arXiv subjects

Esra Akbas

Publications and source records attributed to Esra Akbas.

At least 19 recordsLinked to original sources

arXiv2026

Cross-attentive Cohesive Subgraph Embedding to Mitigate Oversquashing in GNNs

Graph neural networks (GNNs) have achieved strong performance across various real-world domains. Nevertheless, they suffer from oversquashing, where long-range information is distorted as it is compressed through limited message-passing pathways. This bottleneck limits their ability to capture essential global context and decreases their performance, particularly in dense and heterophilic regions of graphs. To address this issue, we propose a novel graph learning framework that enriches node embeddings via cross-attentive cohesive subgraph representations to mitigate the impact of excessive long-range dependencies. This framework enhances the node representation by emphasizing cohesive structure in long-range information but removing noisy or irrelevant connections. It preserves essential global context without overloading the narrow bottlenecked channels, which further mitigates oversquashing. Extensive experiments on multiple benchmark datasets demonstrate that our model achieves consistent improvements in classification accuracy over standard baseline methods.

cs.LG↗

arXiv2025

HyperGCL: Multi-Modal Graph Contrastive Learning via Learnable Hypergraph Views

Recent advancements in Graph Contrastive Learning (GCL) have demonstrated remarkable effectiveness in improving graph representations. However, relying on predefined augmentations (e.g., node dropping, edge perturbation, attribute masking) may result in the loss of task-relevant information and a lack of adaptability to diverse input data. Furthermore, the selection of negative samples remains rarely explored. In this paper, we introduce HyperGCL, a novel multimodal GCL framework from a hypergraph perspective. HyperGCL constructs three distinct hypergraph views by jointly utilizing the input graph's structure and attributes, enabling a comprehensive integration of multiple modalities in contrastive learning. A learnable adaptive topology augmentation technique enhances these views by preserving important relations and filtering out noise. View-specific encoders capture essential characteristics from each view, while a network-aware contrastive loss leverages the underlying topology to define positive and negative samples effectively. Extensive experiments on benchmark datasets demonstrate that HyperGCL achieves state-of-the-art node classification performance.

cs.LG↗

arXiv2025

Self-Clustering Graph Transformer Approach to Model Resting-State Functional Brain Activity

Resting-state functional magnetic resonance imaging (rs-fMRI) offers valuable insights into the human brain's functional organization and is a powerful tool for investigating the relationship between brain function and cognitive processes, as it allows for the functional organization of the brain to be captured without relying on a specific task or stimuli. In this study, we introduce a novel attention mechanism for graphs with subnetworks, named Self-Clustering Graph Transformer (SCGT), designed to handle the issue of uniform node updates in graph transformers. By using static functional connectivity (FC) correlation features as input to the transformer model, SCGT effectively captures the sub-network structure of the brain by performing cluster-specific updates to the nodes, unlike uniform node updates in vanilla graph transformers, further allowing us to learn and interpret the subclusters. We validate our approach on the Adolescent Brain Cognitive Development (ABCD) dataset, comprising 7,957 participants, for the prediction of total cognitive score and gender classification. Our results demonstrate that SCGT outperforms the vanilla graph transformer method and other recent models, offering a promising tool for modeling brain functional connectivity and interpreting the underlying subnetwork structures.

cs.LG↗

arXiv2025

Telegram as a Battlefield: Kremlin-related Communications during the Russia-Ukraine Conflict

Telegram emerged as a crucial platform for both parties during the conflict between Russia and Ukraine. Per its minimal policies for content moderation, Pro-Kremlin narratives and potential misinformation were spread on Telegram, while anti-Kremlin narratives with related content were also propagated, such as war footage, troop movements, maps of bomb shelters, and air raid warnings. This paper presents a dataset of posts from both pro-Kremlin and anti-Kremlin Telegram channels, collected over a period spanning a year before and a year after the Russian invasion. The dataset comprises 404 pro-Kremlin channels with 4,109,645 posts and 114 anti-Kremlin channels with 1,117,768 posts. We provide details on the data collection process, processing methods, and dataset characterization. Lastly, we discuss the potential research opportunities this dataset may enable researchers across various disciplines.

