Geo-tagged Social Media Data Research Articles

Leveraging social media and community science data for environmental niche models: A case study with native Australian bees

Museum occurrence records are popular sources of information for creating Environmental Niche Models (ENMs), which allow the mapping of the potential niche ranges of species. Occurrence data is often downloaded en masse from established databases. However, the use of non-traditional data sources, such as occurrence records from community/citizen science outreach and social media, is increasing in use and abundance. Data from non-traditional data sources are potentially valuable records of information, particularly for species where museum occurrence records may be comparatively scarce. In the current study, we aimed to determine the impact of adding occurrence data from non-traditional databases to ENMs that were originally created using traditional databases with a group of comparatively understudied species, native Australian bees. We used the Maxent algorithm to model the potential environmental niches of eight species. We created three models for each species: 1) one consisting of only location data from museum specimen collection records from the Atlas of Living Australia (ALA) (a traditional database), 2) one combining ALA and geo-tagged social media (Flickr) data, and 3) a model combining ALA and geo-tagged community science data from iNaturalist. This resulted in 24 different models. By comparing the models produced from each of the augmented data sets with the traditional species data set (ALA vs. ALA & Flickr; ALA vs. ALA & iNaturalist) we showed that there were significant differences, not only in predicted ranges, but also in the weighting of environmental variables used by the models to predict the environmental niche. Differences were more greatly influenced by the geographic location of the extra occurrences rather than the number of additional occurrence points. We demonstrate the potential value and risks of including social media and community science geo-tagged image data in supplementing knowledge of species distributions, particularly for relatively under-sampled species such as native bees.

From Tweets to Energy Trends (TwEn): An exploratory framework for machine learning-based forecasting of urban-scale energy behavior leveraging social media data

Understanding energy behavior is crucial in addressing climate change, yet the accuracy of energy predictions is often limited by reliance on oversimplified occupancy data. This study develops an exploratory framework, from Tweets to Energy Trends (TwEn), leveraging machine learning and geo-tagged social media data to investigate the social dynamics of urban energy behavior. TwEn explores the relationship between social media interactions, specifically the frequency of tweets using data from the X Platform, and energy use patterns on an hourly basis. Employing various machine learning models, including artificial neural networks (ANN), decision tree (DTREE), random forest (RDF), and gradient boosting machine (GBM), the study evaluates their efficiency in both static and time-series forecasting of energy use trends and investigates the capability of social media data in predicting urban energy patterns. In addition, the study carries out a series of sensitivity analyses to provide an examination of the data and models. Furthermore, comprehensive data acquisition methods are developed and implemented. Tested on New York City using actual hourly electricity consumption data, the framework demonstrates significant predictive power of tweet frequency on urban electricity use. The framework also exhibits significant seasonality in X data, identifying patterns and trends that can inform time series urban building energy models (UBEM). The results offer new insights into the determinants of urban energy behavior and provide crucial perspectives for augmenting UBEMs, ensuring they are closely aligned with the complex social dynamics of contemporary urban environments. By integrating both digital and physical data, this study sheds light on urban energy behavior, supporting the formulation of effective and sustainable energy policies for urban futures.

Feeling the heat? Analyzing climate change sentiment in Spain using Twitter data

To shed light on the recent debate about climate change in this post-pandemic scenario, we take advantage of a unique dataset that combines geo-tagged social media data from Twitter in Spain from 2017 to 2022. Twitter conversations have been analyzed with natural language processing techniques to obtain sentiment scores related to climate change. These were merged with additional relevant control variables, aiming to understand the role of the contributing factors on the evolution of the hedonic scores, including external temperatures, the occurrence of heat waves, and deaths related to climate. We find a strong negative effect of external temperatures on sentiment, aggravated by recent increases in the frequency of heat waves and deaths related to climate. Further, this negative sentiment is accentuated after experiencing the recent COVID-19.

