Article
Economics
Hali Edison, Hector Carcel
Summary: This paper applies the LDA algorithm to analyze FOMC transcripts, finding that discussions on economic modeling were dominant during the Global Financial Crisis, discussions on the banking system increased post-crisis, and discussions on communication gained relevance recently. The paper suggests that researchers could further utilize LDA analysis to identify topic priorities in relevant documents.
APPLIED ECONOMICS LETTERS
(2021)
Article
Thermodynamics
Doowon Choi, Chul Kim
Summary: This study applies the MOB algorithm to analyze the 2012 CBECS survey data, revealing the heterogeneous effects and mutual influences of energy variables on building energy consumption.
BUILDING SIMULATION
(2021)
Article
Multidisciplinary Sciences
Kelly Ann Coulter
Summary: This paper examines the relationship between international news discourses and Bitcoin price, finding that certain discourses have a negative impact on Bitcoin's market price. The study also shows that the source of the news and the geographical region can amplify this effect.
ROYAL SOCIETY OPEN SCIENCE
(2022)
Article
Green & Sustainable Science & Technology
Marcin Wyskwarski
Summary: This study aimed to identify the most desired project manager competencies by analyzing job advertisements through text mining. The results showed that text mining of job offers can help determine project manager competencies in demand, which can be useful for organizations training future project managers to adapt curricula to labor market needs and monitor current trends in project manager requirements.
Article
Biochemical Research Methods
Alan Min, Timothy Durham, Louis Gevirtzman, William Stafford Noble
Summary: Single cell ATAC-seq (scATAC-seq) enables the mapping of regulatory elements in fine-grained cell types. In this study, the authors propose using latent Dirichlet allocation (LDA) with nonuniform matrix priors to improve the analysis of scATAC-seq data. They demonstrate the effectiveness of this method in capturing cell type information from small scATAC-seq datasets from C. elegans nematodes and mouse skin cells.
PLOS COMPUTATIONAL BIOLOGY
(2023)
Article
Computer Science, Information Systems
Joao Henriques, Joao Ferreira
Summary: The goal of this research is to find associations between trending topics and companies on social media platforms, and a prototype called HotRivers was developed for this purpose. The study applied text mining techniques to process tweets and train personalized company models, resulting in a list of trending topics matched to the target companies. The results showed associations between trending topics and companies.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Review
Mathematics
Javier De la Hoz-M, M. Jose Fernandez-Gomez, Susana Mendes
Summary: LDAShiny is an open source application that provides an interactive graphical user interface for reviewing scientific literature using latent Dirichlet allocation algorithm and machine learning tools. Through analysis, it was found that 14 topics were sufficient to describe the reviewed literature, with research topics on Oreochromis niloticus species mainly related to growth performance, body weight, heavy metals, genetics, and water quality.
Article
Hospitality, Leisure, Sport & Tourism
Jun Liu, Yunyun Yu, Fuad Mehraliyev, Sike Hu, Jiaqi Chen
Summary: This study extends the general sentiment dictionary in Chinese to a restaurant-domain-specific dictionary and uses it to analyze online restaurant reviews. The results identify love and anger as the emotions with the highest impact on online ratings, and reveal the factors that constitute these emotions. These findings have practical implications for restaurant managers, platforms, and consumers.
INTERNATIONAL JOURNAL OF CONTEMPORARY HOSPITALITY MANAGEMENT
(2022)
Article
Hospitality, Leisure, Sport & Tourism
Jun Liu, Yunyun Yu, Fuad Mehraliyev, Sike Hu, Jiaqi Chen
Summary: This study aims to explore discrete emotions in restaurant reviews by building a restaurant-domain-specific sentiment dictionary. The findings highlight that love and anger have the highest effect on online ratings, and identify the factors that constitute these emotions.
INTERNATIONAL JOURNAL OF CONTEMPORARY HOSPITALITY MANAGEMENT
(2022)
Article
Public, Environmental & Occupational Health
Soowon Park, Yaeji Kim-Knauss, Jin-ah Sim
Summary: By analyzing queries on online platforms regarding mental disorders, the study found that people are interested in practical information such as symptoms, treatments, and social welfare benefits, necessitating a diverse range of information in educational programs.
FRONTIERS IN PUBLIC HEALTH
(2021)
Article
Engineering, Industrial
Zheyuan Hu, Wenhao Zhao, Hui Xiong, Xu Zhang
Summary: This paper proposes a hierarchical retrieval approach for the automatic generation of assembly instructions based on previously used technical instruction cards. A case-based reasoning method is used to encode the assembly process and retrieve suitable cases, and an improved weighted latent Dirichlet allocation text mining technique is applied to explore unstructured text topics and recommend the optimal case. The proposed method is tested on automotive assembly process data from 12,034 used instruction cards, and the results show improved quality and speed compared to traditional methods.
JOURNAL OF MANUFACTURING SYSTEMS
(2023)
Article
Biochemistry & Molecular Biology
Sk Abdul Amin, Nilanjan Adhikari, Tarun Jha
Summary: In this study, a diverse set of compounds were analyzed using recursive partitioning (RP) analysis to develop decision trees for discriminating HDAC8 inhibitors from non-inhibitors. Understanding essential structural and physicochemical parameters is crucial for designing potential and selective HDAC8 inhibitors, and the results validate previous findings from Bayesian modeling. This comparative learning will enhance drug discovery efforts related to HDAC8 inhibitors.
