Redefining Event Types and Group Evolution in Temporal Data

TL;DR

The framework is applied to evolving groups from several face-to-face interaction datasets, showing it enables richer, more reliable characterization of group dynamics with respect to state-of-the-art methods, especially when the groups are subject to complex relationships.

Abstract

Groups -- such as clusters of points or communities of nodes -- are fundamental when addressing various data mining tasks. In temporal data, the predominant approach for characterizing group evolution has been through the identification of ``events". However, the events usually described in the literature, e.g., shrinks/growths, splits/merges, are often arbitrarily defined, creating a gap between such theoretical/predefined types and real-data group observations. Moving beyond existing taxonomies, we think of events as ``archetypes" characterized by a unique combination of quantitative dimensions that we call ``facets". Group dynamics are defined by their position within the facet space, where archetypal events occupy extremities. Thus, rather than enforcing strict event types, our approach can allow for hybrid descriptions of dynamics involving group proximity to multiple archetypes. We apply our framework to evolving groups from several face-to-face interaction datasets, showing it enables richer, more reliable characterization of group dynamics with respect to state-of-the-art methods, especially when the groups are subject to complex relationships. Our approach also offers intuitive solutions to common tasks related to dynamic group analysis, such as choosing an appropriate aggregation scale, quantifying partition stability, and evaluating event quality.

Figures (9)

• Figure 1: Three representations for dynamic groups found in the literature. a) Does not describe the relation between the groups; b) Assigns labels (here represented by colors) to each entity/time, yielding a longitudinal group; c) Describe how groups in a timestep are related to those in the next. Note that none of these representations explicit the events occurring on the network.
• Figure 2: A realistic group evolution scenario (a), with corresponding event graph with (b)intersection of union, high threshold, (c) intersection over union, low threshold, (d)intersection over min size
• Figure 3: Representation of the continuous nature of facets $U$ and $I$
• Figure 4: Archetype events according to values of facets $\mathcal{U}$ and $\mathcal{I}$, for $\mathcal{O}=0$. In the middle, an example of an event not clearly affiliated with an archetype according to these facets.
• Figure 5: Typicality distribution of three SocioPatterns datasets using the same daily aggregation window

Theorems & Definitions (7)

• Definition 1: Unicity Facet
• Definition 2: Identity Facet
• Definition 3: Outflow Facet
• Definition 4: Attribute Entropy Change
• Definition 5: Backward Event Weights
• Definition 6: Forward Event Weights
• Definition 7