Search

Population structure has been known to substantially affect evolutionary dynamics. Networks that promote the spreading of fitter mutants are called amplifiers of selection, and those that suppress the spreading of fitter mutants are called suppressors of selection. Research in the past two decades has found various families of amplifiers while suppressors still remain somewhat elusive. It has also been discovered that most networks are amplifiers of selection under the birth-death updating combined with uniform initialization, which is a standard condition assumed widely in the literature. In the present study, we extend the birth-death processes to temporal (i.e., time-varying) networks. For the sake of tractability, we restrict ourselves to switching temporal networks, in which the network structure deterministically alternates between two static networks at constant time intervals or stochastically in a Markovian manner. We show that, in a majority of cases, switching networks are less amplifying than both of the two static networks constituting the switching networks. Furthermore, most small switching networks, i.e., networks on six nodes or less, are suppressors, which contrasts to the case of static networks. [ABSTRACT FROM AUTHOR]

One perspective to view the economic development of cities is through the presence of multinational firms; how subsidiaries of various organizations are set up throughout the globe and how cities are connected to each other through these networks of multinational firms. Analysis of these networks can reveal interesting economical and spatial trends, as well as help us understand the importance of cities in national and regional economic development. This paper aims to study networks of cities formed due to the linkages of multinational firms over a decade (from 2010 to 2019). More specifically we are interested in analyzing the growth and stability of various cities in terms of the connections they form with other cities over time. Our results can be summarized into two key findings: First, we ascertain the central position of several cities due to their economically stable connections; Second, we successfully identify cities that have evolved over the past decade as the presence of multinational firms has increased in these cities. [ABSTRACT FROM AUTHOR]

Abstract Human communication, the essence of collective social phenomena ranging from small-scale organizations to worldwide online platforms, features intense reciprocal interactions between members in order to achieve stability, cohesion, and cooperation in social networks. While high levels of reciprocity are well known in aggregated communication data, temporal patterns of reciprocal information exchange have received far less attention. Here we propose measures of reciprocity based on the time ordering of interactions and explore them in data from multiple communication channels, including calls, messaging and social media. By separating each channel into reciprocal and non-reciprocal temporal networks, we find persistent trends that point to the distinct roles of one-to-one exchange versus information broadcast. We implement several null models of communication activity, which identify memory, a higher tendency to repeat interactions with past contacts, as a key source of temporal reciprocity. When adding memory to a model of activity-driven, time-varying networks, we reproduce the levels of temporal reciprocity seen in empirical data. Our work adds to the theoretical understanding of the emergence of reciprocity in human communication systems, hinting at the mechanisms behind the formation of norms in social exchange and large-scale cooperation.

Graphs have been used in different fields of research for performing structural analysis of various systems. In order to compare the structure of two systems, the correspondence between their graphs has to be verified. The problem of graph matching, especially subgraph isomorphism (SI), has been well studied in case of static graphs. However, many applications require incorporating temporal information, making the corresponding graphs dynamic. In this paper, we apply SI to detect dynamic patterns in dynamic graphs. We propose an algorithm for induced SI to detect all the matchings for a given pattern graph while considering snapshot-based representation of dynamic graphs and taking into account the chronological order of these snapshots. This is the novelty of the proposed approach since the existing state-of-the-art algorithms model dynamic graphs using an aggregated model with time-stamped edges. To the best of our knowledge, there does not exist another approach which considers snapshot-based representation of dynamic pattern and dynamic target graphs for this problem. We discussed the time complexity of our algorithm and tested its performance while comparing it with two existing algorithms using the real-world datasets. It was found that our algorithm is the second best overall in terms of the execution time. The results are promising given the fact that the choice of dynamic graph model affects the algorithmic design for solving the problem of SI. For the applications where aggregated model of dynamic graphs is not applicable and snapshot-based representation is indispensable, our algorithm can be directly applied as opposed to the existing ones. [ABSTRACT FROM AUTHOR]

The ever-growing size of social network information has enhanced research aimed at finding solutions to challenges in this arena. The vastness and complexity of interactions between social network entities render link prediction in these datasets a challenging task. Previous studies often concentrate on only exploring the local node connectivity information neglecting other key network-characterizing properties. In addition, most works assume static networks, yet many real-world graphs evolve. To address these limitations, firstly, we explore topological information from input graph adjacency matrices by computing topological similarity-based convolution feature matrices. Secondly, we leverage the node strength centrality matrix, a more powerful variant of node degree to preserve the node centrality roles and node's structural connectivity information throughout the network. Lastly, we deploy an LSTM layer to explore the underlying network temporal information. The proposed Structural and Topological aware GCN (STP-GCN) is tested on five social network datasets. Based on experimental results, it exhibits a 3% link prediction AUC improvement, negligible training time increment per epoch (0.2s), and a large MSE magnitude (2.5) reduction in structural centrality prediction as compared to the best benchmark. [ABSTRACT FROM AUTHOR]

