Towards Labour Market Intelligence through Topic Modelling

The “Innovating the Use of Labour Market Intelligence within European Universities” (LMI-EUniv) project, set within the Erasmus+ programme’s Key Action 2, represents a pioneering effort to harness Labour Market Intelligence (LMI) for enhancing the alignment between higher education offerings and labour market demands across Europe. This initiative, led by a consortium including the West University of Timisoara, University of Milano-Bicocca, University of Tallinn, Prospektiker, and the Luxembourg Institute of Socio-Economic Research, aimed to explore the current utilisation of labour market information and intelligence by European universities in planning and delivering their educational provisions. With a focus on fostering the congruence of educational supply with labour market demand through innovative learning and teaching methodologies, the project sought to empower Higher Education Institutions (HEIs) with the skills and competencies necessary to meet these challenges. By mapping essential LMI sources at a national level, examining the use of LMI across European HEIs, and developing a comprehensive training course and an online Labour Market Intelligence Hub, this project aspired to create an indispensable reference for HEIs. This article synthesises the project’s key findings, underscoring the critical role of labour market intelligence in adapting academic curricula to meet specific labour market needs, thereby contributing to the broader discourse on the integration of LMI in higher education and its implications for curriculum development, graduate employability, and the overarching alignment between education and labour market expectations.

Labor Market Intelligence (LMI) is an emerging field of study that has been gaining interest as it allows employing Artificial Intelligence (AI) algorithms on labor market information. The goal of LMI is to support decision and policy making activities (e.g., real-time monitoring of Online Job Vacancies (OJV) across countries, forecast skill requested within vacancies, compare similar labor markets across borders, etc.). The European project in which this work is framed can be placed in this field, as it aims at collecting and classifying millions of OJVs from 28 EU Countries, handling 32 languages, and also extracting the requested skills. The result is a huge amount of information useful for understanding labor market dynamics and trends. The goal of this work is to realize a system - namely GraphLMI - that organizes such Labor Market information as a graph, enabling the representation of occupation/skill relevance and similarity over the European Labor Market; another goal is to enrich the European standard taxonomy of occupations and skills (ESCO) to better fit the labor market expectations. We formalize and design the GraphLMI data model, then we implement it as a graph-database, generated by processing 5.3+ million OJVs composed by free text and collected between 2018 and 2019 for France, Germany, and the United Kingdom. Finally, we show how the resulting knowledge can be queried through a declarative query language to understand, compare and evaluate country-based labor market dynamics for supporting policy and decision making activities at European level.

The aim of this paper is to introduce a research project dedicated to identifying gaps in green skills by using the labor market intelligence. Labor Market Intelligence (LMI). The method is primarily descriptive and conceptual, as the authors of this paper intend to develop a theoretical background and justify the planned research using Natural Language Processing (NLP) techniques. This research highlights the role of LMI as a tool for analysis of the green skills gaps and related imbalances. Due to the growing demand for eco-friendly solutions, there arises a need for the identification of green skills. As societies shift towards eco-friendly economic models, changes lead to emerging skill gaps. This study provides an alternative approach for identification of these gaps based on analysis of online job vacancies and online profiles of job seekers. These gaps are contextualized within roles that businesses find difficult to fill due to a lack of requisite green skills. The idea of skill intelligence is to blend various sources of information in order to overcome the information gap related to the identification of supply side factors, demand side factors and their interactions. The outcomes emphasize the urgency of policy interventions, especially in anticipating roles emerging from the green transition, necessitating educational reforms. As the green movement redefines the economy, proactive strategies to bridge green skill gaps are essential. This research offers a blueprint for policymakers and educators to bolster the workforce in readiness for a sustainable future. This article proposes a solution to the quantitative and qualitative mismatches in the green labor market.

