Enrico Steiger's research while affiliated with Universität Heidelberg and other places

Publications (24)

Article
Full-text available
Cities are complex systems, where related Human activities are increasingly difficult to explore within. In order to understand urban processes and to gain deeper knowledge about cities, the potential of location-based social networks like Twitter could be used a promising example to explore latent relationships of underlying mobility patterns. In...
Data
Simulated data (?inclusion? and ?small-scale perspective?) used in Section ?Influence of Scale Differences on the Numbers of Interactions.? See the respective attached file. (ZIP)
Article
Full-text available
Twitter and related social media feeds have become valuable data sources to many fields of research. Numerous researchers have thereby used social media posts for spatial analysis , since many of them contain explicit geographic locations. However, despite its widespread use within applied research, a thorough understanding of the underlying spatia...
Data
Simulated data (?overlap? and ?small-scale perspective?) used in Section ?Influence of Scale Differences on the Numbers of Interactions.? See the respective attached file. (ZIP)
Data
Simulated data used in Sections ?Influences on Spatial Autocorrelation? and ?Increased Topological Variability.? See the respective attached file. (ZIP)
Data
Spatial distribution of the Gaussian attribute values across the single pattern (top) and their histogram (bottom). (TIFF)
Data
Simulated data (?inclusion? and ?large-scale perspective?) used in Section ?Influence of Scale Differences on the Numbers of Interactions.? See the respective attached file. (ZIP)
Data
Spatial distribution of the Gaussian mixture across a combined pattern (top) and their joint histogram (bottom). (TIFF)
Data
Goodness of fit for the fitted functions. Blue: exponential function; red: linear function. Please read the fits in a cumulative way. The exponential function was evaluated from left to right. That is, the determined optimum at 15 means that the first 15 meters of the course follow the respective function. In contrast, the red linear function needs...
Data
Simulated data (?overlap? and ?large-scale perspective?) used in Section ?Influence of Scale Differences on the Numbers of Interactions.? See the respective attached file. (ZIP)
Data
Twitter sample. See the respective attached file. (ZIP)
Data
Heat map of pairwise covariance terms and semivariogram of topic associations. The white semivariogram plotted atop of the heat map refers to the right-hand y-axis. The left-hand y-axis is associated with the underlying color-coded bins of the heat map. This figure is similar to Fig 3, but shows relative heat map values for reasons of comparison (i...
Article
Full-text available
Social networks have been used to overcome the problem of incomplete official data, and to provide a more detailed description of a disaster. However, the filtering of relevant messages on-the-fly remains challenging due to the large amount of misleading, outdated or inaccurate information. This paper presents an approach for the automated geograph...
Conference Paper
Full-text available
Social networks have been used to overcome the problem of incomplete official data, and provide a more detailed description of a disaster. However, the filtering of relevant messages on-the-fly remains challenging due to the large amount of misleading, outdated or inaccurate information. This paper presents an approach for the automated geographic...
Article
Full-text available
The investigation of human activity patterns from location-based social networks like Twitter is an established approach of how to infer relationships and latent information that characterize urban structures. Researchers from various disciplines have performed geospatial analysis on social media data despite the data’s high dimensionality, complex...
Conference Paper
Flood risk management requires updated and accurate information about the overall situation in vulnerable areas. Social media messages are considered to be as a valuable additional source of information to complement authoritative data (e.g. in situ sensor data). In some cases, these messages might also help to complement unsuitable or incomplete s...
Article
Full-text available
The objective of this paper is to conduct a systematic literature review that provides an overview of the current state of research concerning methods and application for spatiotemporal analyses of the social network Twitter. Reviewed papers and their application domains have shown that the study of geographical processes by using spatiotemporal in...
Article
Full-text available
The investigation of human activity in location-based social networks such as Twitter is one promising example of exploring spatial structures in order to infer underlying mobility patterns. Previous work regarding Twitter analysis is mainly focused on the spatiotemporal classification of events. However, since the information about the occurrence...
Technical Report
Full-text available
The increasing number of research contributions covering various topics and research questions from multiple academic disciplines which are relevant to geographical information science (GIScience), are a challenging factor for a successful selection and assessment of corresponding high quality research articles. Heterogeneous multiple electronic da...
Conference Paper
Full-text available
Flood risk management requires updated and accurate information about the overall situation in vulnerable areas. Social media messages are considered to be as a valuable additional source of information to complement authoritative data (e.g. in situ sensor data). In some cases, these messages might also help to complement unsuitable or incomplete s...
Conference Paper
In this paper, we propose a framework to detect human mobility transportation hubs and infer public transport flows from unstructured georeferenced social media data using semantic topic modeling and spatial clustering techniques. An infrastructure for receiving and storing large sets of social media data has been developed together with an ad hoc...

