review of mental illness recovery narratives, depression forums cover most narrative genres and emotional tones, thus create a unique opportunity for integrating the depressing experiences in the self. [ABSTRACT FROM AUTHOR]

Traditional text keyword extraction methods ignore the contextual semantic information and cannot solve the problem of ambiguity of a word, so the extraction effect is not ideal. Based on the LDA and BERT models, this study proposes the LDA-BERT-LightGBM (LB-LightGBM) model. The LDA topic model is selected to obtain the topic of each review and its word distribution, candidate keywords are filtered out according to the threshold, and the filtered words and the original review text are spliced and input into the BERT model. The word vector training is performed to obtain the word vector containing the text topic, so the text keyword extraction problem is converted into a two-classification problem through the LightGBM algorithm. The textrank algorithm, LDA algorithm, LightGBM algorithm and the proposed LB-Light GBM model are compared through experiments on the accuracy rate P, recall rate R and F1 of text keyword extraction in the present study. The results show that when TopN takes 3～6, the average value of F1 is 3.5% higher than that of the optimal method, indicating that the extraction effect of this method is generally better than that of the comparison method selected in the experiment, and the text keywords can be found more accurately. [ABSTRACT FROM AUTHOR]

In international politics, is environmental protection largely a "rich-country" priority? We perceive four reasons why, although individual exceptions are possible, the answer would be yes: as a country meets more of its basic economic needs, it can better take on environmental policy's long-term thinking, policy expenses, collective action problems, and quality-of-life issues. To cut through lip service paid by governments that are not serious about environmental protection, and the fact that the "BASIC" countries (Brazil, South Africa, India, China) occupy a gray area between rich and poor, we employ computer-assisted textual analyses on all 3,774 paragraphs of statements made by national governments between 1995–2012 in the Committee on Trade and Environment (CTE) within the World Trade Organization (WTO). Controlling for other factors, we find a general pattern of environmental discussions increasing as development level increases. This contributes substantively and methodologically to the literatures on the environment/development nexus and rising powers. [ABSTRACT FROM AUTHOR]

Background: Studies of new and expecting parents largely focus on the mother, leaving a gap in knowledge about fathers. Objective: This study aimed to understand web-based conversations regarding new and expecting fathers on social media and to explore whether the COVID-19 pandemic has changed the web-based conversation. Methods: A social media analysis was conducted. Brandwatch (Cision) captured social posts related to new and expecting fathers between February 1, 2019, and February 12, 2021. Overall, 2 periods were studied: 1 year before and 1 year during the pandemic. SAS Text Miner analyzed the data and produced 47% (9/19) of the topics in the first period and 53% (10/19) of the topics in the second period. The 19 topics were organized into 6 broad themes. Results: Overall, 26% (5/19) of the topics obtained during each period were the same, showing consistency in conversation. In total, 6 broad themes were created: fatherhood thoughts, fatherhood celebrations, advice seeking, fatherhood announcements, external parties targeting fathers, and miscellaneous. Conclusions: Fathers use social media to make announcements, celebrate fatherhood, seek advice, and interact with other fathers. Others used social media to advertise baby products and promote baby-related resources for fathers. Overall, the arrival of the COVID-19 pandemic appeared to have little impact on the excitement and resiliency of new fathers as they transition to parenthood. Altogether, these findings provide insight and guidance on the ways in which public health professionals can rapidly gather information about special populations--such as new and expecting fathers via the web--to monitor their beliefs, attitudes, emotional reactions, and unique lived experiences in context (ie, throughout a global pandemic). [ABSTRACT FROM AUTHOR]

