TY - JOUR A1 - Roedenbeck, Marc A1 - Herold, Marcel T1 - Personality Profiles in Online Job Advertisements: Applying a Weighted Closed Vocabulary Approach to Identify Generic Profiles and Clusters JF - SAGE Open N2 - On the background of the supplementary applicant-job fit theory, this study tests the application of a weighted closed vocabulary approach as introduced by Ostendorf on a large corpus of online job advertisements (OJA). Therefore, a large dataset of online job advertisements (OJAs) was scraped from a German job portal (N = 151k), and a sample (n = 3,239) was selected based upon Porter’s Value Chain and corresponding job functions. The dataset was pre-processed with techniques of natural language processing, and the predefined words extracted from the OJAs were weighted with the average subject matter expert rating, as offered in the vocabulary of Ostendorf. The results reveal that, first, 0 to 12 words describing personality are used across OJAs, second, that HR, sales, and finance positions show values higher than sample mean (3.22), third, that HR positions show loadings across all dimensions of personality above the sample means (extroversion: 2.32, agreeableness: 3.98, conscientiousness: 4.28, emotional stability: 3.54, openness: 2.0) followed by sales. The job profiles being significantly parallel on equal levels are HR clerk and referee, as well as IT, R&D, and procurement. While conscientiousness is reported to be an important predictor of performance, our study showed only low to average importance in OJAs. KW - LIWC KW - cluster analysis KW - natural language processing KW - online job advertisement KW - personality KW - profile analysis Y1 - 2026 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:526-opus4-21282 VL - 16 IS - 1 PB - Sage ER - TY - JOUR A1 - Herold, Marcel A1 - Roedenbeck, Marc T1 - Cultural values and the P-O fit: comparative NLP analysis of German online job advertisements JF - Evidence-based HRM N2 - Purpose Within the Person-Organization fit framework and Signalling Theory, this study investigates the performance of word dictionaries detecting cultural values in online job advertisements as one form of external communication of an organization. Based upon a merge of the dictionaries, a corporate value analysis of Germany is conducted. Design/methodology/approach The study builds on a dataset (n > 151 k) of online job advertisements which were scraped from a German job portal. It was pre-processed according to natural language processing standards. For analysing the values of an organization a dictionary based word count was applied. Therefore, the current state-of-the-art dictionaries were tested, and an enhanced dictionary was developed and translated from English to German. Finally, a cluster analysis was conducted. Findings This study supports the possibility of measuring cultural values in texts where the enhanced dictionary based on Ponitzovskiy shows the best results. It thereby supports the use of the Universal Value Structure model (Schwartz, 1992) as well as the Signalling Theory (Guest et al., 2021), that values spread across 10 core or 4 aggregated dimensions are communicated via online job advertisements. Finally, the study offers a profile of the German corporate culture average as well as 4 cultural clusters and separate organizations, all with different profiles. Originality/value This study develops an enhanced dictionary based on a large dataset of online job advertisements for analysing the external communication of values or culture of an organization for improving the Person-Organization fit. KW - culture KW - job advertisement KW - natural language processing KW - person-organization fit Y1 - 2025 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:526-opus4-19987 PB - Emerald ER - TY - JOUR A1 - Herold, Marcel A1 - Roedenbeck, Marc T1 - Matching competency frameworks with job advertisements: a data-driven analysis of its practical application in the healthcare sector JF - Evidence-based HRM N2 - Purpose Competency-based human resource management (CBHRM) is a key component of all organisations but needs to be regularly reviewed and evaluated to ensure the quality of healthcare professionals. One common taxonomy of competency domains for health professions is from Englander et al., where this paper aims to conduct a large-scale analysis based on topic modelling to investigate the extent to which the competency framework for the healthcare sector is applied in the German job market of health professions. Design/methodology/approach The quantitative NLP analysis of a dataset consisting of 3,362 online job advertisements of nurses and doctors was scraped from a German job portal. The data was pre-processed according to Miner et al. For the analysis, the authors applied unsupervised (e.g. HDP, LDA) and supervised (BERTopic) methods and content analysis. Based on the extracted topics a word list was created and these words were coded to existing dimensions of the competency framework of Englander et al. or new dimensions were created. Findings Comparing methodologies, HDP (unsupervised) and BERTopic (supervised) were the best performing while the BERTopic algorithm outperforms