The EU funded Exten(DT)2 project explores how Design Thinking (DT) and emerging technologies have the potential to transform education. This study explores how Swedish students harnessed DT with ChoiCo to design choice-based games tackling media and information literacy related socio-scientific issues (SSI). Students immersed themselves in decision-making scenarios, demonstrating a powerful blend of creative expression and technological application. The intervention highlighted novel aspects of DT, notably fostering abstract reasoning and prompting critical reflection on societal values within simulated contexts. This efforts underscores the potential of innovative educational approaches, directly aligning with the Exten(DT)2 project’s goal of advancing Digital Transformation in education. To better integrate DT into SSI in education, this study focuses on key themes aimed at promoting pedagogical innovative digital transformation.

In this paper we propose a comprehensive analytical pipeline that efficiently processes 1.5 million multilingual tweets about the Ukraine-Russia conflict, relying on multilingual NLP techniques and strategic sampling to produce interpretable sentiment and emotion visualizations. By incorporating temporal and geospatial metadata as multimodal information, our approach reveals complex emotional patterns: filtering predominantly neutral content highlights fear (often over 60%) and anger (10–25%) as dominant emotions, with temporal analysis showing negative sentiment surges during crises. Geographically, conflict zones are fear-dominant, while Western democracies and state-influenced systems exhibit mixedemotional profiles; visualizations further demonstrate how national data can obscure media outlet polarization. Crucially, we enhance explainability by analyzing confusion matrix patterns to identify emotionally ambiguous or surprising predictions, such as fear being frequently assigned to neutral or positive content indicating pervasive underlying emotional currents. The outcomes of our work offer advanced, multimodal, and interpretable tools for analyzing large-scale discursive phenomena.

As artificial intelligence (AI) becomes increasingly integrated into daily life, understanding how young people perceive and emotionally respond to this technology is crucial to leverage AI in novel ways to make education more meaningful and inclusive. This study explores K-12 students from South East Sweden and their sentiments toward AI by analyzing open-ended responses from 183 participants aged 9 to 16 years. Sentiment analysis of the collected data, using both VADER and TextBlob algorithms, revealed a broad spectrum of attitudes, ranging from enthusiastic optimism to explicit distrust. A linear regression analysis examined demographic predictors of sentiment scores, identifying a modest but significant effect of gender and age at first phone acquisition. Male K-12 students expressed more negative sentiments compared to female K-12 students, while students who received their first phone at an older age tended to hold slightly more positive views of AI. These findings indicate that factors such as early exposure to technology and gender influence adolescents' emotional responses to AI. This underscores the importance of addressing emotional and affective aspects in AI literacy programs and research. To deepen our understanding, future studies should broaden sentiment analysis to capture students' perspectives on ethical issues including fairness, bias, and accountability in AI systems. Such an approach will provide a more comprehensive view of how young people engage with both the opportunities and challenges presented by emerging AI technologies.

Background: The landscape of political international relations is undergoing a profound transformation, characterized by the re-emergence and strengthening of strategic alliances amidst escalating geopolitical tensions and evolving global polarities. This dynamic is fundamentally driven by an intense, ongoing competition for power and control over critical resources (Pekkanen, 2025). A compelling illustration of this shift is the fortified relationship between the U.S. and Saudi Arabia (S.A.) under President Trump’s renewed presidency. The U.S.-Saudi alliance holds profound geopolitical and economic significance, impacting global energy markets, regional stability, and international diplomacy, with Saudi Arabia actively seeking continued U.S. commitment to regional security amidst evolving challenges, as evidenced by President Trump's recent visit (Ebrahim & Al Lawati, 2025). Despite the critical nature of the U.S.- S.A. alliance, there's a significant gap in systematically understanding public perceptions, sentiments, and discourse surrounding this relationship, especially as expressed in the vast and dynamic landscape of social media comments (Dalipaj et al., 2024). The sheer volume and unstructured nature of these online interactions pose a substantial analytical challenge for traditional qualitative methods (Hewson, 2014). Consequently, a robust, scalable approach is distinctly needed to efficiently extract and interpret the nuanced public perceptions of these geopolitical dynamics across prominent digital platforms (Dalipaj et al., 2024).

This study aims to critically examine public perceptions and reactions to the evolving U.S.-Saudi Arabian relationship by leveraging text analytics on YouTube comments, thereby offering novel insights into how social media discourse shapes the understanding of crucial geopolitical dynamics.