cs.SI↗

arXiv2024

Tackling Oversmoothing in GNN via Graph Sparsification: A Truss-based Approach

Graph Neural Network (GNN) achieves great success for node-level and graph-level tasks via encoding meaningful topological structures of networks in various domains, ranging from social to biological networks. However, repeated aggregation operations lead to excessive mixing of node representations, particularly in dense regions with multiple GNN layers, resulting in nearly indistinguishable embeddings. This phenomenon leads to the oversmoothing problem that hampers downstream graph analytics tasks. To overcome this issue, we propose a novel and flexible truss-based graph sparsification model that prunes edges from dense regions of the graph. Pruning redundant edges in dense regions helps to prevent the aggregation of excessive neighborhood information during hierarchical message passing and pooling in GNN models. We then utilize our sparsification model in the state-of-the-art baseline GNNs and pooling models, such as GIN, SAGPool, GMT, DiffPool, MinCutPool, HGP-SL, DMonPool, and AdamGNN. Extensive experiments on different real-world datasets show that our model significantly improves the performance of the baseline GNN models in the graph classification task.

cs.LG↗

arXiv2024

DSAM: A Deep Learning Framework for Analyzing Temporal and Spatial Dynamics in Brain Networks

Resting-state functional magnetic resonance imaging (rs-fMRI) is a noninvasive technique pivotal for understanding human neural mechanisms of intricate cognitive processes. Most rs-fMRI studies compute a single static functional connectivity matrix across brain regions of interest, or dynamic functional connectivity matrices with a sliding window approach. These approaches are at risk of oversimplifying brain dynamics and lack proper consideration of the goal at hand. While deep learning has gained substantial popularity for modeling complex relational data, its application to uncovering the spatiotemporal dynamics of the brain is still limited. We propose a novel interpretable deep learning framework that learns goal-specific functional connectivity matrix directly from time series and employs a specialized graph neural network for the final classification. Our model, DSAM, leverages temporal causal convolutional networks to capture the temporal dynamics in both low- and high-level feature representations, a temporal attention unit to identify important time points, a self-attention unit to construct the goal-specific connectivity matrix, and a novel variant of graph neural network to capture the spatial dynamics for downstream classification. To validate our approach, we conducted experiments on the Human Connectome Project dataset with 1075 samples to build and interpret the model for the classification of sex group, and the Adolescent Brain Cognitive Development Dataset with 8520 samples for independent testing. Compared our proposed framework with other state-of-art models, results suggested this novel approach goes beyond the assumption of a fixed connectivity matrix and provides evidence of goal-specific brain connectivity patterns, which opens up the potential to gain deeper insights into how the human brain adapts its functional connectivity specific to the task at hand.

q-bio.NC↗

arXiv2024

DyGCL: Dynamic Graph Contrastive Learning For Event Prediction

Predicting events such as political protests, flu epidemics, and criminal activities is crucial to proactively taking necessary measures and implementing required responses to address emerging challenges. Capturing contextual information from textual data for event forecasting poses significant challenges due to the intricate structure of the documents and the evolving nature of events. Recently, dynamic Graph Neural Networks (GNNs) have been introduced to capture the dynamic patterns of input text graphs. However, these models only utilize node-level representation, causing the loss of the global information from graph-level representation. On the other hand, both node-level and graph-level representations are essential for effective event prediction as node-level representation gives insight into the local structure, and the graph-level representation provides an understanding of the global structure of the temporal graph. To address these challenges, in this paper, we propose a Dynamic Graph Contrastive Learning (DyGCL) method for event prediction. Our model DyGCL employs a local view encoder to learn the evolving node representations, which effectively captures the local dynamic structure of input graphs. Additionally, it harnesses a global view encoder to perceive the hierarchical dynamic graph representation of the input graphs. Then we update the graph representations from both encoders using contrastive learning. In the final stage, DyGCL combines both representations using an attention mechanism and optimizes its capability to predict future events. Our extensive experiment demonstrates that our proposed method outperforms the baseline methods for event prediction on six real-world datasets.