Every epidemic affects the real lives of many people around the world and leads to terrible consequences. Recently, many tweets about the COVID-19 pandemic have been shared publicly on social media platforms. The analysis of these tweets is helpful for emergency response organizations to prioritize their tasks and make better decisions. However, most of these tweets are non-informative, which is a challenge for establishing an automated system to detect useful information in social media. Furthermore, existing methods ignore unlabeled data and topic background knowledge, which can provide additional semantic information. In this paper, we propose a novel Topic-Aware BERT (TABERT) model to solve the above challenges. TABERT first leverages a topic model to extract the latent topics of tweets. Secondly, a flexible framework is used to combine topic information with the output of BERT. Finally, we adopt adversarial training to achieve semi-supervised learning, and a large amount of unlabeled data can be used to improve inner representations of the model. Experimental results on the dataset of COVID-19 English tweets show that our model outperforms classic and state-of-the-art baselines. [ABSTRACT FROM AUTHOR]

While recent developments of neural network (NN) models have led to a series of record-breaking achievements in many applications, the lack of sufficiently good datasets remains a problem for some applications. For such a problem, we can however exploit a large number of unstructured text corpora as an external knowledge to complement the training data, and most prevailing neural network solutions employ word embedding methods for such purposes. In this paper, we propose LDA-Reg, a novel knowledge driven regularization framework based on Latent Dirichlet Allocation (LDA) as an alternative to the word embedding methods to adaptively utilize abundant external knowledge and to interpret the NN model. For the joint learning of the parameters, we propose EM-SGD, an effective update method which incorporates Expectation Maximization (EM) and Stochastic Gradient Descent (SGD) to update parameters iteratively. Moreover, we also devise a lazy update and sparse update method for the high-dimensional inputs and sparse inputs respectively. We validate the effectiveness of our regularization framework through an extensive experimental study over real world and standard benchmark datasets. The results show that our proposed framework not only achieves significant improvement over state-of-the-art word embedding methods but also learns interpretable and significant topics for various tasks. [ABSTRACT FROM AUTHOR]

Health care prescription fraud and abuse result in major financial losses and adverse health effects. The growing budget deficits of health insurance programs and recent opioid drug abuse crisis in the United States have accelerated the use of analytical methods. Unsupervised methods such as clustering and anomaly detection could help the health care auditors to evaluate the billing patterns when embedded into rule-based frameworks. These decision models can aid policymakers in detecting potential suspicious activities. This manuscript proposes an unsupervised temporal learning-based decision frontier model using the real world Medicare Part D prescription data collected over 5 years. First, temporal probabilistic hidden groups of drugs are retrieved using a structural topic model with covariates. Next, we construct combined concentration curves and Gini measures considering the weighted impact of temporal observations for prescription patterns, in addition to the Gini values for the cost. The novel decision frontier utilizes this output and enables health care practitioners to assess the trade-offs among different criteria and to identify audit leads. [ABSTRACT FROM AUTHOR]

The development of Information and Communication Technologies (ICT) and Web 2.0 promotes the emergence of diverse social ecosystems like social Internet of Things (IoT), social media and online communities. User-generated textual content (UGTC), which consists of unstructured texts, is the most important and common type of user-generated content in social ecosystems. UGTC in social ecosystems is generated according to two types of context information—global context (topics) and local context (semantic regularities). For UGTC modeling, topic models just consider global context but ignore semantic regularities, while semantic embedding models are on the opposite. So only utilizing topic models or semantic embedding models to model UGTC suffers from some drawbacks. For this problem, we propose a semantic embedding enhanced topic model named SEE-Twitter-LDA for accurately modeling UGTC in social ecosystems. The core of SEE-Twitter-LDA is that words are generated according to mutual semantic information of topics and semantic regularities. So global context and local context are jointly considered for UGTC modeling. By utilizing 553 098 tweets sampled from Twitter and 211 233 posts sampled from Weibo, we validate SEE-Twitter-LDA's better performance on perplexity, topic divergence and topic coherence versus existing related models. [ABSTRACT FROM AUTHOR]

