Location-based Social Media Data Research Articles

The Impact of Scale on Extracting Individual Mobility Patterns from Location-Based Social Media.

Understanding human movement patterns is crucial for comprehending how a city functions. It is also important for city planners and policymakers to create more efficient plans and policies for urban areas. Traditionally, human movement patterns were analyzed using origin-destination surveys, travel diaries, and other methods. Now, these patterns can be identified from various geospatial big data sources, such as mobile phone data, floating car data, and location-based social media (LBSM) data. These extensive datasets primarily identify individual or collective human movement patterns. However, the impact of spatial scale on the analysis of human movement patterns from these large geospatial data sources has not been sufficiently studied. Changes in spatial scale can significantly affect the results when calculating human movement patterns from these data. In this study, we utilized Weibo datasets for three different cities in China including Beijing, Guangzhou, and Shanghai. We aimed to identify the effect of different spatial scales on individual human movement patterns as calculated from LBSM data. For our analysis, we employed two indicators as follows: an external activity space indicator, the radius of gyration (ROG), and an internal activity space indicator, entropy. These indicators were chosen based on previous studies demonstrating their efficiency in analyzing sparse datasets like LBSM data. Additionally, we used two different ranges of spatial scales-10-100 m and 100-3000 m-to illustrate changes in individual activity space at both fine and coarse spatial scales. Our results indicate that although the ROG values show an overall increasing trend and the entropy values show an overall decreasing trend with the increase in spatial scale size, different local factors influence the ROG and entropy values at both finer and coarser scales. These findings will help to comprehend the dynamics of human movement across different scales. Such insights are invaluable for enhancing overall urban mobility and optimizing transportation systems.

Crowdsourced mapping of tourism distribution and dynamics in Iceland using Flickr

ABSTRACT Social media has become a catalyst for tourism in nature-based destinations, raising management challenges to cope with abrupt visitation growth in sensitive settings. Access to location-based social media data provides valuable material to identify and monitor spatial distribution patterns in tourism visitation and longitudinal trends, especially in vast, remote settings. This paper aims to evaluate the use of Flickr geodata for mapping spatiotemporal dynamics of tourist behaviors in Iceland. The consistency of the Flickr dataset with other proxies supports its validity and relevance. Tourism visitation appears closely related with accessibility factors, which are characterized by a strong seasonal variation. On longer timescales, the data reflect the increase of off-season visitation and the emergence of new tourist sites. This study demonstrates the value and potential of such a dataset to get an overview of the spatial patterns and dynamics of tourism at a national-level and outlines the limits inherent to its use.

Mapping dynamic human sentiments of heat exposure with location-based social media data

Understanding urban heat exposure dynamics is critical for public health, urban management, and climate change resilience. Near real-time analysis of urban heat enables quick decision-making and timely resource allocation, thereby enhancing the well-being of urban residents, especially during heatwaves or electricity shortages. To serve this purpose, we develop a cyberGIS framework to analyze and visualize human sentiments of heat exposure dynamically based on near real-time location-based social media (LBSM) data. Large volumes and low-cost LBSM data, together with a content analysis algorithm based on natural language processing are used effectively to generate near real-time heat exposure maps from human sentiments on social media at both city and national scales with km spatial resolution and census tract spatial unit. We conducted a case study to visualize and analyze human sentiments of heat exposure in Chicago and the United States in September 2021. Enabled with high-performance computing, dynamic visualization of heat exposure is achieved with fine spatiotemporal scales while heat exposure detected from social media data can be used to understand heat exposure from a human perspective and allow timely responses to extreme heat. HIGHLIGHTS Near real-time and high spatial resolution mapping of human sentiments of heat exposure with Twitter data An integrated cyberGIS and machine learning framework for visualizing heat exposure with Twitter data Human sentiment of heat exposure mapping in the City of Chicago and the United States