Many real-world systems can be expressed in temporal networks with nodes playing different roles in structure and function, and edges representing the relationships between nodes. Identifying critical nodes can help us control the spread of public opinions or epidemics, predict leading figures in academia, conduct advertisements for various commodities and so on. However, it is rather difficult to identify critical nodes, because the network structure changes over time in temporal networks. In this paper, considering the sequence topological information of temporal networks, a novel and effective learning framework based on the combination of special graph convolutional and long short-term memory network (LSTM) is proposed to identify nodes with the best spreading ability. The special graph convolutional network can embed nodes in each sequential weighted snapshot and LSTM is used to predict the future importance of timing-embedded features. The effectiveness of the approach is evaluated by a weighted Susceptible-Infected-Recovered model. Experimental results on four real-world temporal networks demonstrate that the proposed method outperforms both traditional and deep learning benchmark methods in terms of the Kendall τ coefficient and top k hit rate. [ABSTRACT FROM AUTHOR]

Representing the nodes of continuous-time temporal graphs in a low-dimensional latent space has wide-ranging applications, from prediction to visualization. Yet, analyzing continuous-time relational data with timestamped interactions introduces unique challenges due to its sparsity. Merely embedding nodes as trajectories in the latent space overlooks this sparsity. However, a natural way to account for this sparsity is to model the uncertainty around the latent positions. In this paper, we propose TGNE (Temporal Gaussian Network Embedding), an innovative method that bridges two distinct strands of literature: the statistical analysis of networks via Latent Space Models (LSM) and temporal graph machine learning. TGNE embeds nodes as piece-wise linear trajectories of Gaussian distributions in the latent space, capturing both structural information and uncertainty around the trajectories. We evaluate TGNE’s effectiveness in reconstructing the original graph and modelling uncertainty. The results demonstrate that TGNE generates time-varying embedding locations that can accurately reconstruct missing parts of the network based on observed ones. Furthermore, the uncertainty estimates align experimentally with the time-varying degree distribution in the network, providing valuable insights into the temporal dynamics of the graph. To facilitate reproducibility, we provide an open-source implementation of TGNE at https://github.com/aida-ugent/tgne/.

One of the key goals of Informetrics is to identify citation-based popular articles among so many other aspects, such as determining popular research topics, identifying influential scholars, and predicting hot trends in science. These can be achieved by applying network science approaches to scientific networks and formulating the problem as a popular (most-cited) node ranking task. To rank the papers based on their future citation gain. In this work a deep learning based framework is proposed. Which helps in automatic node level feature extraction and can make node level prediction in dynamic graphs such as citation networks. To achieve this we have learned global ranking preserve d dimensional node embedding. We have only considered temporal features, which makes it suitable for generalisation to other networks. Although our model can consider node level explicit features also. Further we have given novel cost function which can be easily solve ranking problem for dynamic graphs using probabilistic regression method. Which can be easily optimised. Another novelty of our work is that our model can be trained using different snapshots of the graph and different time. Further trained model can be used to make future prediction. The proposed model has been tested on an arXiv paper citation network using six standard information retrieval-based metrics. The results show that our proposed model outperforms, on average, other state-of-the-art static models as well as dynamic node ranking models. The outcome of this research study leads to informed data-driven decision-making in science, such as the allocation and distribution of research funds and investment in strategic research centers. When considering past time window size as 10 months and making prediction after 10 months our proposed model’s performance on various ranking based evaluation metrics are as follows: AUC-0.974, Kendal’s rank correlation tau-0.455, Precision- 0.643, Novelty-0.0456, Temporal novelty-0.375 and on NDCG-0.949. Our model is able to make long term trend prediction with just training on short time window.