With the advances in natural language processing and big data analytics, the labor market community has introduced the emerging field of Labor Market Intelligence (LMI). This field aims to design and utilize Artificial Intelligence (AI) algorithms and frameworks to analyze data related to the labor market information for supporting policy and decision-making. This paper elaborates on the automatic classification of free-text Web job vacancies on a standard taxonomy of occupations. In achieving this, we draw on well-established approaches for extracting textual features, which subsequently are employed for training machine learning algorithms. The training and evaluation of our machine learning models were performed with data extracted from online sources, pre-processed, and hand-annotated following the ISCO taxonomy. The results showed that the proposed model is very promising. The advantage is its simplicity. After its application to a relatively small and difficult to clean dataset, it achieved a good accuracy. Furthermore, in this paper we discuss how real-life applications for skill anticipation and matching could benefit from our approach.KeywordsNatural language processingLabor marketISCO taxonomy prediction

Las decisiones sobre el aprendizaje y el trabajo deben ubicarse en un contexto particular espacial, de mercado laboral: los individuos toman decisiones dentro de "estructuras de oportunidad" particulares y sus decisiones y aspiraciones se enmarcan dentro de su comprensión de tales estructuras. Este artículo examina formas en las que se puede aplicar el aprendizaje sobre carreras con datos abiertos e inteligencia del mercado laboral. Un estudio de caso ilustrativo del proyecto LMI for All en el Reino Unido muestra la viabilidad técnica del diseño y desarrollo de dichos sistemas y un modelo para su difusión e impacto. La tendencia hacia los datos abiertos y las aplicaciones cada vez más poderosas para procesar y consultar datos han cobrado impulso. Esto, combinado con la necesidad de información sobre el mercado laboral para la toma de decisiones en mercados laborales cada vez más inestables, ha llevado al desarrollo y pilotaje de nuevos sistemas de información del mercado laboral (LMI), que involucran a múltiples grupos de usuarios. Existen desafíos universales debido al uso cada vez mayor de LMI, especialmente en la asignación de empleos y el uso en rápida expansión de datos abiertos en diferentes entornos de educación y empleo. Destacamos seis temas emergentes que deben abordarse para que los datos abiertos y la inteligencia del mercado laboral puedan aplicarse de manera efectiva en diferentes contextos y entornos. Concluimos reflexionando sobre la urgente necesidad de ampliar el cuerpo de investigación y desarrollar nuevos métodos de co-construcción en asociaciones de colaboración innovadoras.

Artificial intelligence in interdisciplinary life science and drug discovery research.

The rapid development of artificial intelligence is transforming social and economic systems, reshaping labor markets, and redefining the competencies required for future professions. As automation, machine learning, and data-driven decision-making become integral parts of organizational practice, the traditional education system faces increasing pressure to adapt. Contemporary learners must acquire new interdisciplinary skills that combine technological literacy, creativity, analytical thinking, and adaptability. Higher education institutions are therefore required to anticipate the changing needs of students and employers, while designing academic programs that reflect long-term labor market trends. Artificial intelligence has the potential to support this transformation by identifying emerging professional fields, forecasting future skill requirements, and enabling personalized educational pathways. Integrating AI into educational planning creates opportunities to redesign curricula, enhance institutional competitiveness, and align national education strategies with global technological change. The objective of this study is to examine how artificial intelligence can support the development of innovative educational proposals and contribute to shaping professions of the future. The study aims to investigate the mechanisms through which AI-based tools can identify labor market trends, forecast competency needs, and optimize the design of new academic programs. It also seeks to determine how AI-driven insights can support strategic decision-making within universities and training institutions. The research applies a mixed-methods approach combining horizon scanning, expert assessment, and analysis of AI-driven forecasting tools. Horizon scanning techniques were used to identify emerging technological, economic, and social trends affecting future professions. Expert interviews with university administrators, HR specialists, and digital transformation experts provided qualitative insights into how AI can guide curriculum design. The study also employed scenario analysis based on data from global platforms such as the OECD Skills Outlook, LinkedIn Workforce Reports, and various AI-powered labor market analytics systems. The methodological framework included content analysis of academic literature on AI in education, as well as case studies of institutions that have adopted AI-driven planning models. The results indicate that artificial intelligence provides substantial opportunities for shaping new educational proposals and responding to the challenges of the digital economy. First, AI algorithms are capable of analyzing large volumes of labor market data to detect emerging professions and forecast demand for new competencies with high accuracy. Second, AI-supported curriculum design allows universities to identify content gaps, reorganize learning trajectories, and create interdisciplinary programs aligned with future workforce needs. Third, personalized learning systems improve access to education by recommending individual skill pathways based on learners’ abilities, interests, and career ambitions. Case studies demonstrate that institutions implementing AI-assisted program planning experience increased student enrollment, higher adaptability to labor market changes, and greater academic innovation. Moreover, AI-supported career simulations and predictive analytics help institutions model potential professional scenarios and design programs relevant to technologically advanced sectors such as robotics, cybersecurity, digital ethics, and sustainable technologies. Artificial intelligence plays a significant role in forming educational strategies aimed at preparing future professionals. Its ability to process complex data and identify long-term trends makes it a powerful tool for designing new academic programs and ensuring the relevance of educational proposals. AI enhances the capacity of universities to anticipate labor market evolution, promote innovation, and implement flexible learning models. Its integration into educational planning strengthens institutional competitiveness and contributes to national human capital development. Further research should investigate the ethical and governance implications of using AI in educational program design, including issues of transparency, bias, and data ownership. Longitudinal studies are needed to evaluate the real impact of AI-generated educational proposals on graduates’ labor market outcomes. Additionally, future research should explore how AI can support international cooperation in shaping global standards for professions of the future.