Empowering health geography research with location-based social media data: innovative food word expansion and energy density prediction via word embedding and machine learning

BackgroundThe exponential growth of location-based social media (LBSM) data has ushered in novel prospects for investigating the urban food environment in health geography research. However, previous studies have primarily relied on word dictionaries with a limited number of food words and employed common-sense categorizations to determine the healthiness of those words. To enhance the analysis of the urban food environment using LBSM data, it is crucial to develop a more comprehensive list of food-related words. Within the context, this study delves into the exploration of expanding food-related words along with their associated energy densities.MethodsThis study addresses the aforementioned research gap by introducing a novel methodology for expanding the food-related word dictionary and predicting energy densities. Seed words are generated from official and crowdsourced food composition databases, and new food words are discovered by clustering food words within the word embedding space using the Gaussian mixture model. Machine learning models are employed to predict the energy density classifications of these food words based on their feature vectors. To ensure a thorough exploration of the prediction problem, ten widely used machine learning models are evaluated.ResultsThe approach successfully expands the food-related word dictionary and accurately predicts food energy density (reaching 91.62%.). Through a comparison of the newly expanded dictionary with the initial seed words and an analysis of Yelp reviews in the city of Toronto, we observe significant improvements in identifying food words and gaining a deeper understanding of the food environment.ConclusionsThis study proposes a novel method to expand food-related vocabulary and predict the food energy density based on machine learning and word embedding. This method makes a valuable contribution to building a more comprehensive list of food words that can be used in geography and public health studies by mining geotagged social media data.

Analyzing changes in travel patterns due to Covid-19 using Twitter data in India

Advancements in technology have enabled researchers to gather large-scale mobility information cost-effectively. In fact, with millions of active users, location-based social media (LBSM) platforms such as Facebook, Twitter, Instagram, Flickr, etc., have become potential big data sources to measure individual behaviour. Despite such passive data collection techniques primarily not providing individual-level information, the sheer volume of such data facilitates a better understanding of aggregate patterns. Besides, conducting the conventional transportation survey during the initial waves of the COVID-19 pandemic was nearly impossible due to social- distancing and lockdown rules. In such context, the present research showcases a method for extracting the mobility traces and identifying the travel patterns of visitors and residents in Delhi from geo-labelled posts on Twitter.Initially, a heuristic classification strategy has been developed based on a few spatiotemporal assumptions to identify and differentiate visitors and residents based on user coordinates. Also, three supervised machine learning techniques, i.e., support vector machine (SVM), k-nearest neighbours (kNN) and decision tree, were used to classify users based on their historical coordinates. Afterwards, the spatial variation of their destination preferences was studied using K-Means, DBSCAN, and Means-Shift clustering techniques, out of which the K- Means clustering method performed best. Lastly, the travel patterns from tweets during pre and during pandemic (COVID-19) were compared using respective clusters.We observed that the performance of the proposed heuristic classifier is comparable with the supervised machine learning (ML) technique used for classification. Furthermore, the results indicate that the proposed model can successfully identify the cluster coordinates for visitors' spots as well as the locations of residents. During the pandemic situation, the mean distance travelled by users is significantly reduced. The study also shows that the number of long-distance trips has also decreased. Also, during COVID, tweets were done from very few unique tourist spots. This suggests lower tendencies of people to travel for tourism purposes. The proposed methods for classification and clustering in the present study will be crucial to obtain individual travel patterns from LBSM data.

Modeling activity spaces using big geo‐data: Progress and challenges

AbstractThe growing availability of big geo‐data, such as mobile phone data and location‐based social media (LBSM), provides new opportunities and challenges for modeling human activity spaces in the big data era. These datasets often cover a large sample size and can be used to model activity spaces more efficiently than traditional travel surveys. However, these data also have inherent limitations, such as the lack of reliable demographic information of individuals and a low sampling rate. This paper first reviews the strengths and weaknesses of various internal and external activity space indicators. We then discuss the pros and cons of using various new data sources (e.g., georeferenced mobile phone data and LBSM data) for activity space modeling. We believe this review paper is a valuable reference not only for researchers who are interested in activity space modeling based on big geo‐data, but also for planners and policy makers who are looking to incorporate new data sources into their future workflow.