Abstract Temporal social networks of human interactions are preponderant in understanding the fundamental patterns of human behavior. In these networks, interactions occur locally between individuals (i.e., nodes) who connect with each other at different times, culminating into a complex system-wide web that has a dynamic composition. Dynamic behavior in networks occurs not only locally but also at the global level, as systems expand or shrink due either to: changes in the size of node population or variations in the chance of a connection between two nodes. Here, we propose a numerical maximum-likelihood method to estimate population size and the probability of two nodes connecting at any given point in time. An advantage of the method is that it relies only on aggregate quantities, which are easy to access and free from privacy issues. Our approach enables us to identify the simultaneous (rather than the asynchronous) contribution of each mechanism in the densification and sparsification of human contacts, providing a better understanding of how humans collectively construct and deconstruct social networks.

The large-scale real-time sequencing of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) genomes has allowed for rapid identification of concerning variants through phylogenetic analysis. However, the nature of phylogenetic reconstruction is typically static, in that the relationships between taxonomic units, once defined, are not subject to alterations. Furthermore, most phylogenetic methods are intrinsically batch mode in nature, requiring the presence of the entire data set. Finally, the emphasis of phylogenetics is on relating taxonomical units. These characteristics complicate the application of classical phylogenetics methods to represent relationships in molecular data collected from rapidly evolving strains of an etiological agent, such as SARS-CoV-2, since the molecular landscape is updated continuously as samples are collected. In such settings, variant definitions are subject to epistemological constraints and may change as data accumulate. Furthermore, representing within-variant molecular relationships may be as important as representing between variant relationships. This article describes a novel data representation framework called dynamic epidemiological networks (DENs) along with algorithms that underpin its construction to address these issues. The proposed representation is applied to study the molecular development underlying the spread of the COVID-19 (coronavirus disease 2019) pandemic in two countries: Israel and Portugal spanning a 2-year period from February 2020 to April 2022. The results demonstrate how this framework could be used to provide a multiscale representation of the data by capturing molecular relationships between samples as well as those between variants, automatically identifying the emergence of high frequency variants (lineages), including variants of concern such as Alpha and Delta, and tracking their growth. Additionally, we show how analyzing the evolution of the DEN can help identify changes in the viral population that could not be readily inferred from phylogenetic analysis. [ABSTRACT FROM AUTHOR]

In network analysis, real-world systems may be represented via graph models, where nodes and edges represent the set of biological objects (e.g., genes, proteins, molecules) and their interactions, respectively. This representative knowledge-graph model may also consider the dynamics involved in the evolution of the network (i.e., dynamic networks), in addition to a classic static representation (i.e., static networks). Bioinformatics solutions for network analysis allow knowledge extraction from the features related to a single network of interest or by comparing networks of different species. For instance, we may align a network related to a well known species to a more complex one in order to find a match able to support new hypotheses or studies. Therefore, the network alignment is crucial for transferring the knowledge between species, usually from simplest (e.g., rat) to more complex (e.g., human). Methods: In this paper, we present Dynamic Network Alignment based on Temporal Embedding (DANTE), a novel method for pairwise alignment of dynamic networks that applies the temporal embedding to investigate the topological similarities between the two input dynamic networks. The main idea of DANTE is to consider the evolution of interactions and the changes in network topology. Briefly, the proposed solution builds a similarity matrix by integrating the tensors computed via the embedding process and, subsequently, it aligns the pairs of nodes by performing its own iterative maximization function. Results: The performed experiments have reported promising results in terms of precision and accuracy, as well as good robustness as the number of nodes and time points increases. The proposed solution showed an optimal trade-off between sensitivity and specificity on the alignments produced on several noisy versions of the dynamic yeast network, by improving by ∼18.8% (with a maximum of  20.6 % ) the Area Under the Receiver Operating Characteristic (ROC) Curve (i.e., AUC or AUROC), compared to two well known methods: DYNAMAGNA++ and DYNAWAVE. From the point of view of quality, DANTE outperformed these by ∼91% as nodes increase and by ∼75% as the number of time points increases. Furthermore, a ∼23.73% improvement in terms of node correctness was reported with our solution on real dynamic networks. [ABSTRACT FROM AUTHOR]

Human communication, the essence of collective social phenomena ranging from small-scale organizations to worldwide online platforms, features intense reciprocal interactions between members in order to achieve stability, cohesion, and cooperation in social networks. While high levels of reciprocity are well known in aggregated communication data, temporal patterns of reciprocal information exchange have received far less attention. Here we propose measures of reciprocity based on the time ordering of interactions and explore them in data from multiple communication channels, including calls, messaging and social media. By separating each channel into reciprocal and non-reciprocal temporal networks, we find persistent trends that point to the distinct roles of one-to-one exchange versus information broadcast. We implement several null models of communication activity, which identify memory, a higher tendency to repeat interactions with past contacts, as a key source of temporal reciprocity. When adding memory to a model of activity-driven, time-varying networks, we reproduce the levels of temporal reciprocity seen in empirical data. Our work adds to the theoretical understanding of the emergence of reciprocity in human communication systems, hinting at the mechanisms behind the formation of norms in social exchange and large-scale cooperation. [ABSTRACT FROM AUTHOR]