cs.SI↗

arXiv2023

HeTriNet: Heterogeneous Graph Triplet Attention Network for Drug-Target-Disease Interaction

Modeling the interactions between drugs, targets, and diseases is paramount in drug discovery and has significant implications for precision medicine and personalized treatments. Current approaches frequently consider drug-target or drug-disease interactions individually, ignoring the interdependencies among all three entities. Within human metabolic systems, drugs interact with protein targets in cells, influencing target activities and subsequently impacting biological pathways to promote healthy functions and treat diseases. Moving beyond binary relationships and exploring tighter triple relationships is essential to understanding drugs' mechanism of action (MoAs). Moreover, identifying the heterogeneity of drugs, targets, and diseases, along with their distinct characteristics, is critical to model these complex interactions appropriately. To address these challenges, we effectively model the interconnectedness of all entities in a heterogeneous graph and develop a novel Heterogeneous Graph Triplet Attention Network (\texttt{HeTriNet}). \texttt{HeTriNet} introduces a novel triplet attention mechanism within this heterogeneous graph structure. Beyond pairwise attention as the importance of an entity for the other one, we define triplet attention to model the importance of pairs for entities in the drug-target-disease triplet prediction problem. Experimental results on real-world datasets show that \texttt{HeTriNet} outperforms several baselines, demonstrating its remarkable proficiency in uncovering novel drug-target-disease relationships.

cs.LG↗

arXiv2023

Brain Networks and Intelligence: A Graph Neural Network Based Approach to Resting State fMRI Data

Resting-state functional magnetic resonance imaging (rsfMRI) is a powerful tool for investigating the relationship between brain function and cognitive processes as it allows for the functional organization of the brain to be captured without relying on a specific task or stimuli. In this paper, we present a novel modeling architecture called BrainRGIN for predicting intelligence (fluid, crystallized, and total intelligence) using graph neural networks on rsfMRI derived static functional network connectivity matrices. Extending from the existing graph convolution networks, our approach incorporates a clustering-based embedding and graph isomorphism network in the graph convolutional layer to reflect the nature of the brain sub-network organization and efficient network expression, in combination with TopK pooling and attention-based readout functions. We evaluated our proposed architecture on a large dataset, specifically the Adolescent Brain Cognitive Development Dataset, and demonstrated its effectiveness in predicting individual differences in intelligence. Our model achieved lower mean squared errors and higher correlation scores than existing relevant graph architectures and other traditional machine learning models for all of the intelligence prediction tasks. The middle frontal gyrus exhibited a significant contribution to both fluid and crystallized intelligence, suggesting their pivotal role in these cognitive processes. Total composite scores identified a diverse set of brain regions to be relevant which underscores the complex nature of total intelligence.

cs.LG↗

arXiv2023

Topology-guided Hypergraph Transformer Network: Unveiling Structural Insights for Improved Representation

Hypergraphs, with their capacity to depict high-order relationships, have emerged as a significant extension of traditional graphs. Although Graph Neural Networks (GNNs) have remarkable performance in graph representation learning, their extension to hypergraphs encounters challenges due to their intricate structures. Furthermore, current hypergraph transformers, a special variant of GNN, utilize semantic feature-based self-attention, ignoring topological attributes of nodes and hyperedges. To address these challenges, we propose a Topology-guided Hypergraph Transformer Network (THTN). In this model, we first formulate a hypergraph from a graph while retaining its structural essence to learn higher-order relations within the graph. Then, we design a simple yet effective structural and spatial encoding module to incorporate the topological and spatial information of the nodes into their representation. Further, we present a structure-aware self-attention mechanism that discovers the important nodes and hyperedges from both semantic and structural viewpoints. By leveraging these two modules, THTN crafts an improved node representation, capturing both local and global topological expressions. Extensive experiments conducted on node classification tasks demonstrate that the performance of the proposed model consistently exceeds that of the existing approaches.