3-D printing is an additive manufacturing process, which enables products to be custom designed. Obtaining the evolution trend and identifying the potential R&D hotspots of 3-D printing technology are important strategic issues for both nations and enterprises. In this article, a combination method is proposed using the Viterbi algorithm to identify technological terms in patents, using the latent Dirichlet allocation model to capture potential technological topics in patents, and using the hidden Markov model to analyze the distribution and evolution pattern of the technological topics. Based on the results, technological topics and future R&D hotspots in 3-D printing are identified. Among the technological topics, extrusion devices and laser powder molding are gradually losing R&D; model acquisition, extrusion devices, rotary printing, and denture are the key topics of technological evolution that are difficult to overcome; laser focusing, print head, color printing, recyclable, feeding, control panel, and denture will become the focus of R&D in the near future. Finally, a comparison and experiments are given to demonstrate the plausibility of our method. [ABSTRACT FROM AUTHOR]

In this study, we conduct a bibliometric review of the Mathematics journal to map its thematic structure, and to identify major research trends for future research to build on. Our review focuses primarily on the bibliometric clusters derived from an application of a bibliographic coupling algorithm and offers insights into how studies included in the review sample relate to one another to form coherent research streams. We combine this analysis with keyword frequency and topic modeling analyses to reveal the discourse that is taking place in the journal more recently. We believe that a systematic/computer-assisted review of the Mathematics journal can open a path for new developments and discoveries in research and help editors assess the performance and historic evolution of the journal and predict future developments. In so doing, the findings should advance our cumulative understanding in those areas consistent with the scope of the Mathematics journal, such as applied mathematics, analytics, and computational sciences. [ABSTRACT FROM AUTHOR]

With the development of the Internet and mobile networks, social networks have gradually become an essential tool and widespread application. Therefore, the research on short text semantic modelling of social networks has attracted widespread attention. However, modelling short texts encounter the semantics sparsity and multiple meanings of a word in social networks. To solve the above problems, we propose a user-based topic model with topical word embeddings semantic modelling method, namely SM-UTM. Firstly, we construct the user topic model to aggregate short text. Secondly, we build word pair in the user topic model to alleviate semantics sparsity in social networks. In addition, we introduce the time information of social networks into the topic model to jointly constrain the generation process of topics, to improve the quality of semantic representation of social network short texts. Finally, we use the topic word embedding learning based on deep learning to train and optimize the word vector according to the learning results of the user topic model, to alleviate the problem of polysemy in social networks. We build multiple groups of quantitative and qualitative experiments based on the crawled real Sina Weibo data. The experimental results show that our SM-UTM is significantly better than the comparison method in the evaluation indicators of topic consistency, purity and entropy. [ABSTRACT FROM AUTHOR]

Manual processing of a large volume of video data captured through closed-circuit television is challenging due to various reasons. First, manual analysis is highly time-consuming. Moreover, as surveillance videos are recorded in dynamic conditions such as in the presence of camera motion, varying illumination, or occlusion, conventional supervised learning may not work always. Thus, computer vision-based automatic surveillance scene analysis is carried out in unsupervised ways. Topicmodelling is one of the emerging fields used in unsupervised information processing. Topic modelling is used in text analysis, computer vision applications, and other areas involving spatio-temporal data. In this article, we discuss the scope, variations, and applications of topic modelling, particularly focusing on surveillance video analysis. We have provided a methodological survey on existing topic models, their features, underlying representations, characterization, and applications in visual surveillance's perspective. Important research papers related to topic modelling in visual surveillance have been summarized and critically analyzed in this article. [ABSTRACT FROM AUTHOR]