Exploring Tourist Flow Patterns through Geotagged Social Media Data: A Case Study from Sri Lanka

Tourism studies often rely on conventional methods such as interviews and site-specific surveys to collect data on tourist behaviour, mobility patterns, and preferences. However, these methods can be expensive, time-consuming, and limited in scope. Additionally, accurately tracing tourist travel paths can be challenging due to recall bias. Recently, user generated content on social media platforms have emerged as an alternative data source in tourism studies. In this study, we used geotagged posts on Flickr to understand the dominant paths taken by inbound tourists in Sri Lanka. The methodology consisted of three steps. First, geotagged photographs were collected from the Flickr API spanning a ten-year period, from 2012 to 2022. These photographs included metadata such as user ID, the timestamp of photo capture, and geo coordinates indicating where the photo was taken. Second, a density-based clustering algorithm was utilized to identify tourist hotspots. Finally, the Markov Chain model was employed to calculate transition probabilities among different attractions, revealing dominant travel routes within the country. Study’s findings indicate that Cultural heritage attractions were the most popular, comprising 55% of all attractions identified by the algorithm, and are particularly popular in districts such as Anuradhapura, Polonnaruwa, Matale, Kandy, and Galle. Nature-based attractions, constituting 37% of the total, were mainly located in Nuwara Eliya and Badulla Districts, as well as Yala National Park in Hambantota District. For coastal tourism, Galle and Matara Districts were the top preferences. The analysis confirms dominant travel patterns between North Central, Central, and Southwest coast regions. Notably, Galle, Kandy, Matale, Matara, Nuwara Eliya, and Colombo attract tourists from a more diverse array of regions compared to other districts in the country, showcasing their significance as tourist hubs. The study findings highlight opportunities in Jaffna for cultural-heritage tourism, the Eastern coast for coastal tourism, and the central highlands for tea tourism. It also emphasizes the need to develop the less-visited natural sites to ease pressure on popular National Parks. This research's significance lies in its contribution to informed decision-making and the sustainable management of tourist destinations. By understanding tourist mobility patterns and identifying popular attractions through social media data, policymakers can effectively manipulate visitor flows and mitigate excessive tourist pressure, while preserving the authenticity and allure of these destinations.
 
 Keywords: Sustainable tourism, Geotagged photographs, Social media, Tourist mobility, Flickr

ITourSPOT: a context-aware framework for next POI recommendation in location-based social networks

ABSTRACT The rising prosperity of Location-based Social Networks (LBSNs) witnessed an explosion in the availability of geo-tagged social media data, which enables tremendous location-aware online services, especially next point of interest (POI) recommendation. However, previous next POI recommendation studies usually adopt fix-length time windows for user check-in sequence modeling, leading to a limited capacity in capturing fine-grained user temporal preferences that easily change over time. Besides, existing methods often directly leverage multi-modal contexts as auxiliary to alleviate the data sparsity issue, which fails to fully exploit the sequential patterns of contextual information for inferring user interest drift. To address the above challenges, we propose a novel framework named iTourSPOT which extends traditional collaborative filtering methods with a context-aware POI embedding architecture. For enhancing temporal interests modeling capacity, we associate the context feature extraction with varying-length sessions and incorporate check-in frequencies of POIs as prior knowledge to instruct the session representation learning of our model. Moreover, a collaborative sequence transduction model is designed for joint context sequence modeling and session-based POI recommendation. Experimental results on a real-world geo-tagged photo dataset clearly demonstrate the effectiveness of the proposed framework when compared with state-of-the-art baseline methods, especially in both sparse and cold-start scenarios.

SocialMedia2Traffic: Derivation of Traffic Information from Social Media Data

Traffic prediction is a topic of increasing importance for research and applications in the domain of routing and navigation. Unfortunately, open data are rarely available for this purpose. To overcome this, the authors explored the possibility of using geo-tagged social media data (Twitter), land-use and land-cover point of interest data (from OpenStreetMap) and an adapted betweenness centrality measure as feature spaces to predict the traffic congestion of eleven world cities. The presented framework and workflow are termed as SocialMedia2Traffic. Traffic congestion was predicted at four tile spatial resolutions and compared with Uber Movement data. The overall precision of the forecast for highly traffic-congested regions was approximately 81%. Different data processing steps including ways to aggregate data points, different proxies and machine learning approaches were compared. The lack of a universal definition on a global scale to classify road segments by speed bins into different traffic congestion classes has been identified to be a major limitation of the transferability of the framework. Overall, SocialMedia2Traffic further improves the usability of the tested feature space for traffic prediction. A further benefit is the agnostic nature of the social media platform’s approach.