Exploring the Relationship among Human Activities, COVID-19 Morbidity, and At-Risk Areas Using Location-Based Social Media Data: Knowledge about the Early Pandemic Stage in Wuhan.

It is significant to explore the morbidity patterns and at-risk areas of the COVID-19 outbreak in megacities. In this paper, we studied the relationship among human activities, morbidity patterns, and at-risk areas in Wuhan City. First, we excavated the activity patterns from Sina Weibo check-in data during the early COVID-19 pandemic stage (December 2019~January 2020) in Wuhan. We considered human-activity patterns and related demographic information as the COVID-19 influencing determinants, and we used spatial regression models to evaluate the relationships between COVID-19 morbidity and the related factors. Furthermore, we traced Weibo users’ check-in trajectories to characterize the spatial interaction between high-morbidity residential areas and activity venues with POI (point of interest) sites, and we located a series of potential at-risk places in Wuhan. The results provide statistical evidence regarding the utility of human activity and demographic factors for the determination of COVID-19 morbidity patterns in the early pandemic stage in Wuhan. The spatial interaction revealed a general transmission pattern in Wuhan and determined the high-risk areas of COVID-19 transmission. This article explores the human-activity characteristics from social media check-in data and studies how human activities played a role in COVID-19 transmission in Wuhan. From that, we provide new insights for scientific prevention and control of COVID-19.

Collecting, analyzing, and visualizing location-based social media data: review of methods in GIS-social media analysis

Collecting, analyzing, and visualizing location-based social media data: review of methods in GIS-social media analysis

Integrating the Eigendecomposition Approach and k-Means Clustering for Inferring Building Functions with Location-Based Social Media Data

Understanding the relationship between human activity patterns and urban spatial structure planning is one of the core research topics in urban planning. Since a building is the basic spatial unit of the urban spatial structure, identifying building function types, according to human activities, is essential but challenging. This study presented a novel approach that integrated the eigendecomposition method and k-means clustering for inferring building function types according to location-based social media data, Tencent User Density (TUD) data. The eigendecomposition approach was used to extract the effective principal components (PCs) to characterize the temporal patterns of human activities at building level. This was combined with k-means clustering for building function identification. The proposed method was applied to the study area of Tianhe district, Guangzhou, one of the largest cities in China. The building inference results were verified through the random sampling of AOI data and street views in Baidu Maps. The accuracy for all building clusters exceeded 83.00%. The results indicated that the eigendecomposition approach is effective for revealing the temporal structure inherent in human activities, and the proposed eigendecomposition-k-means clustering approach is reliable for building function identification based on social media data.

Evaluating the performance of LBSM data to estimate the gross domestic product of China at multiple scales: A comparison with NPP-VIIRS nighttime light data

Regional economic development evaluation is essential for understanding social and environmental issues. Although the nighttime light (NTL) data have been proved to be effective in economic estimation, it cannot reflect the human activities that occur during the daytime. Recently, with the widespread use of smart mobile devices, the location based social media (LBSM) data are increasingly being used as a proxy for real-time human activities. However, little work was carried out to explore the potential of LBSM data in estimating economic development at different scales in China. This study filled this gap by evaluating the effectiveness of Tencent user density (TUD) data, a typical type of LBSM data in China, in Gross Domestic Product (GDP) modeling at the provincial, municipal, and county scales. In this study, we employed holiday and non-holiday TUD sample data to simulate the annual TUD data, and compared it with the new generation NTL data, NPP/VIIRS images. The results showed that although the simulated annual TUD data does not perform better than NPP/VIIRS-NTL data in provincial and municipal GDP estimation, it outperforms NPP/VIIRS-NTL data at the county scale. More importantly, the simulated annual TUD data are much more powerful and reliable than NPP/VIIRS-NTL data in underdeveloped areas with complex terrain, such as the Northwest and Southwest China, as well as in more developed areas with separation of work and housing, such as the North China and South China. This is mainly because TUD data can reduce the impact of natural factors such as terrain on data collection as well as reflect both daytime and nighttime human activities. This study confirmed that the LBSM-TUD data is a potential and promising data source for economic modeling in small scale areas of China, which will help to support China's regional economic evaluation.

