Link Prediction on Evolving Data Using Tensor Factorization

Spiegel, Stephan; Clausen, Jan Hendrik; Albayrak, Şahin; Kunegis, Jérôme

doi:10.1007/978-3-642-28320-8_9

Cited by 32 publications

(27 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The cost of merging two adjacent segments is evaluated by the Singular Value Decomposition (SVD) model of the new segment. SVD-based algorithms are able to detect changes in the mean, variance and correlation structure among several variables [37,38,36]. The proposed approach can be considered as an extension of the sensor fusion algorithm developed by Abonyi [1,2].…”

Section: Time Series Segmentationmentioning

confidence: 99%

“…Principal Component Analysis (PCA) is a matrix factorization technique which is used to reduce the dimensionality of the input space and/or to retrieve latent relations between variables of the observed dataset. PCA can be found in many applications, such as recommender systems [20,33,38], semantic analysis of textual information [9] and link prediction in complex networks [3,24,37]. Furthermore, PCA is a common feature extraction method in signal processing, which is employed for time series segmentation and clustering [1,2,36].…”

Section: Feature Extractionmentioning

confidence: 99%

See 1 more Smart Citation

Pattern recognition in multivariate time series

Spiegel

Jain

Luca

et al. 2011

Proceedings of the 4th Workshop on Workshop for Ph.D. Students in Information &Amp; Knowledge Management

Self Cite

View full text Add to dashboard Cite

Nowadays computer scientists are faced with fast growing and permanently evolving data, which are represented as observations made sequentially in time. A common problem in the data mining community is the recognition of recurring patterns within temporal databases or streaming data. This dissertation proposal aims at developing and investigating efficient methods for the recognition of contextual patterns in multivariate time series in different application domains based on machine learning techniques. To this end, we propose a generic three-step approach that involves (1) feature extraction to build robust learning models based on significant time series characteristics, (2) segmentation to identify internally homogeneous time intervals and change points, as well as (3) clustering and/or classification to group the time series (segments) into the sub-population to which they belong to. To support our proposed approach, we present and discuss first experiments on real-life vehicular data. Furthermore we describe a number of applications, where pattern recognition in multivariate time series is practical or rather necessary.

show abstract

Section: Time Series Segmentationmentioning

confidence: 99%

Section: Feature Extractionmentioning

confidence: 99%

Pattern recognition in multivariate time series

Spiegel

Jain

Luca

et al. 2011

Proceedings of the 4th Workshop on Workshop for Ph.D. Students in Information &Amp; Knowledge Management

Self Cite

View full text Add to dashboard Cite

show abstract

“…The learned weights are used in a reinforcing way for the final prediction. Finally, in Spiegel et al (2011) tensor factorization is used to select the more predictive attributes, while in Lichtenwalter et al (2010) important features for link prediction are examined and it is provided a general, high-performance framework for the prediction task.…”

Section: Related Workmentioning

confidence: 99%

A supervised approach for intra-/inter-community interaction prediction in dynamic social networks

Rossetti

Guidotti

Miliou

et al. 2016

Soc. Netw. Anal. Min.

View full text Add to dashboard Cite

Due to the growing availability of Internet services in the last decade, the interactions between people became more and more easy to establish. For example, we can have an intercontinental job interview, or we can send real-time multimedia content to any friend of us just owning a smartphone. All this kind of human activities generates digital footprints, that describe a complex, rapidly evolving, network structures. In such dynamic scenario, one of the most challenging tasks involves the prediction of future interactions between couples of actors (i.e., users in online social networks, researchers in collaboration networks). In this paper, we approach such problem by leveraging networks dynamics: to this extent, we propose a supervised learning approach which exploits features computed by time-aware forecasts of topological measures calculated between node pairs. Moreover, since real social networks are generally composed by weakly connected modules, we instantiate the interaction prediction problem in two disjoint applicative scenarios: intracommunity and inter-community link prediction.Experimental results on real time-stamped networks show how our approach is able to reach high accuracy. Furthermore, we analyze the performances of our methodology when varying the typologies of features, community discovery algorithms and forecast methods.

show abstract

“…Without tensor factorization, the only way to achieve this goal is to perform clustering directly on the frontal slices of the tensor. Due to highdimensionality of the frontal slices, it is difficult to find distinct cluster structures in Euclidian space of frontal slices [35]. In contrast, the r-dimensional signatures derived from the tensor factorization form a projection subspace with much lower dimensionality, thus strengthen the underlying cluster structure of spatiotemporal traffic dynamic patterns.…”

Section: Clustering and Long-term Temporal Prediction Of Large-scale mentioning

confidence: 99%

“…The multi-way structure of tensor provides a natural way to encode the underlying multiple dependencies in the sequential multivariate data. For example, in video processing, tensors are widely used to represent temporal streams of multi-dimensional data, such as 2D images in video frames [34] and user product rating profiles in recommendation systems [35].…”

Section: Introductionmentioning

confidence: 99%

Analysis of Large-Scale Traffic Dynamics in an Urban Transportation Network Using Non-Negative Tensor Factorization

Han

Moutarde

2014

Int. J. ITS Res.

View full text Add to dashboard Cite

In this paper, we present our work on clustering and prediction of temporal evolution of global congestion configurations in a large-scale urban transportation network. Instead of looking into temporal variations of traffic flow states of individual links, we focus on temporal evolution of the complete spatial configuration of congestions over the network. In our work, we pursue to describe the typical temporal patterns of the global traffic states and achieve long-term prediction of the large-scale traffic evolution in a unified data-mining framework. To this end, we formulate this joint task using regularized Non-negative Tensor Factorization, which has been shown to be a useful analysis tool for spatio-temporal data sequences. Clustering and prediction are performed based on the compact tensor factorization results. The validity of the proposed spatio-temporal traffic data analysis method is shown on experiments using simulated realistic traffic data.

show abstract

Link Prediction on Evolving Data Using Tensor Factorization

Cited by 32 publications

References 17 publications

Pattern recognition in multivariate time series

Pattern recognition in multivariate time series

A supervised approach for intra-/inter-community interaction prediction in dynamic social networks

Analysis of Large-Scale Traffic Dynamics in an Urban Transportation Network Using Non-Negative Tensor Factorization

Contact Info

Product

Resources

About