Do You Believe Your (Social Media) Data? A Personal Story on Location Data Biases, Errors, and Plausibility as Well as Their Visualization.

We present a case study on a journey about a personal data collection of carnivorous plant species habitats, and the resulting scientific exploration of location data biases, data errors, location hiding, and data plausibility. While initially driven by personal interest, our work led to the analysis and development of various means for visualizing threats to insight from geo-tagged social media data. In the course of this endeavor we analyzed local and global geographic distributions and their inaccuracies. We also contribute Motion Plausibility Profiles-a new means for visualizing how believable a specific contributor's location data is or if it was likely manipulated. We then compared our own repurposed social media dataset with data from a dedicated citizen science project. Compared to biases and errors in the literature on traditional citizen science data, with our visualizations we could also identify some new types or show new aspects for known ones. Moreover, we demonstrate several types of errors and biases for repurposed social media data. Please note that people with color impairments may consider our alternative paper version.

Negative Associations in Word Embeddings Predict Anti-black Bias across Regions-but Only via Name Frequency.

The word embedding association test (WEAT) is an important method for measuring linguistic biases against social groups such as ethnic minorities in large text corpora. It does so by comparing the semantic relatedness of words prototypical of the groups (e.g., names unique to those groups) and attribute words (e.g., 'pleasant' and 'unpleasant' words). We show that anti-Black WEAT estimates from geo-tagged social media data at the level of metropolitan statistical areas strongly correlate with several measures of racial animus-even when controlling for sociodemographic covariates. However, we also show that every one of these correlations is explained by a third variable: the frequency of Black names in the underlying corpora relative to White names. This occurs because word embeddings tend to group positive (negative) words and frequent (rare) words together in the estimated semantic space. As the frequency of Black names on social media is strongly correlated with Black Americans' prevalence in the population, this results in spuriously high anti-Black WEAT estimates wherever few Black Americans live. This suggests that research using the WEAT to measure bias should consider term frequency, and also demonstrates the potential consequences of using black-box models like word embeddings to study human cognition and behavior.

Comparing spatial and content analysis of residents and tourists using Geotagged Social Media Data. The Historic Neighbourhood of Alfama (Lisbon), a case study

Tourism flows to large cities have increased drastically in the past few years. The Alfama neighbourhood in Lisbon (Portugal) is facing major changes with respect to land uses, demographic features and social appropriation patterns in public spaces, caused by the intensification of tourism. The consequences of new emerging economic and symbolic values have rapidly given rise to a scenario of touristification and gentrification in the neighbourhood. In order to address such complexities, sustainable urban planning can benefit from real-time data sources that can represent the tourism flows in spatial and temporal perspectives. The research question allows Twitter to be used as an emerging data source and for analysing spatial patterns and content, based on two sample groups: residents and tourists, and their interpretations about the use of space for leisure activities. The research method is based on an analysis of two years of geotagged tweets in the city of Lisbon, differentiating between tourist and resident users, and, in a subsequent step, in the Alfama neighbourhood. The spatial distribution analysis and the content analysis have revealed not only spatio-temporal activity patterns but also emotional responses to new trends in urban tourism uses, consumption and perception of an increasing tourism pressure in Alfama. The results are relevant in the field of tourism and sustainable urban planning.