HDP. For the doctor dataset 46% of one main dimension was identified but with an overall coverage of 69%, for the care dataset is weaker with 30.8% but an overall coverage of 100%. Additionally, the taxonomy was enhanced with supplementary competencies of “personality/characteristics” and “leadership” as well as two facets of job description which are “place of work” and “job conditions”. Originality/value On the one hand selected dimensions of the taxonomy could be clearly identified but on the other hand, there is a documented gap between the taxonomy and the competencies advertised. One cause may lie in the NLP algorithms but applicants may also have the same difficulties when reading the OJAs. Thus, practitioners should carefully review OJAs regarding better separating explicit competencies they are searching for. For the scientific development of new competency frameworks, our data-driven approach exemplified an extension of a given taxonomy. KW - online job advertisement KW - natural language processing KW - competency frameworks KW - content analysis KW - (un)supervised learning Y1 - 2025 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:526-opus4-19971 SN - 2049-3991 PB - Emerald ER - TY - JOUR A1 - Przymus, Piotr A1 - Rykaczewski, Krzysztof A1 - Martín-Segura, Adrián A1 - Truu, Jaak A1 - De Santa Pau, Enrique Carrillo A1 - Kolev, Mikhail A1 - Naskinova, Irina A1 - Gruca, Aleksandra A1 - Sampri, Alexia A1 - Frohme, Marcus A1 - Nechyporenko, Alina T1 - Deep Learning in Microbiome Analysis: A Comprehensive Review of Neural Network Models JF - Frontiers in Microbiology N2 - Microbiome research, the study of microbial communities in diverse environments, has seen significant advances due to the integration of deep learning (DL) methods. These computational techniques have become essential for addressing the inherent complexity and high-dimensionality of microbiome data, which consist of different types of omics datasets. Deep learning algorithms have shown remarkable capabilities in pattern recognition, feature extraction, and predictive modeling, enabling researchers to uncover hidden relationships within microbial ecosystems. By automating the detection of functional genes, microbial interactions, and host-microbiome dynamics, DL methods offer unprecedented precision in understanding microbiome composition and its impact on health, disease, and the environment. However, despite their potential, deep learning approaches face significant challenges in microbiome research. Additionally, the biological variability in microbiome datasets requires tailored approaches to ensure robust and generalizable outcomes. As microbiome research continues to generate vast and complex datasets, addressing these challenges will be crucial for advancing microbiological insights and translating them into practical applications with DL. This review provides an overview of different deep learning models in microbiome research, discussing their strengths, practical uses, and implications for future studies. We examine how these models are being applied to solve key problems and highlight potential pathways to overcome current limitations, emphasizing the transformative impact DL could have on the field moving forward. KW - clasiffication KW - clustering KW - deep learning KW - natural language processing KW - microbiome Y1 - 2025 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:526-opus4-19889 SN - 1664-302X VL - 15 PB - Frontiers ER - TY - THES A1 - Müller, Philipp T1 - Mensch-Maschine-Kommunikation basierend auf Natural Language Processing: Evaluierung von Möglichkeiten für den Einsatz eines humanoiden Roboters im Kontext einer Hochschulbibliothek N2 - Thematisiert wird der Weg, von einer theoretischen Betrachtung computerlinguistischer Grundlagen der Sprachverarbeitung, bis hin zu einer prototypischen Umsetzung einer Software zur natürlich sprachlichen Dialogkommunikation zwischen Mensch und Roboter. Die Arbeit widmet sich dabei zunächst der Bereitstellung theoretischer Grundlagen (Definition, Geschichte, Funktionsweise, Herausforderungen, Aktuelle Lage) von Mensch-Maschine-Kommunikation unter Zuhilfenahme des Natural Language Processings. Darauf basierend werden Anforderungen aufgestellt, die eine softwaremäßige Implementierung eines solchen Systems beschreiben. Dies geschieht unter der Beachtung von Rahmenbedingungen durch den themenstellenden Betrieb dieser Arbeit und den Grenzen des Robotersystems. Es folgt eine ausgiebige Evaluierung verschiedener NLP-Systeme anhand des Abgleichs von Anforderungen und zusätzlicher Kennziffer-Messungen (Precision, Recall, F1 Score). In diesem Zusammenhang entsteht auch ein Testprogramm, welches ein automatisiertes Evaluierungsverfahren bereitstellt. Die beiden besten NLP-Systeme werden anschließend in ihrer Verwendung detailliert beschrieben. Eine prototypische Implementierung stellt die Verwendung eines NLP-Systems, zur Kommunikation von Mensch und Roboter, in einem konkreten Anwendungsfall unter Beweis. KW - Mensch-Maschine-Kommunikation KW - natural language processing KW - NAO Pepper Robot KW - Watson Assistant KW - Rasa Core NLU Y1 - 2018 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:526-opus4-10349 ER -