Methods: Social media discourse from three YouTube channels were investigated by scraping 3300 comments using Python: SkyNews (1000 comments), Fox35Orlando (2000 comments), and Al-Arabiya (300 comments). It is worth noting that comments were collected at an earlier point, reflecting data volumes at that time. The extracted datasets then underwent preprocessing and were analyzed using transformer-based models for emotion-sentiment detection. Additionally, SpaCy was employed for lexical and Named Entity Recognition analysis. To identify key discussion themes, BERTopic was utilized, leveraging RoBERTa embeddings (Mehmood et al., 2024). Finally, these insights were subject to human expert analysis to interpret the mental representation and coherence of the discourse. This qualitative review enhanced our understanding of the empirical output from the digital analysis, culminating in a comparative analysis of public discourse across the different news sources (Dooley & Levinsohn, 2000; Mohammed et al., 2025).

Study Findings: The study analysis reveals distinct audience engagement patterns regarding US-Saudi relations. FOX35Orlando showed strongly positive sentiment (70%) dominated by celebratory Trump-focused discourse, reflecting its conservative viewership. Al-Arabiya presented a dual pattern: neutral sentiment (48%) coexisted with religiously framed positive commentary (32%), though nearly 20% of comments were off-topic. SkyNews balanced neutral (42%) and positive (41%) sentiment, with discourse equally divided between policy analysis and symbolic evaluation of leadership appearances. These patterns demonstrate how channel identity shapes public discourse. FOX35Orlando's emotionally congruent responses reflect partisan alignment, while Al-Arabiya's mixed profile suggests its dual role as news provider and cultural forum. SkyNews' hybrid engagement mirrors its international audience's diverse perspectives. This highlights how media ecosystems transform geopolitical reporting into participatory audience experiences, with platform-specific comment cultures emerging from institutional, demographic, and editorial factors.

Conclusion with Future plans: Our cross-channel analysis of YouTube commentary has yielded three principal insights about digital discourse on US-Saudi relations: First, we observe that news outlets function as ideological filters, with FOX35Orlando's celebratory Trump narratives (70% positive sentiment) contrasting sharply with Al-Arabiya's neutral-religious duality and SkyNews' balanced policy- appearance focus. Second, the study reveals how platform affordances enable distinct participation styles, from coherent partisan engagement to fragmented multi-topic discussions. Third, we identify emotional resonance as a key mechanism through which media channels cultivate audience loyalty, whether through joy- driven conservative solidarity or the more measured responses of international viewership. These findings advance media effects research by demonstrating how geopolitical narratives undergo participatory reformulation in digital comment spaces. The study paves way for three specific contributions: (1) methodological innovation in combining Sentiment-Emotion-Topic analysis, (2) empirical documentation of transnational opinion formation processes, and (3) theoretical development of platform-specific public diplomacy frameworks. Future studies should also investigate longitudinal patterns and cross-linguistic variations to further unpack the complex interplay between news framing and digital audience behavior. One of this study's deliverables is the provision of links to Python implementations via Colab notebooks, and a more detailed similar implementation with a larger dataset can be found in link facilitating reproducibility and serving as a foundation for future research.

Background: Digitalization has emerged as a key driver of circular construction, enhancing project efficiency, collaboration, and innovation across the construction value chain (Aghimien et al., 2018). Key Digital Technologies (DTs) such as Building Information Modeling (BIM), the Internet of Things (IoT), and Artificial Intelligence (AI) are essential in enabling more effective resource tracking, lifecycle management, and process optimization within the industry. (Ezeokoli et al., 2016; Adafin et al., 2021). However, the adoption of these DTs specifically within Nigeria’s construction sector remains limited. This is largely attributable to a confluence of factors, including prevailing economic constraints, weak government support, infrastructural deficits, low levels of digital literacy, insufficient training programs, and broader human-capital challenges (Moshood et al., 2020; Bello and Abdullahi, 2023; Bamgbose, Ogunbayo and Aigbavboa, 2024). While existing research acknowledges the transformative potential of DTs for circular construction, much of it often centers on identifying obstacles rather than exploring systemic enablers. As a result, discussions around how public and industry narratives influence adoption remain underdeveloped, leaving a significant gap.

Public discourse, critically influenced by socio-economic and environmental contexts, plays an indispensable role in shaping technology adoption across all sectors, including construction. Social media platforms, described by Fieseler and Fleck (2013) modern community infrastructures provide highly accessible avenues for such discourse, often surpassing the reach of traditional publications. However, despite their potential, these platforms are underutilized in facilitating dialogue around DT adoption and circularity within the construction domain (Willems and Alizadeh, 2015).

Aim: This study therefore addresses the identified gap by conducting a comparative analysis of narratives on DT adoption for circular construction in Nigeria, scraping data from both social media and construction company websites. The primary aim was to uncover prevailing themes, sentiments, and perspectives articulated across these distinct platforms. The study inquired into three core areas: the representation of DTs for circular construction on social media and company websites in Nigeria; the sentiments surrounding this discourse; and the contextual similarities and differences evident in the narratives exchanged across the two platforms.