Detecting Urban Events by Considering Long Temporal Dependency of Sentiment Strength in Geotagged Social Media Data

The development of location-based services facilitates the use of location data for detecting urban events. Currently, most studies based on location data model the pattern of an urban dynamic and then extract the anomalies, which deviate significantly from the pattern as urban events. However, few studies have considered the long temporal dependency of sentiment strength in geotagged social media data, and thus it is difficult to further improve the reliability of detection results. In this paper, we combined a sentiment analysis method and long short-term memory neural network for detecting urban events with geotagged social media data. We first applied a dictionary-based method to evaluate the positive and negative sentiment strength. Based on long short-term memory neural network, the long temporal dependency of sentiment strength in geotagged social media data was constructed. By considering the long temporal dependency, daily positive and negative sentiment strength are predicted. We extracted anomalies that deviated significantly from the prediction as urban events. For each event, event-related information was obtained by analyzing social media texts. Our results indicate that the proposed approach is a cost-effective way to detect urban events, such as festivals, COVID-19-related events and traffic jams. In addition, compared to existing methods, we found that accounting for a long temporal dependency of sentiment strength can significantly improve the reliability of event detection.

Examination of Urban Landscape Aesthetics through Mapping Using Geotagged Social Media Data from Flickr

Landscapes are part of the earth’s surface, and include terrestrial, aquatic, natural, and anthropogenic features with differing dominance. The evaluation of Landscape Aesthetics (LA) and scenery is helpful for the understanding landscapes. The mapping of LA is increasingly being recognized in urban and landscape planning. Examining the flow and capacity of LA is an essential element in understanding the sustainability of urban landscapes. An explicit methodological approach was proposed for mapping LA using social media data (geolocated photographs) from Flickr to study the spatial distribution of LA in Xi’an, China. The LA flow and LA capacity were conceptually distinguished and assessed. The LA flow was analyzed with 10,278 geotagged photographs from Flickr to express people’s actual appreciation of LA. The LA capacity was examined by spatial criterion evaluation, combined the weight of landscape features. LA flow and LA capacity differ in both spatial extent and quantity, although there was a spatially explicit balance between them. The results show that LA flow largely occurs in the urban core zone and that LA capacity is concentrated in certain parts of Xi’an due to the imbalanced distribution of historical cultural heritage sites. This research contributes to a better understanding of LA values and the relationships between LA values and landscape characteristics. The results can support the integration of cultural services in landscape planning and land use policy.

Find you if you drive: Inferring home locations for vehicles with surveillance camera data

Inferring home locations for users from spatiotemporal data has become increasingly important for real-world applications ranging from security, recommendation, advertisement targeting, to transportation scheduling. Existing home location inference studies are based either on geo-tagged social media data or continuous GPS data. Yet this inference problem in highly sparse vehicle trajectories in urban surveillance systems remains largely unexplored. In this paper, we propose an accurate home location inference framework for vehicles in urban traffic surveillance systems by considering both spatial and temporal characteristics. To the best of our knowledge, we are the first to predict exact home community for vehicles at such a fine granularity using the sparse and noisy surveillance camera data. First, we collect and preprocess multiple contextual datasets to obtain a context-rich road network with residential communities and surveillance cameras. Second, we detect the potential home location areas for each vehicle by clustering Origin–Destination (O-D) pairs extracted in vehicle’s camera-based trajectories. Then we further propose an in∕out time pattern to distinguish the home area candidate from the O-D clusters by leveraging time-aware constraints. Furthermore, to find the exact home community, we propose a Kernel Density Estimation (KDE) based inference method with a local camera selection strategy to effectively identify the home community from the residential communities near/in the home area candidate. Our comprehensive experiments on a large-scale real-world dataset demonstrate the effectiveness of our proposed method.

Perceiving Beijing’s “City Image” Across Different Groups Based on Geotagged Social Media Data

City image in general refers to the perception, the feeling, and the opinion of a city, which contributes great importance to urban management, urban planning, urban cultural perceptions, and tourism resource development. Traditionally, city image is often inferred by the `five-element' model of physical factors while lacking the consideration of subjective perception. With the rising penetration of smart mobile devices and social media, massive data of location-related texts has been generated for a variety of urban areas. The accessibility to the big data leads to a new approach of understanding the subjective perception of city image, which is important since the new approach takes the subjective heterogeneity into account. Based on the Beijing's Weibo (microblog) data in the year of 2016, we use a random forest model to categorize user backgrounds into locals and non-locals. Meanwhile, spatial clustering is applied to identify hotspots. Then two text analysis methods-term frequency-inverse document frequency (TF-IDF) and latent Dirichlet allocation (LDA)-are adopted to abstract topics regarding the different geographical hotspots in the city across the different groups of individuals. Our research shows text mining on geotagged big data for city image makes it possible to accommodate the heterogeneity of the activities of different groups of people and to understand their preferences for different points of interests in the city, and thereby reveals the socio-cultural and functional features for the city.