Artificial intelligence applied to medicine: There is an “elephant in the room”

The rapid growth of Web usage for advertising job positions provides a great opportunity for real-time labour market monitoring. This is the aim of Labour Market Intelligence (LMI), a field that is becoming increasingly relevant to EU Labour Market policies design and evaluation. The analysis of Web job vacancies, indeed, represents a competitive advantage to labour market stakeholders with respect to classical survey-based analyses, as it allows for reducing the time-to-market of the analysis by moving towards a fact-based decision making model. In this paper, we present our approach for automatically classifying million Web job vacancies on a standard taxonomy of occupations. We show how this problem has been expressed in terms of text classification via machine learning. Then, we provide details about the classification pipelines we evaluated and implemented, along with the outcomes of the validation activities. Finally, we discuss how machine learning contributed to the LMI needs of the European Organisation that supported the project.

Classifying online Job Advertisements through Machine Learning

Rapid transformations in the labor market, driven by digitalization, automation, and artificial intelligence, have intensified skill mismatches and workforce instability across urban economies. Traditional labor analytics platforms rely on static reports and siloed indicators, limiting their ability to support strategic workforce planning and policy intervention. To address these challenges, this paper proposes Skill-Pulse AI, an integrated, intelligence-driven decision support system for real-time labor market analysis, forecasting, and reskilling optimization. The proposed system employs a hybrid analytical architecture that combines geospatial intelligence, time-series demand forecasting, anomaly detection, graph-based career path optimization, and semantic resume analysis within a unified Streamlit-based interface. Market dynamics are modeled using synthetic-real hybrid data, enhanced with seasonality-aware neural forecasting and Isolation Forest–based risk detection. Network graph algorithms are applied to compute optimal reskilling pathways, while policy simulations and persona-based analytics enable both job-seeker and institutional decision support. Experimental evaluation demonstrates that Skill-Pulse AI effectively identifies high-demand skill clusters, talent supply gaps, and labor market risk zones with improved interpretability and responsiveness compared to conventional systems. Forecasting modules capture short-term demand trends with scenario-based projections, while reskilling recommendations reduce transition cost by optimizing salary and skill distance metrics. The system further enables quantitative policy impact assessment through grant-to-employment ROI simulations. The implications of this work extend to government agencies, educational institutions, enterprises, and individual professionals, providing actionable intelligence for workforce resilience, equitable labor mobility, and strategic investment planning. By integrating multiple analytical lenses into a single platform, Skill-Pulse AI supports evidence-based decision-making in complex labor ecosystems. Despite its effectiveness, the current implementation relies partially on simulated market signals and city-level aggregation, which may limit fine-grained sectoral accuracy. Future enhancements will incorporate real-time job portal feeds, deep learning–based semantic embeddings, and cross-country labor mobility modeling to further strengthen predictive precision and global applicability. Keywords: - Labor Market Intelligence; Workforce Analytics; Skill Gap Analysis; AI-Based Demand Forecasting; Reskilling Recommendation System; Geospatial Labor Analysis; Policy Impact Simulation; Anomaly Detection; Career Path Optimization