Citations

... Additionally, computational tools such as computer vision techniques, and performance analysis are designed to help urban planners to analyze data from social media to unveil specific aspects of a city (Jenkins et al., 2016;Kelley, 2013) as diverse as tourist flows , and eating habits of citizens (Mejova et al., 2016). Moreover, geographic, temporal and functional aspects present in social media posts or check-ins are employed in a variety of applications including risk reporting (Velasco et al., 2017), traffic/congestion prediction (Liao et al., 2018;Pourebrahim et al., 2018), event detection Angaramo & Rossi, 2018;Ferracani et al., 2017), disasters prevention (Assis et al., 2016;Kankanamge et al., 2019), location recommendation, urban functional zone study, crime prediction (Yang & Eickhoff, 2018), and delivery of municipal services (Yigitcanlar, Kankanamge, et al., 2020). These studies, analyses and applications can inform computational models and offer new possibilities for a wide variety of professionals linked to urban planning, city branding, mobility studies, and city management as well as artists and activists concerned about raising awareness of the issues of public interest (Candeia et al., 2017;. ...
... Steiger [19] commented that "the user-generated, textual content of tweets is noisy, making it challenging to apply natural language processing (NLP) techniques to identify meaningful information". He identified a few frequently repeating, daily patterns with similar time-dependent disruption characteristics along major arterial (ring) roads (similar to road centrality), as a proxy indicator of mobility behaviour. ...
... The second investigation conducted reflects the inhomogeneous case and comprises a tweet data set extracted from a one-year Twitter corpus from London, UK. The data is available online (see Westerholt et al. 2016b) and has been used in previous studies (e.g., Steiger et al. 2015; Steiger, Resch, and Zipf 2016; Westerholt 2021). The full preprocessing chain can be found in Steiger et al. (2015) and includes stop word removal, tokenization, and stemming. ...
... To some extent, this even applies to mixtures of differently distributed random variables, although this would require even larger sample sizes. For the latter case, however, it has been shown that the spatial arrangement of the random variables involved has an influence on both the mean and the variance of I, especially when the underlying means and variances of the distributions that enter mixtures differ greatly (Westerholt 2018;Westerholt et al. 2016). Therefore, caution is still required when drawing conclusions about Moran's I using non-normal data. ...
... The user can unfollow pages or people, but they cannot unfollow broader themes like 'politics'. This is a problem because users' posts are a mixture of themes [5,31,41] ⎯ for example, a data scientist professional may post about data science related topics 80% of the time and about politics for the rest of the 20%. However, the user following this person might only be interested in the data science content that the individual posts. ...
... Text mining and semantic analysis from tweets has been a prolific area of investigation over the last decade thanks to the easy collection of data samples with a wealth of opinions and feelings within a short time. Text mining has been used to map and compare the frequency of feelings in cities during the day (Kocich, 2017;Lansley & Longley, 2016;Steiger, Resch, & Zipf, 2016;Wachowicz & Liu, 2016), to detect possible natural phenomena like hurricanes or earthquakes (Hiltz et al., 2014;Sakaki, Okazaki, & Matsuo, 2010), or to extract the spatial patterns of feelings in different events, such as the 2016 United States Elections (Chin, Zappone, & Zhao, 2016), 2015 baseball games in Boston (USA) (Steiger, Ellersiek, Resch, & Zipf, 2015), 2014 Sochi Winter Olympic Games (Kirilenko & Stepchenkova, 2017), or 2012London Summer Olympic Games (Kovacs-Gyori, Ristea, Havas, Resch, & Cabrera-Barona, 2018. Another field of growing importance has been studying the perceptions and feelings about green spaces during different times of the day such as urban green spaces and parks in Melbourne (Lim et al., 2018) or in London (Kovacs-Györi et al., 2018). ...
... Lloyd and Cheshire [50] collected Twitter data about retail stores and used adaptive KDE to visualize the flow of people related to retail stores, an approach that is less biased and smoother than traditional KDE methods. Steiger et al. [51] proposed a self-organizing map (Geo-SOM and Geo-H-SOM) for the visualization of human activity patterns. Secondly, there are studies that investigate spatiotemporal activity characteristics and influencing factors of urban space. ...
... The keywords in Brazilian-Portuguese were "chuv*" (chuva, chuvisco, chuvarada, etc.), "garoa*" (garoando, etc.), "temp*" (temporal, tempestade, tempo ruim, etc.), "alag*" (alagamento, alagado, etc.), and "inund*" (inundação, inundado, etc.). These keywords were chosen and extended from a list of previous studies (ASSIS et al., 2015). After this, a manual search was carried out for the subset of potential rain-related geotagged tweets to remove those false-positive (2,916 geotagged tweets were removed). ...
... With the development and popularization of location-based services and mobile social networks, a large amount of trajectories or geotag data are constantly accumulating, which brings unprecedented opportunities to create computational characterization of urban residents (Goodchild, 2011). Geospatial big data that can be used to capture spatiotemporal patterns of human activities include taxi trajectories (Kang & Qin, 2016), cell phone records (Lee et al., 2018;Xiong et al., 2021), social media or social network data (Steiger et al., 2015), smart card records in public transportation systems (Long & Thill, 2015;Wang et al., 2017), and so on. From the activity time rhythm presented in the data, we can find people's daily activities, social network coverage, the spatial distribution of urban occupation and housing, etc. (Cai et al., 2019;Cao et al., 2021;Gao et al., 2018;Gong et al., 2017;Liu et al., 2021;Zhang et al., 2020). ...
... Extracting social media content relating to an event To aid situational awareness, one approach to using social media to obtain impact information is to extract the relevant tweet content relating to the event for further review. This includes extracting the relevant text and/or images, relating to reports of flooding (Oktafiani et al., 2012;Herfort et al., 2014b,a;Rossi et al., 2018;Moumtzidou et al., 2018;Rodavia et al., 2018;Huang et al., 2019b,a;Shi et al., 2019;Jony et al., 2019;Wang et al., 2020a); damage assessment during floods (Brovelli et al., 2014;Assis et al., 2015); and creating a dataset of floodrelated tweets in Arabic languages (Shannag & Hammo, 2019;Hamoui et al., 2020). Accessing tweets relating to reports of landslides has also been explored by Musaev & Hou (2017). ...