Methodology: This study utilized a quantitative and comparative content analysis methodology, drawing data from both social media and construction company websites. The initial step involved compiling a comprehensive list of Nigerian construction companies, which was efficiently gathered using a browser extension and subsequently exported to an Excel spreadsheet. Each identified company's website was then meticulously reviewed for content related to DTs and circularity. However, recognizing the limited availability of dedicated blog sections on many sites, relevant textual posts were ultimately sourced from three particularly active websites. For the social media component, a total of 197 posts were scraped from Platform X (formerly Twitter) using targeted search terms. The collected data then underwent rigorous processing and analysis using Orange Data Miner software. Texts were initially imported as corpus and subjected to essential preprocessing steps, including tokenization, removal of stop-words, and lemmatization. To identify dominant themes within the dataset, word clouds were generated. Sentiment analysis was a crucial aspect of the study, conducted using two distinct models: the Ekman model, which identified six fundamental emotions, and the VADER model, which categorized sentiment polarity. The findings from this analysis were effectively visualized through box plots, heat maps, and linear projections, allowing for a thorough exploration of emotional trends and correlations within the data.

Results and Findings: The comparative analysis of thematic content derived from both social media (Platform X) and construction company websites unveiled distinct perspectives on digital transformation within circular construction. Across both platforms, shared foundational themes such as "construction," "projects," "infrastructure," and "sustainability" were consistently present, indicating a common understanding of the broader industry context. However, social media discourse demonstrated a broader emphasis on overarching DT terms, including "digital," "innovation," and "data," reflecting a more general engagement with the concept of digitalization. In contrast, company websites primarily utilized traditional industry terminology and conveyed a more explicit vision for industry progress, employing terms like "company," "estate," and "efficiency," which align with corporate objectives and operational improvements. Notably, specific DT tools (e.g., "AI," "BIM," "IoT") appeared with less frequency, particularly on social media, suggesting a more conceptual rather than tool-specific discussion among the public. Furthermore, social media conversations frequently referenced leadership, political dynamics, and economic challenges, thereby reflecting a wider spectrum of public concerns that extend beyond technical adoption.

Sentiment profiling, using the Ekman and VADER models, highlighted “joy” as a dominant emotion on both platforms indicating a general positive engagement with the topic. Social media discourse, while also displayed a notable presence of “fear” along with limited expression of “anger” and “sadness”, which suggested public apprehension, diverse viewpoints, and polarized opinions regarding feasibility, associated costs, and practical implementation challenges of DT in circular construction. Conversely, "Joy" overwhelmingly dominated the sentiments expressed on company website, with negative sentiments, including "fear," occurring minimally. This striking disparity indicated a largely positive, often controlled narrative from the industry. Such messaging seems strategically crafted to promote progress and appeal to stakeholders while downplaying potential drawbacks.

The heatmaps and linear projections confirmed these contrasts. Social media showed diverse emotional expressions, with scattered negative sentiment and clustered positive themes around DT’s potential. While website sentiment remained uniformly positive, with limited acknowledgement of challenges. This disparity highlights a communication gap: while the public expresses hope and skepticism, industry messaging remains polished and overly optimistic.

Conclusions and Recommendations: The study concluded that social media discourse serves as an effective conduit for capturing diverse sentiments regarding digital technology in circular construction, including enthusiasm for tools like IoT, BIM, and AI, alongside notable concerns about economic and regulatory barriers. Conversely, company websites primarily maintain positive, business-aligned narratives but often lack actionable strategies that explicitly link digital transformation to sustainability objectives. Despite these differences, both platforms reveal shared challenges, such as security and regulatory issues, indicating potential avenues for collaborative efforts. Therefore, there is need for more balanced, transparent industry communication that aligns with public sentiment to support DT adoption for circular construction in Nigeria as well as fostering open dialogue among stakeholders.

Limitations and Future Work: This study was limited by its narrow data scope, relying on a single social media platform and three company websites, and lacked direct stakeholder input. Future studies would expand on the data sources to include diverse textual materials and stakeholder interviews within the Nigerian construction industry.

