PurposeIn traditional Chinese medicine (TCM), the mechanism of disease (MD) constitutes an essential element of syndrome differentiation and treatment, elucidating the mechanisms underlying the occurrence, progression, alterations and outcomes of diseases. However, there is a dearth of research in the field of intelligent diagnosis concerning the analysis of MD.Design/methodology/approachIn this paper, we propose a supervised Latent Dirichlet Allocation (LDA) topic model, termed MD-LDA, which elucidates the process of MDs identification. We leverage the label information inherent in the data as prior knowledge and incorporate it into the model’s training. Additionally, we devise two parallel parameter estimation algorithms for efficient training. Furthermore, we introduce a benchmark MD identification dataset, named TMD, for training MD-LDA. Finally, we validate the performance of MD-LDA through comprehensive experiments.FindingsThe results show that MD-LDA is effective and efficient. Moreover, MD-LDA outperforms the state-of-the-art topic models on perplexity, Kullback–Leibler (KL) and classification performance.Originality/valueThe proposed MD-LDA can be applied for the MD discovery and analysis of TCM clinical diagnosis, so as to improve the interpretability and reliability of intelligent diagnosis and treatment.
PurposeSocial media platforms that disseminate scientific information to the public during the COVID-19 pandemic highlighted the importance of the topic of scientific communication. Content creators in the field, as well as researchers who study the impact of scientific information online, are interested in how people react to these information resources and how they judge them. This study aims to devise a framework for extracting large social media datasets and find specific feedback to content delivery, enabling scientific content creators to gain insights into how the public perceives scientific information.Design/methodology/approachTo collect public reactions to scientific information, the study focused on Twitter users who are doctors, researchers, science communicators or representatives of research institutes, and processed their replies for two years from the start of the pandemic. The study aimed in developing a solution powered by topic modeling enhanced by manual validation and other machine learning techniques, such as word embeddings, that is capable of filtering massive social media datasets in search of documents related to reactions to scientific communication. The architecture developed in this paper can be replicated for finding any documents related to niche topics in social media data. As a final step of our framework, we also fine-tuned a large language model to be able to perform the classification task with even more accuracy, forgoing the need of more human validation after the first step.FindingsWe provided a framework capable of receiving a large document dataset, and, with the help of with a small degree of human validation at different stages, is able to filter out documents within the corpus that are relevant to a very underrepresented niche theme inside the database, with much higher precision than traditional state-of-the-art machine learning algorithms. Performance was improved even further by the fine-tuning of a large language model based on BERT, which would allow for the use of such model to classify even larger unseen datasets in search of reactions to scientific communication without the need for further manual validation or topic modeling.Research limitations/implicationsThe challenges of scientific communication are even higher with the rampant increase of misinformation in social media, and the difficulty of competing in a saturated attention economy of the social media landscape. Our study aimed at creating a solution that could be used by scientific content creators to better locate and understand constructive feedback toward their content and how it is received, which can be hidden as a minor subject between hundreds of thousands of comments. By leveraging an ensemble of techniques ranging from heuristics to state-of-the-art machine learning algorithms, we created a framework that is able to detect texts related to very niche subjects in very large datasets, with just a small amount of examples of texts related to the subject being given as input.Practical implicationsWith this tool, scientific content creators can sift through their social media following and quickly understand how to adapt their content to their current user’s needs and standards of content consumption.Originality/valueThis study aimed to find reactions to scientific communication in social media. We applied three methods with human intervention and compared their performance. This study shows for the first time, the topics of interest which were discussed in Brazil during the COVID-19 pandemic.
PurposeEducational data mining (EDM) discovers significant patterns from educational data and thus can help understand the relations between learners and their educational settings. However, most previous data mining techniques focus on prediction of learning performance of learners without integrating learning patterns identification techniques.Design/methodology/approachThis study proposes a new framework for identifying learning patterns and predicting learning performance. Two modules, the learning patterns identification module and the deep learning prediction models (DNN), are integrated into this framework to identify the difference of learning performance and predicting learning performance from profiles of students.FindingsExperimental results from survey data indicate that the proposed identifying learning patterns module could facilitate identifying valuable difference (change) patterns from student's profiles. The proposed learning performance prediction module which adapts DNN also performs better than traditional machine techniques in prediction performance metrics.Originality/valueTo our best knowledge, the framework is the only educational system in the literature for identifying learning patterns and predicting learning performance.
PurposeThis study explores the career trajectories of Chief Executive Officers (CEOs) to uncover unique characteristics that contribute to their success. By utilizing web scraping and machine learning techniques, over two thousand CEO profiles from LinkedIn are analyzed to understand patterns in their career paths. This study offers an alternative approach compared to the predominantly qualitative research methods employed in previous research.Design/methodology/approachThis study proposes a framework for analyzing CEO career patterns. Job titles and company information are encoded using the Standard Occupational Classification (SOC) scheme. The study employs the Needleman-Wunsch optimal matching algorithm and an agglomerative approach to construct distance matrices and cluster CEO career paths.FindingsThis study gathered data on the career transition processes of graduates from several renowned public and private universities in the United States via LinkedIn. Employing machine learning techniques, the analysis revealed diverse career trajectories. The findings offer career guidance for individuals from various academic backgrounds aspiring to become CEOs.Research limitations/implicationsThe building of a career sequence that takes into account the number of years requires integers. Numbers that are not integers have been rounded up to facilitate the optimal matching process but this approach prevents a perfectly accurate representation of time worked.Practical implicationsThis study makes an original contribution to the field of career pattern analysis by disclosing the distinct career path groups of CEOs using the rich LinkedIn online dataset. Note that our CEO profiles are not restricted in any industry or specific career paths followed to becoming CEOs. In light of the fact that individuals who hold CEO positions are usually perceived by society as successful, we are interested in finding the characteristics behind their success and whether either the title held or the company they remain at show patterns in making them who they are today.Originality/valueAs a matter of fact, nearly all CEOs had previous experience working for a non-Fortune organization before joining a Fortune company. Of those who have worked for Fortune firms, the number of CEOs with experience in Fortune 500 forms exceeded those with experience in Fortune 1,000 firms.
PurposeThis study aims to provide measurable information that evaluates a company’s ESG performance based on the conceptual connection between ESG, non-financial elements of a company and the UN Sustainable Development Goals (SDGs) for resolving global issues.Design/methodology/approachA novel data processing method based on the BERT is presented and applied to analyze the changes and characteristics of SDG-related ESG texts from companies’ disclosures over the past decade. Specifically, ESG-related sentences are extracted from 93,277 Form 10-K filings disclosed between 2010 and 2022 and the similarity between these extracted sentences and SDGs statements is calculated through sentence transformers. A classifier is created by fine-tuning FinBERT, a financial domain-specific pre-trained language model, to classify the sentences into eight ESG classes.FindingsThe quantified results obtained from the classifier reveal several implications. First, it is observed that the trend of SDG-related ESG sentences shows a slow and steady increase over the past decade. Second, large-cap companies relatively have a greater amount of SDG-related ESG disclosures than small-cap companies. Third, significant events such as the COVID-19 pandemic greatly impact the changes in disclosure content.Originality/valueThis study presents a novel approach to textual analysis using neural network-based language models such as BERT. The results of this study provide meaningful information and insights for investors in socially responsible investment and sustainable investment and suggest that corporations need a long-term plan regarding ESG disclosures.