In the continuously changing labor market, understanding the dynamics of online job postings is crucial for economic and workforce development. With the increasing reliance on Online Job Portals, analyzing online job postings has become an essential tool for capturing real-time labor-market trends. This paper presents a comprehensive methodology for processing online job postings to generate labor-market intelligence. The proposed methodology encompasses data source selection, data extraction, cleansing, normalization, and deduplication procedures. The final step involves information extraction based on employer industry, occupation, workplace, skills, and required experience. We address the key challenges that emerge at each step and discuss how they can be resolved. Our methodology is applied to two use cases: the first focuses on the analysis of the Greek labor market in the tourism industry during the COVID-19 pandemic, revealing shifts in job demands, skill requirements, and employment types. In the second use case, a data-driven ontology is employed to extract skills from job postings using machine learning. The findings highlight that the proposed methodology, utilizing NLP and machine-learning techniques instead of LLMs, can be applied to different labor market-analysis use cases and offer valuable insights for businesses, job seekers, and policymakers.

The rapid advancements in artificial intelligence and machine learning have led to the development of highly sophisticated models capable of superhuman performance in a variety of tasks. However, the increasing complexity of these models has also resulted in them becoming "black boxes", where the internal decision-making process is opaque and difficult to interpret. This lack of transparency and explainability has become a significant barrier to the widespread adoption of these models, particularly in sensitive domains such as healthcare and finance. To address this challenge, the field of Explainable AI has emerged, focusing on developing new methods and techniques to improve the interpretability and explainability of machine learning models. This review paper aims to provide a comprehensive overview of the research exploring the combination of Explainable AI and traditional machine learning approaches, known as "hybrid models". This paper discusses the importance of explainability in AI, and the necessity of combining interpretable machine learning models with black-box models to achieve the desired trade-off between accuracy and interpretability. It provides an overview of key methods and applications, integration techniques, implementation frameworks, evaluation metrics, and recent developments in the field of hybrid AI models. The paper also delves into the challenges and limitations in implementing hybrid explainable AI systems, as well as the future trends in the integration of explainable AI and traditional machine learning. Altogether, this paper will serve as a valuable reference for researchers and practitioners working on developing explainable and interpretable AI systems. Keywords: Explainable AI (XAI), Traditional Machine Learning (ML), Hybrid Models, Interpretability, Transparency, Predictive Accuracy, Neural Networks, Ensemble Methods, Decision Trees, Linear Regression, SHAP (Shapley Additive Explanations), LIME (Local Interpretable Model-agnostic Explanations), Healthcare Analytics, Financial Risk Management, Autonomous Systems, Predictive Maintenance, Quality Control, Integration Techniques, Evaluation Metrics, Regulatory Compliance, Ethical Considerations, User Trust, Data Quality, Model Complexity, Future Trends, Emerging Technologies, Attention Mechanisms, Transformer Models, Reinforcement Learning, Data Visualization, Interactive Interfaces, Modular Architectures, Ensemble Learning, Post-Hoc Explainability, Intrinsic Explainability, Combined Models

As the attack types become more sophisticated, the ones in use today are losing their touch due to various reasons. Chief among these include zero-day exploits, AI-driven phishing, and polymorphic malware. This study explores incorporating artificial intelligence (AI) in cyber security frameworks to counter such threats, thereby proposing to shift the focus from reactive to proactive and adaptive mechanisms. It employs machine learning (ML) algorithms, neural networks, and natural language processing (NLP) to show how AI can better threat detection, automate incident response, and predict vulnerabilities in real-time. A new AI-based framework marries supervised learning for anomaly detection, reinforcement learning for adaptive protocol optimization, and generative adversarial networks (GANS) to simulate and counter advanced persistent threats (APTs).A set of examples is provided that validates the real-life functionality of the proposed framework in NIDS and cloud security environments and reveals a 40% speed improvement in threat identification and a 35% decrease in false positives compared to rule-based systems. Simultaneously, the study also deals with other ethical and operational issues such as adversarial attacks on AI models, privacy of valid data, and the "black box" problem of ML in decision-making. Using explainable AI (XAI) techniques and federated learning for distributed data processing, the proposed framework contends with the balancing act between transparency and robust security.This study presents the potential of AI to craft self-healing, context-sensitive cyber security infrastructures and summons standard regulatory guidelines governing AI on critical systems. The findings performed aim to empower departments to adopt intelligent, scale able defenses as the cyber warfare continues escalating. Keywords: AI-Driven Cyber security, Proactive Threat Detection, Adaptive Security Frameworks, Explainable AI (XAI), Machine Learning in Intrusion Detection