Understanding the evolutionary patterns of real-world complex systems such as human interactions, biological interactions, transport networks, and computer networks is important for our daily lives. Predicting future links among the nodes in these dynamic networks has many practical implications. This research aims to enhance our understanding of the evolution of networks by formulating and solving the link-prediction problem for temporal networks using graph representation learning as an advanced machine learning approach. Learning useful representations of nodes in these networks provides greater predictive power with less computational complexity and facilitates the use of machine learning methods. Considering that existing models fail to consider the temporal dimensions of the networks, this research proposes a novel temporal network-embedding algorithm for graph representation learning. This algorithm generates low-dimensional features from large, high-dimensional networks to predict temporal patterns in dynamic networks. The proposed algorithm includes a new dynamic node-embedding algorithm that exploits the evolving nature of the networks by considering a simple three-layer graph neural network at each time step and extracting node orientation by using Given's angle method. Our proposed temporal network-embedding algorithm, TempNodeEmb, is validated by comparing it to seven state-of-the-art benchmark network-embedding models. These models are applied to eight dynamic protein–protein interaction networks and three other real-world networks, including dynamic email networks, online college text message networks, and human real contact datasets. To improve our model, we have considered time encoding and proposed another extension to our model, TempNodeEmb++. The results show that our proposed models outperform the state-of-the-art models in most cases based on two evaluation metrics. [ABSTRACT FROM AUTHOR]

Within the 6G vision, the future of mobile communication networks is expected to become more complex, heterogeneous, and characterized by denser deployments with a myriad of users in an ever-more dynamic environment. There is an increasing intent to provide services following the microservice architecture, thus gaining from higher scalability and significant reliability. Microservices introduce novel challenges and the level of granularity impacts performances, due to complex composition patterns. This openness in design demands service requirements be heterogeneous and dynamic. To this end, we propose a framework and a mathematical approach to investigate the complex quality of services. We exploit the temporal multilayer network representation and analysis jointly, with the spreading dynamics of user experience. We study the joint impact of structural heterogeneity and the evolutionary dynamics of the temporal multilayer quality network, composed of networked parameters, and a temporal multilayer social network, populated by a social layered structure of users. We conducted simulations to display our findings on how this modeling approach enables evaluation of otherwise-overlooked information on quality arising from a profound investigation of the structural-complexity and social-dynamics measurements. [ABSTRACT FROM AUTHOR]

Many dynamical systems can be successfully analyzed by representing them as networks. Empirically measured networks and dynamic processes that take place in these situations show heterogeneous, non-Markovian, and intrinsically correlated topologies and dynamics. This makes their analysis particularly challenging. Randomized reference models (RRMs) have emerged as a general and versatile toolbox for studying such systems. Defined as random networks with given features constrained to match those of an input (empirical) network, they may, for example, be used to identify important features of empirical networks and their effects on dynamical processes unfolding in the network. RRMs are typically implemented as procedures that reshuffle an empirical network, making them very generally applicable. However, the effects of most shuffling procedures on network features remain poorly understood, rendering their use nontrivial and susceptible to misinterpretation. Here we propose a unified framework for classifying and understanding microcanonical RRMs (MRRMs) that sample networks with uniform probability. Focusing on temporal networks, we survey applications of MRRMs found in the literature, and we use this framework to build a taxonomy of MRRMs that proposes a canonical naming convention, classifies them, and deduces their effects on a range of important network features. We furthermore show that certain classes of MRRMs may be applied in sequential composition to generate new MRRMs from the existing ones surveyed in this article. We finally provide a tutorial showing how to apply a series of MRRMs to analyze how different network features affect a dynamic process in an empirical temporal network. Our taxonomy provides a reference for the use of MRRMs, and the theoretical foundations laid here may further serve as a base for the development of a principled and automatized way to generate and apply randomized reference models for the study of networked systems. [ABSTRACT FROM AUTHOR]