Cultural ecosystem services (CES) are of significant interest due to their effectiveness in assessing the impact of ecosystem degradation on human well-being. The evaluation of spatial coordination between the supply and demand of urban CES is crucial for guiding the sustainable development of urban areas. However, current research and survey methods are time-consuming and laborious, lacking in full integration of beneficiary opinions, and insufficient in examining multiple CES in an integrated manner. In this study, we have developed a novel framework that integrates an investigation-driven approach with a natural language processing topic model to assess the supply and demand of CES and their spatial correspondence. Using the main urban area of Xi'an as a case study, we analyzed the spatial coherence of supply and demand including aesthetic, cultural, educational, recreational, and spiritual services. Our findings demonstrate that the fusion of multiple models can effectively evaluate the CES supply from a natural perspective of nature and the demand for CES from a beneficiary perspective. High-supply areas for CES are concentrated in the center region of the study area and in proximity to roads and rivers. Aesthetic service supply was higher than that of other service types, followed by recreational and spiritual services supply. Demand is notably high in the city's hinterland, with substantial interest in recreational and aesthetic services. We observed mismatches between supply and demand across service categories ranging from 33.58% to 67.64%, with supply exceeding demand in most regions. Therefore, tailored measures should be implemented to address the supply and demand mismatch of CES based on local conditions. Our research introduces a new technical tool for analyzing the supply-demand matching relationship of CES and establishes a theoretical foundation for sustainable urban management. • Topic model can effectively study cultural ecosystem services (CES) demand. • The distribution of supply and demand for CES are highly localized. • A framework for supply and demand assessment by urban CES. • The framework addresses the difficulty of measuring services at the city scale. [ABSTRACT FROM AUTHOR]

Benchmarking is one of the key management approaches to support companies to continuously improve and gain competitive advantage. However, previous benchmarking management suffered from subjectivity and one-sidedness, and the benchmarking evaluation was only to identify the benchmark and its own deficiencies, with a high degree of independence between the sub-processes. The framework is the first to jointly apply topic mining and relation extraction techniques to establish an objective index system and mine the mechanistic relations among the indexes, and finally provides a concise and interactive display of the results based on knowledge graph visualization. In this study, 50 oil and gas companies are taken as examples to carry out the case study of sustainable development benchmarking management, and a sustainable development program with strong logic and systematicity is proposed at both macro and micro levels. This study results verify that the framework can effectively improve the scientific, comprehensive and systematic nature of sustainable development benchmarking management. At the same time, the framework extends the role of the index system so that it can provide the necessary mechanistic guidance for subsequent sustainable development programmatic to achieve programmatic with a high degree of synergy. In addition, the framework has a high degree of coherence with the United Nations Sustainable Development Goals (SDGs), and the proposals put forward can not only be used to improve the sustainable development capability of companies themselves, but also contribute to the realization of the SDGs. [Display omitted] • A novel sustainable development benchmarking framework for energy companies. • A first joint application of topic mining and relation extraction for benchmarking. • Energy companies should promote localization and diversification of the workforce. • Energy companies should expand joint venture projects with developing country. [ABSTRACT FROM AUTHOR]

Stigma surrounding suicide is a massive problem in Indonesia. Thus, it is important to study how conversations about suicide take place. We take a machine learning approach and study tweets with suicide keywords to understand how people converse about suicide or express suicide ideation. Tweets with suicide-related keywords were extracted from May to June 2023. 20,057 tweets were subject to topic modelling with an 11-topic solution. While most topics contain negative messages, no purely stigmatizing topics emerge, despite prior research suggesting overwhelming stigma. Various kinds of existential, emotional, and social tweets about suicide take place among Indonesian users, indicating that Indonesian Twitter users utilize the platform to express their thoughts and emotions. Notably, religious-spiritual keywords are highly prevalent, suggesting that in a highly religious society, there is a need for policy makers and awareness campaigns to frame their positive messaging within the society’s religious context. [ABSTRACT FROM AUTHOR]

Quantitative text analysis is a set of procedures for grouping documents and the words and expressions that make up the content of documents, supported by statistical methods and tools. Its main advantage is that in addition to the information that can be extracted from numerical data, it provides a solution for accessing additional information that previously seems unavailable. Thus, text analysis is most commonly used where the text contains more information than could be obtained from quantified data. Text and content analysis is today one of the most exciting and focused areas of data mining and data analysis, both in terms of mathematical, statistical, methodological, and software fit. In this study, latent Dirichlet allocation (LDA) is presented as a possible methodological approach through a concrete example. The primary aim of the article is to show that text analysis can provide long-term and novel perspectives in the fields of management science. [ABSTRACT FROM AUTHOR]