Assessing Place Type Similarities Based on Functional Signatures Extracted from Social Media Data

Place types are often used to query places or retrieve data in gazetteers. Existing gazetteers do not use the same place type classification schemes, and the various typing schemes can cause difficulties in data alignment and matching. Different place types may share some level of similarities. However, previous studies have paid little attention to the place type similarities. This study proposes an analytical approach to measuring similarities between place types in multiple typing schemes based on functional signatures extracted from web-harvested place descriptions. In this study, a functional signature consists of three component signature factors: place affordance, events, and key-descriptors. The proposed approach has been tested in a case study using Twitter data. The case study finds high similarity scores between some pairs of types and summarizes the situations when high similarities could occur. The research makes two innovative contributions: First, it proposes a new analytical approach to measuring place type similarities. Second, it demonstrates the potential and benefits of using location-based social media data to better understand places.

Analytics of location-based big data for smart cities: Opportunities, challenges, and future directions

The growing ubiquity of location/activity sensing technologies and location-based services (LBS) has led to a large volume and variety of location-based big data (LocBigData), such as location tracking or sensing data, social media data, and crowdsourced geographic information. The increasing availability of such LocBigData has created unprecedented opportunities for research on urban systems and human environments in general. In this article, we first review the common types of LocBigData: mobile phone network data, GPS data, Location-based social media data, LBS usage/log data, smart card travel data, beacon log data (WiFi or Bluetooth), and camera imagery data. Secondly, we describe the opportunities fueled by LocBigData for the realization of smart cities, mainly via answering questions ranging from “what happened” and “why did it happen” to “what's likely to happen in the future” and “what to do next”. Thirdly, pitfalls of dealing with LocBigData are summarized, such as high volume/velocity/variety; non-random sampling; messy and not clean data; and correlations rather than causal relationships. Finally, we review the state-of-the-art research trends in this field, and conclude the article with a list of open research challenges and a research agenda for LocBigData research to help achieve the vision of smart and sustainable cities.

Understanding gender difference in perceptions toward transit services across space and time: A social media mining approach

Location-based social media data can offer useful insights on the spatial and temporal dynamics of public attitudes. In this study, we aim to investigate the gendered attitudes toward transit services in China, utilizing the case of Shenzhen. We collected 44,257 Weibo microblogs, a major source of social media data in China, and applied a series of text mining and visualization techniques to examine the gender differences among our focused themes. The microblogs reveal a distinct gender gap in terms of quantity, as nearly 74% are posted by women. While women tend to be more concerned about the comfort of transit environment (e.g., temperature, crowdedness, and safety, especially at night), men tend to be more interested in transit systems’ e-payment services and reporting traffic incidents. Overall, this study presents a methodological framework and empirical case study about how we can utilize certain social media mining techniques to investigate gendered, subjective travel experiences, providing researchers and practitioners with an innovative way to gather customer service feedback and build more inclusive service systems.

