Claudia Morais Viana

Rapid urbanization and population growth have intensified interest in understanding how urban environments relate to people’s sentiments and emotions. In recent years, social media data have increasingly been used to investigate these relationships, while a wide range of analytical approaches has been employed to identify and quantify sentiment. This study presents a systematic review of the associations between characteristics of the urban environment and human sentiments and emotions, while also examining the methodological approaches used to assess these relationships. The review was reported in accordance with the PRISMA 2020 guidelines. From an initial set of 655 records, 59 studies met the inclusion criteria. These were published between 2018 and 2025, with most case studies conducted in China (64%), followed by North America (12%), with additional studies in Europe and other Asian contexts. Weibo, a Chinese microblog comparable to Twitter (now X), was the most frequently used data source. Machine-learning and deep-learning (ML/DL) approaches were the most frequently used for sentiment analysis, followed by lexicon-based, hybrid, and manual approaches. Findings reveal generally positive associations between green and blue spaces and positive sentiments, as well as with scenic landscapes, urban features that facilitate social interaction, and social activities, while industrial and transport environments tend to be associated with more negative sentiments. On the other hand, several spatial and density-related indicators show heterogeneous relationships with sentiments, with the association types varying across studies and urban contexts. External stressors are generally associated with negative sentiment, including air pollution, heat-related conditions, extreme precipitation, and signs of environmental degradation. However, some indicators, such as air temperature and noise, show mixed relationships. Higher levels of walkability and perceived safety are also associated with more positive sentiments. Overall, several recurring findings across the reviewed studies can guide urban planning, including improving access to green and blue spaces, enhancing spatial configuration and public infrastructure, ensuring adequate maintenance and accessibility,, reducing exposure to environmental stressors, and improving conditions around transport hubs and stations.

Historical document collections are increasingly available as digitised images and PDFs, but their conversion into reliable text remains affected by optical character recognition (OCR) errors, degraded pages, heterogeneous layouts, and domain-specific terminology. This study proposes a pathway-level framework for documenting and comparing conventional OCR, OCR followed by large language model (LLM)-assisted correction, and direct multimodal transcription. The framework is demonstrated using the Portuguese Agricultural and Forestry Surveys (1950–1958). A stratified validation sample of 45 pages was selected by visual quality, page type, and geographic coverage. Outputs were evaluated against manually verified reference transcriptions using content-normalised character error rate (CER) and word error rate (WER), document-condition analysis, paired statistical tests, and an entity-level semantic preservation assessment focused on place names, agricultural terms, and measurement expressions. Under the evaluated model and interface conditions, both LLM-based pathways produced lower mean CER and WER than the conventional OCR baseline, with the lowest values observed for direct multimodal transcription. Semantic preservation was also higher for the LLM-based pathways, although measurement expressions remained the most persistent risk, particularly in table-based pages. Downstream tasks were not directly evaluated. The findings support the framework as a method for validating text-extraction pathways before reuse, rather than establishing a universal ranking of tools.

This paper offers a comprehensive overview of academic research on sentiment analysis in urban built environments from 2000 to 2025. Based on data from the scientific database Scopus and drawing on bibliometric tools like Bibliometrix (R) and VOSviewer for performance analysis and scientific mapping, it identifies publication trends, key influential works, leading authors and institutions, funding sources, and thematic clusters. The final dataset comprises 1315 English-language documents authored by 3855 researchers across 160 sources, with a total of 14,058 citations worldwide. The academic production increased after 2009, peaking in 2025. Keyword and network analyses highlight central themes (and methodological approaches) to the study of sentiment analysis in urban built environments. These include social media platforms like Twitter/X, machine learning, smart cities, artificial intelligence, mental health, and urban planning. China, the USA, and India lead in publication output. Over the last twenty-five years, key publication outlets included Sustainability (Switzerland), Cities, and the International Journal of Environmental Research and Public Health, while the National Natural Science Foundation of China has been the main funder. The paper discusses how sentiment analysis can support urban planning and public health by linking environmental features to well-being and explores emerging methodological trends like deep learning, multimodal approaches, and context-aware models. Overall, it maps the field’s intellectual landscape and argues in future directions for human-centered, data-driven urban decision-making.

Identifying the flow regime of springs—whether perennial or intermittent—is essential for effective water resource management, especially in headwater regions. This study explores the integration of climatic and morphological data to classify spring flow regimes in first-order drainage basins in southern Brazil. Using Random Forest (RF) as the main classification algorithm, we analyzed 164 basins with field-verified spring regimes and evaluated 152 predictive variables derived from ERA5-Land climatic data and terrain morphology metrics, particularly the Index of Concentration of Roughness (RCI). The model achieved an average classification accuracy of 78.2% (ROC = 84%), surpassing traditional photointerpretation and automatic extraction methods. Variable importance analysis revealed that RCI, runoff, precipitation, and vegetation-related metrics were the most influential factors. The accuracy remained stable even when the number of variables was reduced to eight, indicating the robustness and computational efficiency of the model and importance of variables. Our findings confirm that basin area alone is insufficient to determine flow regime and highlight the relevance of local morphological features and climatic data in understanding local flow in basins. The proposed method offers valuable insights for conservation planning, spring protection, and water management policy development.