Existing literature regarding punctuated equilibrium theories (PET) and agenda diversity has examined the patterns of cross-domain attention allocation based on western democracies; however, few studies have systematically discussed the situation within a specific policy domain in transitioning countries. To address this research gap, this study empirically explored the punctuations and diversity in attention allocation within China's national e-government issue from 2001 to 2018. By employing the latent Dirichlet allocation topic model to analyse relevant articles on e-government in the People's Daily, we derived 10 distinct dimensions hidden in the corpora and constructed an original dataset of the e-government agenda. The empirical findings revealed that the information processing in the domain of the e-government in China are leptokurtic, and a gradual decrease in the intensity of punctuations with the enhancement of central coordination and civic participation was also observed. We also found that the dynamics of attention diversity partially followed the theoretical expectations of classic scholarships. By examining patterns of attention allocation across multiple dimensions within a certain policy, our findings speak to literature on PET and agenda diversity and increases the externality of relevant theories derived from the cross-domain analysis. [ABSTRACT FROM AUTHOR]

With the rapid growth of social network users, the social network has accumulated massive social network topics. However, due to the randomness of content, it becomes sparse and noisy, accompanied by many daily chats and meaningless topics, which brings challenges to bursty topics discovery. To deal with these problems, this paper proposes the spatial-temporal topic model with sparse prior and recurrent neural networks (RNN) prior for bursty topic discovering (ST-SRTM). The semantic relationship of words is learned through RNN to alleviate the sparsity. The spatial-temporal areas information is introduced to focus on bursty topics for further weakening the semantic sparsity of social network context. Besides, we introduced the "Spike and Slab" prior to decouple the sparseness and smoothness. Simultaneously, we realized the automatic discovery of social network bursts by introducing the burstiness of words as the prior and binary switching variables. We constructed multiple sets of comparative experiments to verify the performance of ST-SRTM by leveraging different evaluation indicators on real Sina Weibo data sets. The experimental results confirm the superiority of our ST-SRTM. [ABSTRACT FROM AUTHOR]

At present, we live in prosperity contrary to the past times. As income increases, people enjoy wealth, but more people tend to pursue their own inner happiness: travel. People go to other places or visit foreign countries for business or journey. This study aims to identify the best tour route for foreign tourists in South Korea. Based on the review analysis results, this paper also aims to put forward techniques and methodologies required in practical affairs when developing a travel site or a travel application. On this note, it collected tourists' reviews at the Tripadvisor official website and conducted text mining technique as well as network analysis using R and Tagxedo, which is a big data analytic tool. The analysis results displayed that there were differences in travel preference, and especially, individual travelers had difficulty traveling by public transportation and selecting travel destinations. Therefore, customized travel routes were suggested for convenient use among travelers. [ABSTRACT FROM AUTHOR]

Recently, the proposed non-parametric Bayesian based techniques which aim to model short-length textual documents through the multinomial distribution on the bag-of-words (BOW), aka mixture model-based approach. Although existing model can effectively deal with the topic/concept drift and textual sparsity problems, they are unable to exploit the semantic sequential representation of text as well as the co-occurrence relationships between words. To meet these challenges, we propose a novel approach called as GOWSeqStream. Our proposed model is a joint integration of graph-of-words (GOW) and deep sequential encoding within the Dirichlet Process Mixture Model (DPMM) framework to improve the performance of text clustering task. Extensive experiments in benchmark real-world datasets demonstrate the effectiveness of our proposed GOWSeqStream model in comparing with recent state-of-the-art baselines. Experimental outputs in terms of NMI standard metric demonstrate the outperformances of proposed GOWSeqStream model over the recent well-known text stream clustering baselines, such as MStream, NPMM and OSDM. [ABSTRACT FROM AUTHOR]