In this abstract we present and discuss the central themes of a collaborative research project that examines the role of emotional mobilization in shaping democratic discourse and enhancing societal resilience amidst ongoing digital transformation. In today’s increasingly fragmented and hybrid media landscape, emotionally charged narratives—particularly those propagated by populist actors via alternative media and social platforms (Holt, 2023)—frequently challenge evidence-based communication, thereby contributing to heightened polarization and a decline in public trust. With the continued rise of populist and far-right political movements, it is crucial to understand how to elicit emotional engagement with active citizenship and forms of democratic deliberation. While emotional appeals are often portrayed as antithetical to rational deliberation, they may also serve as catalysts for democratic engagement, social cohesion, and civic participation. Our research questions how emotional narratives in the media may interact with and challenge evidence-based communication in democratic discourse, what role emotional mobilization has in shaping user engagement, polarization and trust in journalistic expertise, and how can an understanding of affective mobilization contribute to building more resilient, inclusive and trustworthy democratic communication practices.The core ideas that guide our research emerge from a collaboration between the A Questioned Democracy and Digital Transformations Knowledge Environments at Linnaeus University. Our research approach adopts an intermedial and cross-disciplinary methodology to analyse the interplay between emotional and factual appeals across diverse media ecologies (for the emotional turn in media studies, see Wahl-Jorgensen, 2019). Drawing on digital media content from Sweden—including alternative news sites, mainstream journalism, and social media platforms—the study combines computational methods (e.g., sentiment and network analysis), qualitative approaches (e.g., discourse and frame analysis), and participatory workshops to investigate how emotions and facts co-constitute public discourse (Acheampong et al, 2021).By moving beyond the conventional dichotomy between reason and emotion, the project seeks to advance a more nuanced understanding of affective dynamics in digital communication. In doing so, it contributes to ongoing debates on democratic resilience by examining how emotional mobilization can both distort and strengthen deliberative processes. The broader objective is to inform the design of inclusive, trustworthy, and emotionally attuned communication strategies—particularly within a European context characterized by epistemic uncertainty, algorithmic amplification, and rising populist influence (Moffitt, 2016).Although the project ideas described above are in currently in their early stages, this presentation will outline its conceptual underpinnings, research design, and collaborative framework. Emphasis will be placed on the methodological challenges and opportunities associated with studying affect, evidence, and trust in an intermedial, data-rich environment. The expected outcomes range from strengthening democratic resilience through improved media practices, informing public communication strategies that balance reason and affect, and developing educational resources and policy recommendations to enhance societal trust in democratic institutions and knowledge systems.

As artificial intelligence (AI) becomes increasingly embedded in everyday life and educational systems, understanding how students conceptualize AI is essential for designing responsive and inclusive AI literacy curricula. This study investigates the perceptions of AI among Swedish students aged 13-16, using a qualitative survey instrument comprising background questions and five open-ended questions adapted from previous research on student conceptions of AI. Using a BERT-based topic modelling approach, we analyzed responses from 75 students, identifying nine key themes that reflect how students think about, relate to, and emotionally react to AI technologies. The identified topics, ranging from functional views of AI as a tool or helper, to more anthropomorphic notions like robots and AI being "just like"humans, to expressions of uncertainty and mixed emotions, highlight the diversity and complexity of student perceptions. We compare these findings with previous data collected in Azerbaijan to examine cultural and contextual differences in the way young learners understand AI. This research contributes to the growing body of work on AI literacy by providing evidence-based insights into student conceptions and the sociocultural shaping of AI understanding.

Recent advances in Information and Communication Technologies (ICT), especially AI-powered tools, are driving innovation and productivity across academic and industrial sectors. These technologies are also transforming education by enabling personalized learning, redefining teaching methods, and enhancing educational outcomes. Given the transformative potential of Artificial Intelligence in Education (AIEd), it is essential to examine its benefits and implications for the future of education. AI tools offer substantial benefits to various educational stakeholders by enhancing cognitive engagement, improving teaching assessments, and fostering digital literacy. Additionally, they support policy development, address ethical considerations, and increase educational efficiency and accessibility. However, despite AI’s potential in education, the rapid pace of technological change can create uncertainty and overwhelm, even among experts. As AIEd advances, it is crucial to develop strategies that help educators and institutions stay updated on new developments and integrate them effectively into educational settings. This study explores the historical development of AIEd and identifies current research trends through a dual-method approach. First, a bibliometric analysis using VOSViewer uncovers patterns in AIEd research over the past 50 years. Second, text mining techniques using term frequency analysis and topic modeling—examine over 2,204 abstracts from Scopus, covering publications from 1970 to the present. By combining these methods, the study highlights key AIEd research areas that impact stakeholders and influence future educational practices. Understanding the evolution of AIEd and current trends is essential for anticipating its future trajectory, preparing us to address emerging challenges and opportunities in education. The findings trace the progression of AIEd research, revealing shifts in focus, underlying learning theories, and application paradigms. This comprehensive overview helps identify research gaps and emerging themes, supporting the development of strategies to leverage AIEd’s potential in transforming education and meeting the complex, evolving needs of twenty-first century stakeholders.