cs.LG↗

arXiv2023

Hypergraph Classification via Persistent Homology

Persistent homology is a mathematical tool used for studying the shape of data by extracting its topological features. It has gained popularity in network science due to its applicability in various network mining problems, including clustering, graph classification, and graph neural networks. The definition of persistent homology for graphs is relatively straightforward, as graphs possess distinct intrinsic distances and a simplicial complex structure. However, hypergraphs present a challenge in preserving topological information since they may not have a simplicial complex structure. In this paper, we define several topological characterizations of hypergraphs in defining hypergraph persistent homology to prioritize different higher-order structures within hypergraphs. We further use these persistent homology filtrations in classifying four different real-world hypergraphs and compare their performance to the state-of-the-art graph neural network models. Experimental results demonstrate that persistent homology filtrations are effective in classifying hypergraphs and outperform the baseline models. To the best of our knowledge, this study represents the first systematic attempt to tackle the hypergraph classification problem using persistent homology.

math.AT↗

arXiv2023

MPool: Motif-Based Graph Pooling

Graph Neural networks (GNNs) have recently become a powerful technique for many graph-related tasks including graph classification. Current GNN models apply different graph pooling methods that reduce the number of nodes and edges to learn the higher-order structure of the graph in a hierarchical way. All these methods primarily rely on the one-hop neighborhood. However, they do not consider the higher- order structure of the graph. In this work, we propose a multi-channel Motif-based Graph Pooling method named (MPool) captures the higher-order graph structure with motif and local and global graph structure with a combination of selection and clustering-based pooling operations. As the first channel, we develop node selection-based graph pooling by designing a node ranking model considering the motif adjacency of nodes. As the second channel, we develop cluster-based graph pooling by designing a spectral clustering model using motif adjacency. As the final layer, the result of each channel is aggregated into the final graph representation. We perform extensive experiments on eight benchmark datasets and show that our proposed method shows better accuracy than the baseline methods for graph classification tasks.

cs.LG↗

arXiv2023

Seq-HyGAN: Sequence Classification via Hypergraph Attention Network

Sequence classification has a wide range of real-world applications in different domains, such as genome classification in health and anomaly detection in business. However, the lack of explicit features in sequence data makes it difficult for machine learning models. While Neural Network (NN) models address this with learning features automatically, they are limited to capturing adjacent structural connections and ignore global, higher-order information between the sequences. To address these challenges in the sequence classification problems, we propose a novel Hypergraph Attention Network model, namely Seq-HyGAN. To capture the complex structural similarity between sequence data, we first create a hypergraph where the sequences are depicted as hyperedges and subsequences extracted from sequences are depicted as nodes. Additionally, we introduce an attention-based Hypergraph Neural Network model that utilizes a two-level attention mechanism. This model generates a sequence representation as a hyperedge while simultaneously learning the crucial subsequences for each sequence. We conduct extensive experiments on four data sets to assess and compare our model with several state-of-the-art methods. Experimental results demonstrate that our proposed Seq-HyGAN model can effectively classify sequence data and significantly outperform the baselines. We also conduct case studies to investigate the contribution of each module in Seq-HyGAN.

cs.LG↗

arXiv2022

DDI Prediction via Heterogeneous Graph Attention Networks

Polypharmacy, defined as the use of multiple drugs together, is a standard treatment method, especially for severe and chronic diseases. However, using multiple drugs together may cause interactions between drugs. Drug-drug interaction (DDI) is the activity that occurs when the impact of one drug changes when combined with another. DDIs may obstruct, increase, or decrease the intended effect of either drug or, in the worst-case scenario, create adverse side effects. While it is critical to detect DDIs on time, it is timeconsuming and expensive to identify them in clinical trials due to their short duration and many possible drug pairs to be considered for testing. As a result, computational methods are needed for predicting DDIs. In this paper, we present a novel heterogeneous graph attention model, HAN-DDI to predict drug-drug interactions. We create a heterogeneous network of drugs with different biological entities. Then, we develop a heterogeneous graph attention network to learn DDIs using relations of drugs with other entities. It consists of an attention-based heterogeneous graph node encoder for obtaining drug node representations and a decoder for predicting drug-drug interactions. Further, we utilize comprehensive experiments to evaluate of our model and to compare it with state-of-the-art models. Experimental results show that our proposed method, HAN-DDI, outperforms the baselines significantly and accurately predicts DDIs, even for new drugs.