This paper is a study on an industrial engineering approach for the formulation of digital transformation (DX) strategies in the manufacturing industry. When a company is working on the formulation of a DX strategy, it is not uncommon for the stakeholders in the relevant departments to have different opinions and conflict with each other. Therefore, companies invite outside specialized consultants as facilitators and organize workshops with stakeholders to try to compile fair and objective DX requirements. However, not all of the stakeholders who participated in the workshop are satisfied with the results compiled by the consultant. In this paper, we propose to incorporate an augmented digital process of text mining for the content of stakeholders' requirements into the conventional real in-person discussion process of organizing requirements. To evaluate the usefulness of this proposal, we conducted an industrial experiment analyzing stakeholders’ opinions collected during a workshop at an industrial company. In other words, we evaluated whether or not it is possible to discover new insights and potential requirements that the consultant was not aware of. As a result of this experiment, we were able to confirm the possibility of revealing stakeholder opinions that were overlooked by the consultant during the workshop at the time, doing so by applying the proposed approach. From this, we concluded that incorporating text mining into the organization of stakeholder requirements would be effective in the requirements gathering process of DX strategy development. [ABSTRACT FROM AUTHOR]

The purpose of this research is to clarify the relationship between word-of-mouth content and hotel rankings in a Japanese travel information website. We conducted a latent semantic analysis of word-of-mouth data based on the Latent Dirichlet Allocation (LDA) model and created variables for word-of-mouth content (topic indexes). A hierarchical multiple regression analysis was performed on the ranking order of the objective variable, with the control variable (customer and hotel characteristics) as the explanatory variables in the first stage and the topic index as the explanatory variable in the second stage. We used 3,769 word-of-mouth data written for 79 hotels between October 1, 2019 and September 30, 2020 for the analysis. As a result, five topics (bath and hot springs, cleanness, virus protection, staff, and meals with family) were extracted from the word-of-mouth data and five topic indexes were created. We found a statistically significant positive impact of the "virus protection," “staff” and “meals with family” topic indexes on hotel rankings. It is important for hotel operators to provide personalized and attentive services to each guest, to improve its ranking and for guests to enjoy special moments with their family and friends on a vacation under a secure environment. An analytical model for clarifying the relationship between word-of-mouth content (unstructured data) and ranking order (structured data) is presented, and can be practically applied to show hotel operators how to improve their hotel services to increase their ranking on travel information websites. [ABSTRACT FROM AUTHOR]

In the recent world of diverse products, it is important to identify the consumer segments that should be targeted for each product. In order to identify the consumer segments to be targeted from among all consumers, it is common practice to use target attributes such as “male office workers in their 30s.” If data including attributes, hobbies and preferences of all consumers are available, target attributes can be identified using an analytical process. However, it is not realistic for each company to collect such data from all consumers for their products because of the huge cost involved. It is therefore, common to ask a consulting company that have the web browsing and purchasing histories of various sample consumers to analyze the sample data to identify the target. In particular, clustering behavioral data such as web browsing histories of sample consumers and discovering appropriate attributes that characterize target clusters (i.e. cluster attributes) is one of the most commonly used approaches. In real situations, cluster attributes are often assigned by the analyst based on the attribute statistics of the sample consumers belonging to the cluster, and multiple cluster attributes are often assumed for each cluster. When such qualitative analysis and judgment are involved, the selection of cluster attributes strongly depends on the experience and skills of the analyst. In this study, we formulate a model for clustering sample consumers based on their web browsing history, including various interests and preferences, and assigning effective cluster attributes as targets to each cluster. In addition, we propose a method to search for the best target attributes for a given objective function. By demonstrating an analysis of an actual data set, the effectiveness of the proposed method using real data is clarified. [ABSTRACT FROM AUTHOR]