Constructing and analyzing spatial-social networks from location-based social media data

ABSTRACT People interact with each other in space and time. Improved understanding of human interactions in spatial, temporal, and social dimensions are highly beneficial for research and practices in public health, urban planning, and other fields. Traditional methods of collecting social interaction data are time-intensive and resource-consuming, resulting in relatively small sample sizes and limited information. Furthermore, traditional methods often oversimplify the dynamics of human interactions and fail to capture the characteristics of places where the interactions occur. With the popularity of location-based social media (LBSM) platforms, people can publish information about their social events such as time, location, and other participants. This research introduces a framework that formalizes terminologies and concepts related to spatial-social connections for the construction of spatial-social networks from LBSM data in GIS. Supported by the framework, the study presents methods of collecting, analyzing, and visualizing LBSM data in spatial-social dimensions. The methods are implemented and tested in a case study with Facebook data. The case study demonstrates that location-based social media data can be transformed into spatial-social networks and then be analyzed and visualized to answer innovative types of scientific inquiries.

An Efficient Algorithm for Spatio-Textual Object Cluster Join

With the proliferation of GPS-based equipments and location-based services, spatio-textual objects have been playing an indispensable role in spatial data management. It is of great importance to enable the join operation among spatio-textual object groups. In this paper, we propose to study a novel problem of spatio-textual object cluster join (STOC-Join). Given two sets of spatio-textual objects D 1 and D 2 and a similarity threshold θ , the STOC-Join problem finds all object cluster pairs whose spatio-textual similarities are no less than θ . The problem of STOC-Join is practical in a variety of application scenarios, including location-based event detection, location-based data cleaning, and location-based social media data pre-processing in general. Efficient processing of STOC-Join is challenging in the following three aspects: (1) How to define and compute the spatio-textual similarity between two clusters of spatio-textual objects effectively; (2) How to efficiently cluster a large number of spatio-textual objects; (3) How to efficiently find similar cluster pairs and filter out unqualified pair candidates. To address the challenges, we define an effective and easy-to-compute similarity metric that measures the aggregated similarities between two groups of spatio-textual objects. Based on the similarity metric, we propose a novel two-phase matching algorithm that is able to cluster a large number of spatio-textual objects and find all cluster pairs efficiently. Our experiments on large real-life datasets confirm that our proposed two-phase matching algorithm is capable of achieving high efficiency compared with straightforward methods.

A Deep Neural Network for Crossing-City POI Recommendations

With the popularity of location-aware devices (e.g., smart phones), large amounts of location-based social media data such as check-ins are generated. This stimulates plenty of studies for POI recommendations by applying machine learning techniques. However, most of the existing studies focus on POI recommendations in the same city or region, and fail to recommend POIs for users when they travel to a new city. In this paper, we propose a novel deep neural network, named as ST-TransRec, for crossing-city POI recommendations. It integrates the deep neural network, transfer learning technique, and density-based resampling method into a unified framework. In this model, the deep neural network is used to capture users’ preferences for POIs and learn the embeddings of POIs. Besides, the transfer learning technique is employed to bridge the gap between cities that results from the city-dependent features. As the distributions over POIs are imbalanced, we design a density-based spatial resampling model which enables POIs to be well matched across cities. We conduct extensive experiments on two real-world datasets. The experimental results show the advantages of ST-TransRec over the state-of-the-art methods for crossing-city POI recommendations.

Urban Scale Trade Area Characterization for Commercial Districts with Cellular Footprints

Understanding customer mobility patterns to commercial districts is crucial for urban planning, facility management, and business strategies. Trade areas are a widely applied measure to quantify where the visitors are from. Traditional trade area analysis is limited to small-scale or store-level studies, because information such as visits to competitor commercial entities and place of residence is collected by labour-intensive questionnaires or heavily biased location-based social media data. In this article, we propose CellTradeMap, a novel district-level trade area analysis framework using mobile flow records (MFRs), a type of fine-grained cellular network data. We show that compared to traditional cellular data and social network check-in data, MFRs can model customer mobility patterns comprehensively at urban scale. CellTradeMap extracts robust location information from the irregularly sampled, noisy MFRs, adapts the generic trade area analysis framework to incorporate cellular data, and enhances the original trade area model with cellular-based features. We evaluate CellTradeMap on two large-scale cellular network datasets covering 3.5 million and 1.8 million mobile phone users in two metropolis in China, respectively. Experimental results show that the trade areas extracted by CellTradeMap are aligned with domain knowledge and CellTradeMap can model trade areas with a high predictive accuracy.