cs.LG↗

arXiv2022

HyGNN: Drug-Drug Interaction Prediction via Hypergraph Neural Network

Drug-Drug Interactions (DDIs) may hamper the functionalities of drugs, and in the worst scenario, they may lead to adverse drug reactions (ADRs). Predicting all DDIs is a challenging and critical problem. Most existing computational models integrate drug-centric information from different sources and leverage them as features in machine learning classifiers to predict DDIs. However, these models have a high chance of failure, especially for the new drugs when all the information is not available. This paper proposes a novel Hypergraph Neural Network (HyGNN) model based on only the SMILES string of drugs, available for any drug, for the DDI prediction problem. To capture the drug similarities, we create a hypergraph from drugs' chemical substructures extracted from the SMILES strings. Then, we develop HyGNN consisting of a novel attention-based hypergraph edge encoder to get the representation of drugs as hyperedges and a decoder to predict the interactions between drug pairs. Furthermore, we conduct extensive experiments to evaluate our model and compare it with several state-of-the-art methods. Experimental results demonstrate that our proposed HyGNN model effectively predicts DDIs and impressively outperforms the baselines with a maximum ROC-AUC and PR-AUC of 97.9% and 98.1%, respectively.

q-bio.QM↗

arXiv2022

Liars are more influential: Effect of Deception in Influence Maximization on Social Networks

Detecting influential users, called the influence maximization problem on social networks, is an important graph mining problem with many diverse applications such as information propagation, market advertising, and rumor controlling. There are many studies in the literature for influential users detection problem in social networks. Although the current methods are successfully used in many different applications, they assume that users are honest with each other and ignore the role of deception on social networks. On the other hand, deception appears to be surprisingly common among humans within social networks. In this paper, we study the effect of deception in influence maximization on social networks. We first model deception in social networks. Then, we model the opinion dynamics on these networks taking the deception into consideration thanks to a recent opinion dynamics model via sheaf Laplacian. We then extend two influential node detection methods, namely Laplacian centrality and DFF centrality, for the sheaf Laplacian to measure the effect of deception in influence maximization. Our experimental results on synthetic and real-world networks suggest that liars are more influential than honest users in social networks.

cs.SI↗

arXiv2021

Identifying critical higher-order interactions in complex networks

Information diffusion on networks is an important concept in network science observed in many situations such as information spreading and rumor controlling in social networks, disease contagion between individuals, cascading failures in power grids. The critical interactions in networks are the ones that play critical roles in information diffusion and primarily affect network structure and functions. Besides, interactions can occur between not only two nodes as pairwise interactions, i.e., edges, but also three or more nodes, described as higher-order interactions. This report presents a novel method to identify critical higher-order interactions. We propose two new Laplacians that allow redefining classical graph centrality measures for higher-order interactions. We then compare the redefined centrality measures using the Susceptible-Infected-Recovered (SIR) simulation model. Experimental results suggest that the proposed method is promising in identifying critical higher-order interactions.

cs.SI↗

arXiv2021

Hypergraph Laplacians in Diffusion Framework

Networks are important structures used to model complex systems where interactions take place. In a basic network model, entities are represented as nodes, and interaction and relations among them are represented as edges. However, in a complex system, we cannot describe all relations as pairwise interactions, rather should describe as higher-order interactions. Hypergraphs are successfully used to model higher-order interactions in complex systems. In this paper, we present two new hypergraph Laplacians based on diffusion framework. Our Laplacians take the relations between higher-order interactions into consideration, hence can be used to model diffusion on hypergraphs not only between vertices but also higher-order structures. These Laplacians can be employed in different network mining problems on hypergraphs, such as social contagion models on hypergraphs, influence study on hypergraphs, and hypergraph classification, to list a few.

cs.SI↗