Biased Dynamic Sampling for Temporal Network Streams

Tabassum, Shazia; Gama, João

doi:10.1007/978-3-030-05411-3_42

Cited by 3 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the streaming scenario, the distribution of the incoming data tends to evolve over time, which implies that recent instances are more relevant to describe the state of the stream than older ones [45]. A first approach to deal with this issue could be to use a sliding window with the most recent instances [46].…”

Section: Adlstream Frameworkmentioning

confidence: 99%

Asynchronous dual-pipeline deep learning framework for online data stream classification

Lara-Benítez¹,

Carranza-García²,

García-Gutiérrez³

et al. 2020

ICA

View full text Add to dashboard Cite

Data streaming classification has become an essential task in many fields where real-time decisions have to be made based on incoming information. Neural networks are a particularly suitable technique for the streaming scenario due to their incremental learning nature. However, the high computation cost of deep architectures limits their applicability to high-velocity streams, hence they have not yet been fully explored in the literature. Therefore, in this work, we aim to evaluate the effectiveness of complex deep neural networks for supervised classification in the streaming context. We propose an asynchronous deep learning framework in which training and testing are performed simultaneously in two different processes. The data stream entering the system is dual fed into both layers in order to concurrently provide quick predictions and update the deep learning model. This separation reduces processing time while obtaining high accuracy on classification. Several time-series datasets from the UCR repository have been simulated as streams to evaluate our proposal, which has been compared to other methods such as Hoeffding trees, drift detectors, and ensemble models. The statistical analysis carried out verifies the improvement in performance achieved with our dual-pipeline deep learning framework, that is also competitive in terms of computation time.

show abstract

Section: Adlstream Frameworkmentioning

confidence: 99%

Asynchronous dual-pipeline deep learning framework for online data stream classification

Lara-Benítez¹,

Carranza-García²,

García-Gutiérrez³

et al. 2020

ICA

View full text Add to dashboard Cite

show abstract

“…For predicting recurring links with limited space, efficiency and fastness, we propose our model using the dynamic sampling technique (SBias) given by Tabassum & Gama (2018). If we consider picking a uniform random sample from a power law distribution (which most of the real-world networks follow), it will likely result in getting all low degree nodes from the long tail.…”

Section: Rlp: Recurring Link's Prediction Using Temporal Bias and Frementioning

confidence: 99%

“…On the contrary, the samples generated using (SBias) are proved to preserve certain important properties of temporal networks. They are less biased to disconnected components, low degree nodes, and follow true network distribution (Tabassum & Gama, 2018). Henceforth, we make use of this strategy, and moreover it applies to multi-graphs.…”

Section: Rlp: Recurring Link's Prediction Using Temporal Bias and Frementioning

confidence: 99%

“…A description of the data sets used appears in Table 1. Remember that we define Stability of an edge as the number of time steps it is present in a graph stream over the total number of time steps (Tabassum & Gama, 2018). The average recurrence times of an edge are defined below.…”

Section: Lastfm Bandmentioning

confidence: 99%

“…In an attempt to address the issue of scalability and performance, we implement the method presented by Tabassum & Gama (2018) to provide online samples. Here, we use this methodology to predict recurring links.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On fast and scalable recurring link’s prediction in evolving multi-graph streams

Tabassum¹,

Veloso²,

Gama³

2020

Net Sci

Self Cite

View full text Add to dashboard Cite

The link prediction task has found numerous applications in real-world scenarios. However, in most of the cases like interactions, purchases, mobility, etc., links can re-occur again and again across time. As a result, the data being generated is excessively large to handle, associated with the complexity and sparsity of networks. Therefore, we propose a very fast, memory-less, and dynamic sampling-based method for predicting recurring links for a successive future point in time. This method works by biasing the links exponentially based on their time of occurrence, frequency, and stability. To evaluate the efficiency of our method, we carried out rigorous experiments with massive real-world graph streams. Our empirical results show that the proposed method outperforms the state-of-the-art method for recurring links prediction. Additionally, we also empirically analyzed the evolution of links with the perspective of multi-graph topology and their recurrence probability over time.

show abstract