Space-Time Drift Point Detection in Mobility Patterns

Location-aware information is now commonplace, as the ubiquity and pervasiveness of technology enabled its generation and storage at large scale. These data constitute a rich representation of entities’ whereabouts and behavior as they move on the map. Although several studies reported considerable predictability of such mobility patterns, several factors may impose significant changes on moving behavior. Being able to detect these changes can benefit several applications. In this article, we formalize and address the problem of detecting mobility drifts in mobility patterns. This problem is particularly challenging due to the noisy and incomplete nature of the data. We design non-parametric tests and present two algorithms to detect mobility drifts when the putative drift point is known in advance and there is no previous knowledge about the existence of potential changes, and we need to search for the most likely drift point rigorously. To evaluate our algorithms, we perform an extensive experimental study with real-world data coming from a variety of scenarios, such as geo-tagged social media data and GPS traces of connected vehicles. The results show the effectiveness of our algorithms, being able to identify existing drift points on spatial mobility patterns correctly.

Visual abstraction and exploration of large-scale geographical social media data

A great deal of text and geographical information is provided in the geo-tagged social media data, which offers unprecedented opportunities to get insights into the social behaviors across different local areas. With the increasing size of geo-tagged social media data, a large number of visual mapping elements overlap with each other, which makes it difficult to visually capture topics of interest as well as their spatial distribution. In this paper, we propose a visual abstraction framework for the exploration of large scale geo-tagged social media data. Probabilistic topic modeling is firstly utilized to summarize the semantics of texts and extract a set of topic features of interest. Then, a multi-objective sampling model is designed to generate a subset of original dataset, which will not only reduce the visual clutter of large-scale social media data visualization, but also retain the ordering of topic features of interest as well as the geographical distribution of original social media datasets. A rich set of visual designs such as word cloud, text stream and heat map are integrated into the visual abstraction framework, enabling users to evaluate the sampled results from different perspectives including semantic topics, temporal changes and spatial distribution. Case studies based on real-world datasets and interviews with domain experts have demonstrated the effectiveness of our system in simplifying the geographical visualization of large scale geo-tagged social media data and exploring the social behaviors across different local areas.

Research on event perception based on geo-tagged social media data

Abstract. Technological advancement makes information dissemination more convenient. When a notable event occurs, social media serves a popular platform for citizens to share event-related information. Therefore, in the information age, how to effectively observe the event and improve event management ability is an open question worthy of attention. Traditional social survey methods and various automatic sensors have been widely used to monitor the specific event. However, widely used social media service provides a unique approach for the event study with individuals as smart sensors. How to perceive an event through social media data has triggered a series of researches. Currently, we can find when, where what happened and induced impact based on geo-tagged social media data. However, event study based on social media is still in its infancy. This paper provides an overview of event study based on geo-tagged social media data. Firstly, we introduce the event model and the characteristics of social media data. Then, how to detect and trace event, how to analyze event impact and visually express obtained knowledge are discussed respectively. Subsequently, based on the existing researches, we propose further questions and conclude.

Assessing wild fire risk in the United States using social media data

Massive Geo-tagged social media data provide new opportunities for disaster risk assessment, prevention, and management. This article presents a proof of concept for assessing wildfire risk using Geo-tagged social media data, by taking wildfire risk as a function of wildfire hazard and social–ecological vulnerability. The case study of the United States shows that the regions with the highest wildfire hazard are concentrated in the Western, while the most vulnerable areas are mainly distributed in the Eastern, the Western Coast, and the Southern parts of the nation. Areas with high wildfire risk are mainly located in the Northwestern and Southeastern United States. It shows that the wildfire risk level has significant linear relationship with population density. Massive and vulnerable population might result in significant increase in wildfire risk perception. We conclude that Geo-tagged social media data have great potential in disaster risk studies.