Modeling User Activity Space from Location-Based Social Media: A Case Study of Weibo

Activity space studies are beneficial for discovering meaningful activity patterns and providing a deeper understanding of human behaviors. There is insufficient research, however, on how reliable location-based social media (LBSM) is as a new data source for discovering user activity spaces. To this end, this research calculates four external and three internal activity space indicators based on Weibo data from three Chinese cities. We compared the strengths and weaknesses of these indicators for approximating user activity spaces from LBSM data with a low sampling resolution. We also tested how different amounts of check-in data affect the calculation of these activity space indicators. The results provide a useful reference for future experimental design in human activity modeling based on social media data.

Urban sprawl and its impact on sustainable urban development: a combination of remote sensing and social media data

ABSTRACT Urbanization is one of the most impactful human activities across the world today affecting the quality of urban life and its sustainable development. Urbanization in Africa is occurring at an unprecedented rate and it threatens the attainment of Sustainable Development Goals (SDGs). Urban sprawl has resulted in unsustainable urban development patterns from social, environmental, and economic perspectives. This study is among the first examples of research in Africa to combine remote sensing data with social media data to determine urban sprawl from 2011 to 2017 in Morogoro urban municipality, Tanzania. Random Forest (RF) method was applied to accomplish imagery classification and location-based social media (Twitter usage) data were obtained through a Twitter Application Programming Interface (API). Morogoro urban municipality was classified into built-up, vegetation, agriculture, and water land cover classes while the classification results were validated by the generation of 480 random points. Using the Kernel function, the study measured the location of Twitter users within a 1 km buffer from the center of the city. The results indicate that, expansion of the city (built-up land use), which is primarily driven by population expansion, has negative impacts on ecosystem services because pristine grasslands and forests which provide essential ecosystem services such as carbon sequestration and support for biodiversity have been replaced by built-up land cover. In addition, social media usage data suggest that there is the concentration of Twitter usage within the city center while Twitter usage declines away from the city center with significant spatial and numerical increase in Twitter usage in the study area. The outcome of the study suggests that the combination of remote sensing, social sensing, and population data were useful as a proxy/inference for interpreting urban sprawl and status of access to urban services and infrastructure in Morogoro, and Africa city where data for urban planning is often unavailable, inaccurate, or stale.

Improving population mapping using Luojia 1-01 nighttime light image and location-based social media data

Fine-resolution population mapping, which is vital to urban planning, public health, and disaster management, has gained considerable attention in socioeconomic and environmental studies. Although population distribution has been considered highly correlated with urban facilities, the quantitative relationship between the two has yet to be revealed when considering huge heterogeneity. To address this problem, the present study proposed a novel population mapping method by adopting Luojia 1-01 nighttime light imagery, points of interest (POI), and social media check-in data. A grid-based attraction degree (AD) model was built to quantify the possibility of population concentration in each geographic unit with a comprehensive consideration of the distribution and the popularity of facilities. On the basis of kernel density estimation, 16 attraction indexes were extracted by matching POI and check-in data. Multiple variables were used to train a random forest model, through which fine-scale population mapping was conducted in Zhejiang, China. The comparison between demographic and WorldPop data proved the high accuracy of our approach (R2 = 0.75 and 0.58). To explore the characteristics of the model further, the most appropriate search distance (650 m) and acquisition time (19:00–08:00) of the check-in data were discussed. The contrast experiment revealed that the model could outperform those from previous studies on rural and suburban areas with a few check-in points and low AD; thus, the mapping error caused by heterogeneity considerably decreased. The results indicated the proposed method has great potential in fine-scale population mapping.