The LBSN structure is a common data scheme that aims to systematically describe objects, bases and their relations currently found on Location Bases Social Networks.
For example, a Post is considered an object and may refer to a tweet on Twitter, a photo on Flickr, or a video shared on YouTube. However, despite some differences, all posts are characterized by a set of common attributes (called bases), such as a description, the content itself, (hash)tags, or the date of publication (etc.).
Based on a 4-Facet-Model introduced in Dunkel et al. 2018, objects, bases and relations that exist in various LBSN are assigned to the following five key dimensions.
The social facet describes a specific user identity, but also encompassing wider affiliation with social groups or cultures. The underlying assumption is that events, to which people react to on Social Media, involve or affect different groups of people differently. In other words, whether someone feels affected or unaffected, is considered a participant, observer or witness, or takes a positive or negative stance, depends, to some degree, on the social background of this individual. This may encompass complex aspects including political orientation, beliefs, values, norms and preferences, which express a continuum of people’s relationships towards an event that (often unconsciously) affect reactions.
Objects organized under the Social Facet currently include:
|Origin||A Location Based Social Network consisting of a large group of people|
|CompositeKey||A Composite Key used to reference unique objects across different LBSN|
|User||A single user (e.g. a profile or an account) on a location based social network (LBSN)|
|UserGroup||A single group of users on a LBSN|
|Language||A common language used on LBSN, relating to a larger group of people sharing the same language|
The difference between the thematic and social facet is in the relation to the user. Thematic attributes include immediate situational aspects that affect reactions from an in a particular situation (e.g. sentiments, feelings, emotions, or any other attributes of the reaction environment). Therefore, unlike social attributes, thematic attributes change frequently from one reaction to another. Possible questions explored in visual analytics include but are not limited to emotional states of the actor (e.g. positive, neutral, negative), as inferred from emoticons or based on sentiment analysis (Bai et al. 2019), or the stance of different actors to events as inferred from semantics such as titles, comments or descriptions etc. (Zeng et al. 2016). Keywords such as hashtags, for example, may indicate what aspects of an event were perceived as being of particular importance (Towne et al. 2016), or refer to individual event consequences or actions people have undertaken or plan to undertake (Gao et al. 2014).
Objects organized under the Topical Facet currently include:
|Post||An single post on a location based social network (LBSN) providing original (new) content|
|PostReaction||A reaction on a location based social network (LBSN) such as like, quote, share etc.|
Both time and space are referenced from many objects in LBSN, such as the location of a post, the date when a user joins a Social Network or when a place is added to a map. Beyond these references, independent spatial objects of LBSN Networks are frequently found at different levels of granularity. Spatial information found on LBSN is of high relevancy to user privacy, because most location data can be used to track or identify users. In the base LBSN Structure we consider objects on four levels of spatial granularity, Coordinate, Place, City and Country, which can be further subdivided based on personal needs.
Objects organized under the Spatial Facet currently include:
|Place||A particular (named) place on a location based social network (LBSN).|
|City||A city on a location based social network (LBSN).|
|Country||A country on a location based social network (LBSN).|
Both time and space are referenced from many objects in LBSN such as the time of sharing a post, the date when a user joins a Social Network or when someone reacts to something. Beyond these references, an explicit temporal object frequently found on LBSN is the Event. Similar to objects of the spatial facet, events may cover a range of granularities crossing different hierarchical levels.
Objects organized under the Temporal Facet currently include:
|Event||An (named) event with a representation on LBSN.|
Interlinkage and the spread of information
Interlinkage and relations between objects in LBSN can be considered as an additional, fifth facet. Relationships on LBSN are the cause of many privacy concerns, because they can be used to combine information in ways other than originally anticipated by the user.
There exist direct references between different objects of the LBSN structure, such as the user-id linking several posts to the same user, but also more complex relationships such as many-to-many.
Implicit and explicit relationships are considered that link entities between two different Origins (e.g. different LBSN) or between implicit social media and external sources (explicit VGI, e.g. Wikipedia, OSM).
A specific example for a relationship is the spread of information, an important characteristic of LBSM, frequently evaluated in visual analytics. The spread of information occurs when reactions become the referent event for other reactions.
Bai, Q., Dan, Q., Mu, Z., & Yang, M. (2019). A Systematic Review of Emoji: Current Research and Future Perspectives. Frontiers in Psychology, 10(October). https://doi.org/10.3389/fpsyg.2019.02221
Dunkel, A., Andrienko, G., Andrienko, N., Burghardt, D., Hauthal, E., & Purves, R. (2018). A conceptual framework for studying collective reactions to events in location-based social media. International Journal of Geographical Information Science, 00(00), 1–25. https://doi.org/10.1080/13658816.2018.1546390
Gao, H., et al., 2014. Modeling user attitude toward controversial topics in online social media. Eighth International AAAI Conference on Weblogs and Social Media, 27–29 May 2015 Oxford, 121–130. Available from http://www.aaai.org/ocs/index.php/ICWSM/ICWSM14/paper/ viewPDFInterstitial/8058/8112 [Accessed 9 August 2017].
Towne, W., Rosé, C., and Herbsleb, J., 2016. Measuring similarity similarly: LDA and human perception. ACM transactions on intelligent systems and technology ACM reference format ACM trans. Intell. Syst. Technol, 7(2), 25, 1–25:29. doi:10.1145/0000000.0000000.
Zeng, L., Starbird, K., and Spiro, E.S., 2016. #Unconfirmed: classifying rumor stance in crisis-related social media messages. Proceedings of the International AAAI Conference on Weblogs and Social Media, 17–20 May 2016, Maternushaus, Germany: AAAI Press. 747–750. Available from http:// www.aaai.org/ocs/index.php/ICWSM/ICWSM16/paper/viewFile/13075/12845 [Accessed 9 August 2017].