In temporal networks, PageRank-based methods are usually used to calculate the importance of nodes. However, almost all the methods focus on the first-order relationships between nodes while ignore higher-order interactions between nodes in the graph. Considering that temporal motifs are recurring, higher-order and significant network connectivity patterns, which can capture both temporal and higher-order structural features in dynamic networks, this paper proposes a novel Personalized Temporal Motif PageRank (PTMP) algorithm to measure the importance of nodes in temporal networks. Specifically, to capture temporal information and higher-order features, we develop a method extracting temporal motif instances from temporal networks, and design an algorithm to compute the weighted motif adjacency matrix and the diagonal motif out-degree matrix, then define a motif transition matrix, which contains the personalized feature and can be used to compute the importance score of nodes. Finally, we make the steady-state analysis for the PTMP algorithm and compare it with other state-of-the-art baselines on multiple real-world datasets. The experimental results demonstrate that the PTMP algorithm is capable of mining much richer important nodes information accurately and effectively. [ABSTRACT FROM AUTHOR]

Temporality has been recently identified as a useful feature to exploit when controlling a complex network. Empirical evidence has in fact shown that, with respect to their static counterparts, temporal networks (i) are often endowed with larger reachable sets and (ii) require less control energy when steered towards an arbitrary target state. However, to date, we lack conditions guaranteeing that the dimension of the controllable subspace of a temporal network is larger than that of its static counterpart. In this work, we consider the case in which a static network is input connected but not controllable. We show that when the structure of the graph underlying the temporal network remains the same throughout each temporal snapshot while the (nonvanishing) edge weights vary, then the temporal network will be completely controllable almost always, even when its static counterpart is not. An upper bound on the number of snapshots needed to achieve controllability is also provided. [ABSTRACT FROM AUTHOR]

Graph neural networks (GNNs) are rapidly becoming the dominant way to learn on graph-structured data. Link prediction is a near-universal benchmark for new GNN models. Many advanced models such as Dynamic graph neural networks (DGNNs) specifically target dynamic graphs. However, these models, particularly DGNNs, are rarely compared to each other or existing heuristics. Different works evaluate their models in different ways, thus one cannot compare evaluation metrics and their results directly. Motivated by this, we perform a comprehensive comparison study. We compare link prediction heuristics, GNNs, discrete DGNNs, and continuous DGNNs on the dynamic link prediction task. In total we summarize the results of over 3200 experimental runs ( $\approx 1.5$ years of computation time). We find that simple link prediction heuristics perform better than GNNs and DGNNs, different sliding window sizes greatly affect performance, and of all examined graph neural networks, that DGNNs consistently outperform static GNNs. This work is a continuation of our previous work, a foundation of dynamic networks and theoretical review of DGNNs. In combination with our survey, we provide both a theoretical and empirical comparison of DGNNs.

Abstract Path homology is a powerful method for attaching algebraic invariants to digraphs. While there have been growing theoretical developments on the algebro-topological framework surrounding path homology, bona fide applications to the study of complex networks have remained stagnant. We address this gap by presenting an algorithm for path homology that combines efficient pruning and indexing techniques and using it to topologically analyze a variety of real-world complex temporal networks. A crucial step in our analysis is the complete characterization of path homologies of certain families of small digraphs that appear as subgraphs in these complex networks. These families include all digraphs, directed acyclic graphs, and undirected graphs up to certain numbers of vertices, as well as some specially constructed cases. Using information from this analysis, we identify small digraphs contributing to path homology in dimension two for three temporal networks in an aggregated representation and relate these digraphs to network behavior. We then investigate alternative temporal network representations and identify complementary subgraphs as well as behavior that is preserved across representations. We conclude that path homology provides insight into temporal network structure, and in turn, emergent structures in temporal networks provide us with new subgraphs having interesting path homology.

Mobility is a critical element for understanding human contact networks. In many studies, the researchers use random processes to model human mobility. However, people do not move randomly in their environment. Their interactions do not depend only on spatial constraints but on their temporal, social, economic, and cultural activities. The topological structure of the physical and/or proximity contact networks depends, therefore, entirely on the mobility patterns. This paper performs an extensive comparative analysis of real-world temporal contact networks and synthetic networks based on influential mobility models. Results show that the various topological properties of most of the synthetic datasets depart from those observed in real-world contact networks because the randomness of some mobility parameters tends to move away from human contact properties. However, it appears that data generated using Spatio-Temporal Parametric Stepping (STEPS) mobility model reveals similarities with real temporal contact networks such as heavy-tailed distribution of contact duration, frequency of pairs of contacts, and the bursty phenomenon. These results pave the way for further improvement of mobility models to generate meaningful artificial contact networks.