Semantic-Aware Visual Abstraction of Large-Scale Social Media Data With Geo-Tags

With the rapid growth of geo-tagged social media data, it has become feasible to explore topics across different areas through text mining and geographical visualization. However, the visual elements of social media data always overlap with each other in the map view, which largely disturbs visual perception of semantic features and their geographical distribution. Thus, it is of great significance to reduce the visual clutter of large-scale social media data, and enhance the visibility of semantic features across local areas. In this paper, we utilize a doc2vec model to transform geo-tagged social media data into high-dimensional vectors, and the semantic correlation can be easily characterized in the dimensionality reduction space. Aiming at the reduction of visual clutter of geographical visualization, a dual-objective blue noise sampling model is proposed to select a subset of social media data, by means of which both the semantic correlation and spatial distribution of large scale social media data are well retained. A rich set of visual designs are implemented enabling users to evaluate the sampled results from multiple perspectives and explore the changes of semantic features across areas, such as heatmap, word cloud and text stream. The effectiveness and validity of the proposed visualization system are further demonstrated through case studies and expert reviews.

Geo-Tagged Social Media Data-Based Analytical Approach for Perceiving Impacts of Social Events

Studying the impact of social events is important for the sustainable development of society. Given the growing popularity of social media applications, social sensing networks with users acting as smart social sensors provide a unique channel for understanding social events. Current research on social events through geo-tagged social media is mainly focused on the extraction of information about when, where, and what happened, i.e., event detection. There is a trend towards the machine learning of more complex events from even larger input data. This research work will undoubtedly lead to a better understanding of big geo-data. In this study, however, we start from known or detected events, raising further questions on how they happened, how they affect people’s lives, and for how long. By combining machine learning, natural language processing, and visualization methods in a generic analytical framework, we attempt to interpret the impact of known social events from the dimensions of time, space, and semantics based on geo-tagged social media data. The whole analysis process consists of four parts: (1) preprocessing; (2) extraction of event-related information; (3) analysis of event impact; and (4) visualization. We conducted a case study on the “2014 Shanghai Stampede” event on the basis of Chinese Sina Weibo data. The results are visualized in various ways, thus ensuring the feasibility and effectiveness of our proposed framework. Both the methods and the case study can serve as decision references for situational awareness and city management.

Normalization Strategies for Enhancing Spatio-Temporal Analysis of Social Media Responses during Extreme Events: A Case Study based on Analysis of Four Extreme Events using Socio-Environmental Data Explorer (SEDE)

Abstract. With social media becoming increasingly location-based, there has been a greater push from researchers across various domains including social science, public health, and disaster management, to tap in the spatial, temporal, and textual data available from these sources to analyze public response during extreme events such as an epidemic outbreak or a natural disaster. Studies based on demographics and other socio-economic factors suggests that social media data could be highly skewed based on the variations of population density with respect to place. To capture the spatio-temporal variations in public response during extreme events we have developed the Socio-Environmental Data Explorer (SEDE). SEDE collects and integrates social media, news and environmental data to support exploration and assessment of public response to extreme events. For this study, using SEDE, we conduct spatio-temporal social media response analysis on four major extreme events in the United States including the “North American storm complex” in December 2015, the “snowstorm Jonas” in January 2016, the “West Virginia floods” in June 2016, and the “Hurricane Matthew” in October 2016. Analysis is conducted on geo-tagged social media data from Twitter and warnings from the storm events database provided by National Centers For Environmental Information (NCEI) for analysis. Results demonstrate that, to support complex social media analyses, spatial and population-based normalization and filtering is necessary. The implications of these results suggests that, while developing software solutions to support analysis of non-conventional data sources such as social media, it is quintessential to identify the inherent biases associated with the data sources, and adapt techniques and enhance capabilities to mitigate the bias. The normalization strategies that we have developed and incorporated to SEDE will be helpful in reducing the population bias associated with social media data and will be useful for researchers and decision makers to enhance their analysis on spatio-temporal social media responses during extreme events.