Currently, the majority of retrieval-based question-answering systems depend on supervised training using question pairs. However, there is still a significant need for further exploration of how to employ unsupervised methods to improve the accuracy of retrieval-based question-answering systems. From the perspective of question topic keywords, this paper presents TFCSG, an unsupervised question-retrieval approach based on topic keyword filtering and multi-task learning. Firstly, we design the topic keyword filtering algorithm, which, unlike the topic model, can sequentially filter out the keywords of the question and can provide a training corpus for subsequent unsupervised learning. Then, three tasks are designed in this paper to complete the training of the question-retrieval model. The first task is a question contrastive learning task based on topic keywords repetition strategy, the second is questions and its corresponding sequential topic keywords similarity distribution task, and the third is a sequential topic keywords generation task using questions. These three tasks are trained in parallel in order to obtain quality question representations and thus improve the accuracy of question-retrieval task. Finally, our experimental results on the four publicly available datasets demonstrate the effectiveness of the TFCSG, with an average improvement of 7.1%, 4.4%, and 3.5% in the P@1, MAP, and MRR metrics when using the BERT model compared to the baseline model. The corresponding metrics improved by 5.7%, 3.5% and 3.0% on average when using the RoBERTa model. The accuracy of unsupervised similar question-retrieval task is effectively improved. In particular, the values of P@1, P@5, and P@10 are close, the retrieved similar questions are ranked more advance. [ABSTRACT FROM AUTHOR]

Bicycle-metro integration, in which bicycling is used as a flexible feeder mode to connect with public transport nodes presents new opportunities for sustainable transportation. It is known that the built environment can influence travel attitudes and choice, yet the empirical evidence for the role of built environment features in shaping the bicycle-metro integration remains rare. Inspired by the idea of text mining, this article is an attempt to demonstrate a data-driven semantic framework to capture key topic-based features of land use and bicycle-metro integrated usage in the vicinity of metro stations as well as their interactions. Latent Dirichlet Allocation topic modeling is analogously implemented here to generate a range of probability-based land use patterns and mobility patterns, and the associations between them are investigated by multivariate linear regression. A case study from Shanghai shows that the mixed land use and diversification of urban functions in the catchment areas of the metro stations can be detected effectively by 11 identified land use patterns. Based on 7 derived mobility patterns, this paper gives a probabilistic explanation to the time-varying properties of the bicycle-metro usage. All of the above thematic topics exhibit notably heterogeneous patterns in spatial distribution. The topic compositions in terms of land use pattern and mobility pattern at the station level reveal the current performance of station areas. Plus, results from the regression analysis confirm that most of the land use patterns that are related to various mixed use have close relationships with mobility patterns of bicycle-metro integration. Yet it is noteworthy that the effects of land use patterns often differ and change over time, namely affecting different mobility patterns. This study gives rise to alternative insights into the synergy between bike sharing and metro, which may help policymakers to develop more targeted TOD strategies. [ABSTRACT FROM AUTHOR]

The Transformer-based summarization models rely solely on the attention mechanism for document encoding, making it difficult to accurately capture long-range dependencies in long documents due to the presence of attention redundancy. To address this issue, we propose an extractive summarization framework guided by a topic model (TopicSum) that utilizes a heterogeneous graph neural network to leverage the topic information as document-level features during the sentence selection process, thereby capturing the long-range dependencies among sentences. The sentence-level features in this topic model align with the basic unit of the extractive summarization task. Additionally, a memory mechanism is employed to dynamically store and update the memory module, reducing the potential of repetitive information guiding sentence selection. We evaluated the model on three large document datasets, namely Pubmed, arXiv, and GovReport, and achieved significantly higher Rouge scores than previous works, including extractive and abstractive models. Furthermore, our experiments demonstrate that recent highly regarded large language models such as ChatGPT are insufficient to handle the long document summarization task directly. The proposed approach in this paper exhibits sufficient competitiveness in terms of both generation quality and deployment conditions. [ABSTRACT FROM AUTHOR]

Modelling short text is challenging due to the small number of word co-occurrence and insufficient semantic information that affects downstream Natural Language Processing (NLP) tasks, for example, text classification. Gathering information from external sources is expensive and may increase noise. For efficient short text classification without depending on external knowledge sources, we propose Expressive Short text Classification (EStC). EStC consists of a novel document context-aware semantically enriched topic model called the Short text Topic Model (StTM) that captures words, topics and documents semantics in a joint learning framework. In StTM, the probability of predicting a context word involves the topic distribution of word embeddings and the document vector as the global context, which obtains by weighted averaging of word embeddings on the fly simultaneously with the topic distribution of words without requiring an additional inference method for the document embedding. EStC represents documents in an expressive (number of topics × number of word embedding features) embedding space and uses a linear support vector machine (SVM) classifier for their classification. Experimental results demonstrate that EStC outperforms many state-of-the-art language models in short text classification using several publicly available short text data sets. [ABSTRACT FROM AUTHOR]