JOURNAL OF BIOMOLECULAR STRUCTURE & DYNAMICS
(2021)
Article
Computer Science, Interdisciplinary Applications
Francesco Lupi, Mohammed M. Mabkhot, Eleonora Boffa, Pedro Ferreira, Dario Antonelli, Antonio Maffei, Niels Lohse, Michele Lanzetta
Summary: With the rapid advancements in technology and the evolving competences required in engineering, there is a need to harmonize and unify engineering professional figures. The current limitations in defining and updating engineers' archetypes are due to the absence of a structured and automated approach. This study aims to enhance the definition of professional figures in engineering by automating archetype definitions through text mining and adopting a objective and structured methodology based on topic modeling.
COMPUTERS IN INDUSTRY
(2023)
Article
Engineering, Civil
Ahmed Elsayed, Sarah Rixon, Christina Zeuner, Jana Levison, Andrew Binns, Pradeep Goel
Summary: Loss of nutrients can cause environmental problems in surface water and groundwater. Protecting these water sources is crucial. Existing studies on nutrient transport often focus on specific water systems and neglect other systems. This study used a text mining algorithm to identify key topics and research gaps in nutrient transport. Five research gaps were identified, including the need for integrated models and the study of climate change and crop management on nutrient transport.
JOURNAL OF HYDROLOGY
(2023)
Article
Computer Science, Theory & Methods
Anupam Singh, Aldona Glinska-Newes
Summary: This study applies big data and text mining techniques to analyze the topics and sentiment of Twitter posts about organic foods. The results show that people generally have a positive attitude towards organic foods, although there are also skeptics. The study fills gaps in existing research and provides a fresh perspective on public attitude towards sustainable food consumption.
JOURNAL OF BIG DATA
(2022)
Article
Multidisciplinary Sciences
Patrick Mair, Eva Hofmann, Kathrin Gruber, Reinhold Hatzinger, Achim Zeileis, Kurt Hornik
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2015)
Article
Computer Science, Interdisciplinary Applications
Laura Vana, Ronald Hochreiter, Kurt Hornik
Article
Statistics & Probability
Thomas Rusch, Kurt Hornik, Patrick Mair
JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS
(2018)
Article
Statistics & Probability
Kurt Hornik, Bettina Gruen
JOURNAL OF MULTIVARIATE ANALYSIS
(2014)
Article
Environmental Sciences
Stefan Bachhofner, Ana-Maria Loghin, Johannes Otepka, Norbert Pfeifer, Michael Hornacek, Andrea Siposova, Niklas Schmidinger, Kurt Hornik, Nikolaus Schiller, Olaf Kaehler, Ronald Hochreiter
Article
Statistics & Probability
Thomas Rusch, Patrick Mair, Kurt Hornik
Summary: This paper introduces the Cluster Optimized Proximity Scaling (COPS) method, aiming to find a low-dimensional configuration with clusteredness to improve the clustering of objects, enabling visual identification of clusters of mental states.
JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS
(2021)
Article
Statistics & Probability
Florian Schwendinger, Bettina Grun, Kurt Hornik
Summary: This paper systematically compares different optimization algorithms to obtain the maximum likelihood estimates for the regression coefficients in log-binomial regression, finding that conic optimizers emerge as the preferred choice due to their reliability, lack of requirement to tune hyperparameters, and speed.
COMPUTATIONAL STATISTICS
(2021)
Article
Computer Science, Interdisciplinary Applications
Julian Amon, Kurt Hornik
Summary: This paper takes an alternative approach to studying the factors associated with scientific prestige by examining the relationship between linguistic and meta characteristics of academic papers and the rankings of the journals they appear in. The study uses text mining tools to extract features from a large corpus of economics journal articles and estimates regression models to analyze the relationship between these features and journal rankings. The results identify several predictors, including paper length, coreference chain span, writing style, density of the article, collaboration in research teams, and references cited, as the most informative drivers of scientific prestige.
JOURNAL OF INFORMETRICS
(2022)
Article
Business, Finance
Rainer Hirk, Laura Vana, Kurt Hornik
Summary: This paper proposes a longitudinal credit rating model that considers the serial correlation in ratings. By adding an autoregressive structure to a multivariate ordinal regression model, the model significantly improves the goodness-of-fit and predictive performance compared to static models. The model allows for conditional predictions based on a firm's past rating history, outperforming unconditional predictions in both in-sample and out-of-sample scenarios. Additionally, the model is capable of handling missing rating observations. An empirical analysis using US publicly traded corporates rated by S&P from 1985-2016 shows that S&P exhibits procyclical aspects in their rating behavior.
JOURNAL OF EMPIRICAL FINANCE
(2022)
Article
Economics
Paul Hofmarcher, Jesus Crespo Cuaresma, Bettina Gruen, Stefan Humer, Mathias Moser
JOURNAL OF MACROECONOMICS
(2018)
Article
Economics
Jesus Crespo Cuaresma, Bettina Gruen, Paul Hofmarcher, Stefan Humer, Mathias Moser
EUROPEAN ECONOMIC REVIEW
(2016)
Article
Economics
Paul Hofmarcher, Jesus Crespo Cuaresma, Bettina Gruen, Kurt Hornik
JOURNAL OF FORECASTING
(2015)
Article
Computer Science, Interdisciplinary Applications
Kurt Hornik, Bettina Gruen
JOURNAL OF STATISTICAL SOFTWARE
(2014)
Correction
Anthropology
Angela Bohn, Christian Buchta, Kurt Hornik, Patrick Mair