Time Series Prediction for Graphs in Kernel and Dissimilarity Spaces

Paaßen B, Göpfert C, Hammer B (2018)
Neural Processing Letters 48(2): 669-689.

Zeitschriftenaufsatz | Veröffentlicht | Englisch
Abstract / Bemerkung
Graphs are a flexible and general formalism providing rich models in various important domains, such as distributed computing, intelligent tutoring systems or social network analysis. In many cases, such models need to take changes in the graph structure into account, that is, changes in the number of nodes or in the graph connectivity. Predicting such changes within graphs can be expected to yield important insight with respect to the underlying dynamics, e.g. with respect to user behaviour. However, predictive techniques in the past have almost exclusively focused on single edges or nodes. In this contribution, we attempt to predict the future state of a graph as a whole. We propose to phrase time series prediction as a regression problem and apply dissimilarity- or kernel-based regression techniques, such as 1-nearest neighbor, kernel regression and Gaussian process regression, which can be applied to graphs via graph kernels. The output of the regression is a point embedded in a pseudo-Euclidean space, which can be analyzed using subsequent dissimilarity- or kernel-based processing methods. We discuss strategies to speed up Gaussian Processes regression from cubic to linear time and evaluate our approach on two well-established theoretical models of graph evolution as well as two real data sets from the domain of intelligent tutoring systems. We find that simple regression methods, such as kernel regression, are sufficient to capture the dynamics in the theoretical models, but that Gaussian process regression significantly improves the prediction error for real-world data.
Erscheinungsjahr
Zeitschriftentitel
Neural Processing Letters
Band
48
Zeitschriftennummer
2
Seite
669-689
ISSN
eISSN
PUB-ID

Zitieren

Paaßen B, Göpfert C, Hammer B. Time Series Prediction for Graphs in Kernel and Dissimilarity Spaces. Neural Processing Letters. 2018;48(2):669-689.
Paaßen, B., Göpfert, C., & Hammer, B. (2018). Time Series Prediction for Graphs in Kernel and Dissimilarity Spaces. Neural Processing Letters, 48(2), 669-689. doi:10.1007/s11063-017-9684-5
Paaßen, B., Göpfert, C., and Hammer, B. (2018). Time Series Prediction for Graphs in Kernel and Dissimilarity Spaces. Neural Processing Letters 48, 669-689.
Paaßen, B., Göpfert, C., & Hammer, B., 2018. Time Series Prediction for Graphs in Kernel and Dissimilarity Spaces. Neural Processing Letters, 48(2), p 669-689.
B. Paaßen, C. Göpfert, and B. Hammer, “Time Series Prediction for Graphs in Kernel and Dissimilarity Spaces”, Neural Processing Letters, vol. 48, 2018, pp. 669-689.
Paaßen, B., Göpfert, C., Hammer, B.: Time Series Prediction for Graphs in Kernel and Dissimilarity Spaces. Neural Processing Letters. 48, 669-689 (2018).
Paaßen, Benjamin, Göpfert, Christina, and Hammer, Barbara. “Time Series Prediction for Graphs in Kernel and Dissimilarity Spaces”. Neural Processing Letters 48.2 (2018): 669-689.
Alle Dateien verfügbar unter der/den folgenden Lizenz(en):
Copyright Statement:
This Item is protected by copyright and/or related rights. [...]
Link(s) zu Volltext(en)
Access Level
OA Open Access
Material in PUB:
In sonstiger Relation
Time Series Prediction for Relational and Kernel Data
Paaßen B (2017)
Bielefeld University.
Externe Forschungsdaten:
Beschreibung
MiniPalindrome Dataset
Beschreibung
Sorting Dataset

Export

Markieren/ Markierung löschen
Markierte Publikationen

Open Data PUB

Quellen

arXiv: 1704.06498

Suchen in

Google Scholar