In recent years, the field of Artificial Intelligence (AI) and Machine Learning (ML) has witnessed remarkable advancements, revolutionizing various industries and domains. The proliferation of data availability, computational power, and algorithmic innovations has propelled the development of highly sophisticated AI models, particularly in the realm of Deep Learning (DL). These DL models have demonstrated unprecedented levels of accuracy and performance across a wide range of tasks, including image recognition, natural language processing, and complex decision-making. However, amidst these impressive achievements, a critical challenge has emerged - the lack of interpretability.Highly accurate AI models, including DL models, are often referred to as black boxes because their internal workings and decision-making processes are not readily understandable to humans. While these models excel in generating accurate predictions or classifications, they do not provide clear explanations for their reasoning, leaving users and stakeholders in the dark about how and why specific decisions are made. This lack of interpretability raises concerns and limits the trust that humans can place in these models, particularly in safety-critical or high-stakes applications where accountability, transparency, and understanding are paramount.To address the challenge of interpretability, Explainable AI (xAI) has emerged as a multidisciplinary field that aims to bridge the gap in understanding between machines and humans. xAI encompasses a collection of methods and techniques designed to shed light on the decision-making processes of AI models, making their outputs more transparent, interpretable, and comprehensible to human users.The main objective of this paper is to enhance the explainability of AI-based systems that involve user interaction by employing various xAI methods. The proposed approach revolves around a comprehensive ML workflow, beginning with the utilization of real-world data to train a machine learning model that learns the behavior of a simulated driver. The training process encompasses a diverse range of real-world driving scenarios, ensuring that the model captures the intricacies and nuances of different driving situations. This training data serves as the foundation for the subsequent phases of the workflow, where the model's predictive performance is evaluated.Following the training and testing phases, the predictions generated by the ML model are subjected to explanation using different xAI methods, such as LIME (Local Interpretable Model-Agnostic Explanations) and SHAP (SHapley Additive exPlanations). These xAI methods operate at both the global and local levels, providing distinct perspectives on the model's decision-making process. Global explanations offer insights into the overall behavior of the ML model, enabling a broader understanding of the patterns, relationships, and features that the model deems significant across different instances. These global explanations contribute to a deeper comprehension of the decision-making process employed by the model, allowing users to gain insights into the underlying factors driving its predictions.In contrast, local explanations offer detailed insights into specific instances or predictions made by the model. By analyzing these local explanations, users can better understand why the model made a particular prediction in a given case. This granular analysis facilitates the identification of potential weaknesses, biases, or areas for improvement in the model's performance. By pinpointing the specific features or factors that contribute to the model's decision in individual instances, local explanations offer valuable insights for refining the model and enhancing its accuracy and reliability.In conclusion, the lack of explainability in AI models, particularly in the realm of DL, presents a significant challenge that hinders trust and understanding between machines and humans. Explainable AI (xAI) has emerged as a vital field of research and practice, aiming to address this challenge by providing methods and techniques to enhance the interpretability and transparency of AI models. This paper focuses on enhancing the explainability of AI-based systems involving user interaction by employing various xAI methods. The proposed ML workflow, coupled with global and local explanations, offers valuable insights into the decision-making processes of the model. By unraveling the scenario-based behavior of a self-learning function with user interaction, this paper aims to contribute to the understanding and interpretability of AI-based systems. The insights gained from this research can pave the way for enhanced user trust, improved model performance, and further advancements in the field of explainable AI.