Currently, Android malware detection methods always focus on one kind of app feature, such as structural, semantic, or other statistical features. This paper proposes a novel Android malware detection method that integrates multiple features of Android applications. First, to effectively extract the structural and semantic features, we propose a new type of call graph named the class-set call graph (CSCG) that uses the sets of Java classes as nodes and the call relationships between class sets as edges, and we design a dynamic adaptive CSCG construction method that can automatically determine the node size for applications with different scales. The topic model is used to mine the source code semantics from the class sets as the node features. Then, we use a graph attention network (GAT) with max pooling to extract the CSCG feature that encompass both the semantic and structural features of the Android application. Furthermore, we construct a deep multimodal feature fusion network to fuse the CSCG features with permission features. Experimental results show that our method achieves a detection accuracy of 97.28%–99.54% on the three constructed datasets, which is better than the existing methods. • Proposing multimodal features deep fusion method for Android malware detection. • Designing class-set call graph to integrates structural and semantic features. • Adaptive class merging method for class-set call graph construction. • Introducing GAT and max pooling to extract salient graph features. • Proposing fusion network for graph features and permission features. [ABSTRACT FROM AUTHOR]

It is necessary to reduce the cognitive overhead of interpreting the native topic term list of the Latent Dirichlet Allocation (LDA) style topic model. In this regard, automatic topic labeling has become an effective approach to generate meaningful alternative representations of topics discovered for end-users. In this study, we introduced a novel two-phase neural embedding framework with the redundancy-aware graph-based ranking process. It demonstrated how pre-trained neural embedding could be usefully applied in topic terms, sentence presentations, and automatic topic labeling tasks. Moreover, reranking the topic terms optimized the discovered topics with fewer yet more representative terms while retaining the topic information integrality and fidelity. It further decreased the burden of computation caused by neural embedding and improved the overall effectiveness of the labeling system. Compared with the prevailing state-of-the-art and classical labeling systems, our efficient model boosted the quality of the topic labels generated and discovered more meaningful topic labels. [ABSTRACT FROM AUTHOR]

This article presents analysis of data from scientific articles and patents to identify the evolving trends and underlying topics in research on humanoid robots. We used topic modelling based on latent Dirichlet allocation analysis to identify underlying topics in sub-areas in the field. We also used social network analysis to measure the centrality indices of publication keywords to detect important and influential sub-areas and used co-occurrence analysis of keywords to visualise relationships among subfields. The research result is useful to identify evolving topics and areas of current focus in the field of humanoid technology. The results contribute to identify valuable research patterns from publications and to increase understanding of the hidden knowledge themes that are revealed by patents. [ABSTRACT FROM AUTHOR]

Automated sentiment analysis is becoming increasingly recognized due to the growing importance of social media and e-commerce platform review websites. Deep neural networks outperform traditional lexicon-based and machine learning methods by effectively exploiting contextual word embeddings to generate dense document representation. However, this representation model is not fully adequate to capture topical semantics and the sentiment polarity of words. To overcome these problems, a novel sentiment analysis model is proposed that utilizes richer document representations of word-emotion associations and topic models, which is the main computational novelty of this study. The sentiment analysis model integrates word embeddings with lexicon-based sentiment and emotion indicators, including negations and emoticons, and to further improve its performance, a topic modeling component is utilized together with a bag-of-words model based on a supervised term weighting scheme. The effectiveness of the proposed model is evaluated using large datasets of Amazon product reviews and hotel reviews. Experimental results prove that the proposed document representation is valid for the sentiment analysis of product and hotel reviews, irrespective of their class imbalance. The results also show that the proposed model improves on existing machine learning methods. [ABSTRACT FROM AUTHOR]