This paper offers a comprehensive overview of academic research on sentiment analysis in urban built environments from 1999 to 2024. Based on data from the scientific database Scopus and drawing on bibliometric tools like Bibliometrix (R) and VOSviewer for performance analysis and scientific mapping, it identifies publication trends, key influential works, leading authors and institutions, funding sources, and thematic clusters. The final dataset comprises 871 English‐language documents authored by 2,068 researchers across 307 sources in 70 countries, with a total of 5,642 citations worldwide. The academic production increased after 2009, peaking in 2024. Keyword and network analyses highlight central themes (and methodological approaches?) to the study of sentiment analysis in urban built environments. These include social media platforms like Twitter/X/X, machine learning, Natural Language Processing, smart cities, and tourism. China, the USA, and India lead in publication output. Over the last twenty-five years, key publication outlets include the International Journal of Environmental Research and Public Health, Cities, and Lecture Notes in Computer Science, while the National Natural Science Foundation of China is the most common funder. The paper discusses how sentiment analysis can support urban planning and public health by linking environmental features to well-being and explores methodological emerging trends like deep learning, multimodal approaches, and context-aware models. Overall, it maps the intellectual landscape of the field and argues for future directions for human-centred, data-driven urban decision-making.

Wildfire regimes in Mediterranean landscapes are becoming increasingly unpredictable, driven by the combined effects of climate change, land-use transitions, and socio-economic pressures. Traditional metrics such as burned area or ignition points often fail to capture the complexity of the temporal and spatial recurrence of fire events. To address this gap, we apply, for the first time, a trajectory analysis framework to wildfire occurrence data across mainland Portugal (1975–2024), using pixel-level binary time series at 100 m resolution. Originally developed for land cover change detection, this method classifies each pixel into sequences representing distinct temporal patterns (e.g., stability, gains, losses, or alternations) over defined periods. Results reveal a predominance of stable absence and alternation-type trajectories, particularly “All alternation gain first”, which points to recurrent yet irregular fire activity. Regional differences further highlight the influence of divergent socio-ecological contexts. The findings suggest that fire regimes in Portugal are not only recurrent but structurally dynamic, and that trajectory-based classification offers a novel and valuable tool for long-term monitoring and regionally adapted fire management. Applying this method to wildfire data required specific adjustments to account for the unique temporal and thematic characteristics of fire regimes, ensuring a meaningful interpretation of the results.

Accessibility to community pharmacies is crucial for ensuring timely access to medications and essential health services. While accessibility to community pharmacies is critical, disparities driven by temporal and spatial factors persist, resulting in inequities in healthcare access. This study aims to comprehensively assess spatiotemporal and multimodal accessibility to community pharmacies in Lisbon, highlighting the influence of transport modes and time of day on accessibility disparities. The study employed a methodology that considered five daily time slots and two modes of transport—walking and public transport—to evaluate accessibility to community pharmacies. Data was sourced from road and pedestrian networks, Google API, and GTFS data. Descriptive statistics and spatial analysis were utilized to assess travel time and accessibility disparities across different regions of Lisbon. The analysis focused on both the percentage of residents able to access pharmacies within 10 min and the total number of pharmacies accessible. ndings reveal significant temporal variations in accessibility, with public transport consistently improving access compared to walking. Accessibility peaks in the evening (6–7 PM), when 83.3% of residential buildings are within a 10-min walking distance of a pharmacy, and 92.7% are reachable by public transport. In contrast, early morning hours (4–5 AM) show the lowest accessibility, with only 8.9% of buildings accessible by walking and 16.1% by public transport. During the daytime (8–9 AM), notable disparities emerge across the city: public transport enhances access in the southwest, northwest, and central areas, while limited pharmacy opening hours constrain accessibility in the north and southeast, where only 108 of 258 pharmacies are operational. Finally, travel time to pharmacy services for city residents highlight significant spatial and temporal disparities in pharmacy accessibility, emphasizing the role of transport modes and service hours in shaping urban healthcare access. This study underscores the importance of addressing both temporal and spatial factors to ensure equitable accessibility to community pharmacies. The findings suggest the need for targeted policies to improve public transport services during off-peak hours and to extend pharmacy operating hours. Future research should focus on comparative studies across different urban contexts and incorporate more granular data to better understand accessibility to urban services.

Advances in digital sensors and Information flow have created an abundance of data generated by users under various emotional states in different situations. Although this opens up a new facet in spatial research, the large amount of data makes it difficult to analyze and obtain complete and comprehensive information leading to an increase in the demand for sentiment analysis. In this study, the Canadian National Research Council (NRC) of Sentiment and Emotion Lexicon (EmoLex) was used, based on data from the social network Twitter (now X), thus enabling the identification of the places in Lisbon where both positive and negative sentiment prevails. From the results obtained, the Portuguese are happy in spaces associated with leisure and consumption, such as museums, event venues, gardens, shopping centres, stores, and restaurants. The high score of words associated with negative sentiment have more bias, since the lexicon sometimes has difficulties to identify the context in which the word appears, ending up giving it a negative score (e.g., war, terminal).

Open access peer-reviewed chapter