In the last decade, temporal networks and static and temporal hypergraphs have enabled modelling connectivity and spreading processes in a wide array of real-world complex systems such as economic transactions, information spreading, brain activity and disease spreading. In this manuscript, we present the Reticula C++ library and Python package: A comprehensive suite of tools for working with real-world and synthetic static and temporal networks and hypergraphs. This includes various methods of creating synthetic networks and randomised null models based on real-world data, calculating reachability and simulating compartmental models on networks. The library is designed principally on an extensible, cache-friendly representation of networks, with an aim of easing multi-thread use in the high-performance computing environment.

Interactions among individual animals - and between these individuals and their environment - yield complex, multifaceted systems. The development of multilayer network analysis offers a promising new approach for studying animal social behavior and its relation to eco-evolutionary dynamics.

Temporal social networks of human interactions are preponderant in understanding the fundamental patterns of human behavior. In these networks, interactions occur locally between individuals (i.e., nodes) who connect with each other at different times, culminating into a complex system-wide web that has a dynamic composition. Dynamic behavior in networks occurs not only locally but also at the global level, as systems expand or shrink due either to: changes in the size of node population or variations in the chance of a connection between two nodes. Here, we propose a numerical maximum-likelihood method to estimate population size and the probability of two nodes connecting at any given point in time. An advantage of the method is that it relies only on aggregate quantities, which are easy to access and free from privacy issues. Our approach enables us to identify the simultaneous (rather than the asynchronous) contribution of each mechanism in the densification and sparsification of human contacts, providing a better understanding of how humans collectively construct and deconstruct social networks. [ABSTRACT FROM AUTHOR]

We provide a detailed case study showing how social network analysis allows scholars to detect an event affecting the entire historical network under consideration and identify the responsible actors. We study the middle 13th century in Czech lands, where a rigid political structure of noble families surrounding the monarchs led to the uprising of part of the nobility. Having collected data on approximately 2,400 noblemen from 576 charters, we attempted to uncover social network features pointing to the rebellion and expose the noblemen who joined it. We observed, among other such quantifiable features, assortativity increasing before and resetting to random after the rebellion, a drop in the number of stable connections and subgraph similarity between yearly networks and regional titles (burgraves) rising in centrality above royal court officials in that period. The presented methods can be directly translated to other person-document data of comparable or larger sizes, and we hope it can help detect or disambiguate the timing of similar major events and the roles of people involved in them. [ABSTRACT FROM AUTHOR]

Network representation learning is a promising direction towards applying machine learning over graph-structured data. Most of the recent researches focus on embedding static networks to a low dimensional vector space and performing traditional network mining tasks using this latent space. But, to study many complex interactions between real-world entities, we need to model the data into a time-varying network (temporal network), where the edge connectivity patterns may vary with time. We focus on the problem of temporal network representation learning, where the network is represented with edges time-stamped with the time of interaction. We design a random surfing model for the temporal network using a non-homogeneous Markov chain to generate a node similarity matrix. Further, we perform non-linear dimensionality reduction on the node similarity matrix using a denoising deep autoencoder to generate node representations. We also evaluate the quality of the embeddings generated using a temporal link prediction benchmark. [ABSTRACT FROM AUTHOR]

Social Network Analysis (SNA) has enabled researchers to understand and optimize the key dimensions of collaborative learning. A majority of SNA research has so far used static networks, ie, aggregated networks that compile interactions without considering when certain activities or relationships occurred. Compressing a temporal process by discarding time, however, may result in reductionist oversimplifications. In this study, we demonstrate the potentials of temporal networks in the analysis of online peer collaboration. In particular, we study: (1) social interactions by analysing learners' collaborative behaviour, part of a case study in which they worked on academic writing tasks, and (2) cognitive interactions through the analysis of students' self‐regulated learning tactics. The study included 123 students and 2550 interactions. By using temporal networks, we show how to analyse the longitudinal evolution of a collaborative network visually and quantitatively. Correlation coefficients with grades, when calculated with time‐respecting temporal measures of centrality, were more correlated with learning outcomes than traditional centrality measures. Using temporal networks to analyse the co‐temporal and longitudinal development, reach, and diffusion patterns of students' learning tactics has provided novel insights into the complex dynamics of learning, not commonly offered through static networks. [ABSTRACT FROM AUTHOR]