Tail Node Research Articles

A Packet Content-Oriented Remote Code Execution Attack Payload Detection Model

In recent years, various Remote Code Execution vulnerabilities on the Internet have been exposed frequently; thus, more and more security researchers have begun to pay attention to the detection of Remote Code Execution attacks. In this paper, we focus on three kinds of common Remote Code Execution attacks: XML External Entity, Expression Language Injection, and Insecure Deserialization. We propose a packet content-oriented Remote Code Execution attack payload detection model. For the XML External Entity attack, we propose an algorithm to construct the use-definition chain of XML entities, and implement detection based on the integrity of the chain and the behavior of the chain’s tail node. For the Expression Language Injection and Insecure Deserialization attack, we extract 34 features to represent the string operation and the use of sensitive classes/methods in the code, and then train a machine learning model to implement detection. At the same time, we build a dataset to evaluate the effect of the proposed model. The evaluation results show that the model performs well in detecting XML External Entity attacks, achieving a precision of 0.85 and a recall of 0.94. Similarly, the model performs well in detecting Expression Language Injection and Insecure Deserialization attacks, achieving a precision of 0.99 and a recall of 0.88.

MRME-Net: Towards multi-semantics learning and long-tail problem of efficient event detection from social messages

Discovering trending social events (e.g., major meetings, political scandals, natural disasters, etc.) from social messages is vital because it emphasizes important events and can help people comprehend the world. However, the heterogeneous semantics enrichment, severe long-tail problems, and sparse text contents of social messages pose great challenges to event detection, often leading to limited generalization ability and accuracy. In this paper, we propose a novel Multi-Relational Meta-Enhanced Network (MRME-Net) architecture to learn social events. First, we model social messages into a multi-relational message graph, incorporating abundant meta-semantics along with various meta-relations. Second, we present a multi-relational graph attention network based on Sophia by using a dual-step message aggregation mechanisms to capture the local features of neighboring messages and global semantics of mutiple relations and ultimately learn social message embeddings. We use Sophia optimizer to reduce the massive time and cost of training. Third, in order to address the long-tail problem, we introduce a locally-adapted meta-learning framework in social event detection for the first time and propose a novel META-TAILENH embedding enhancement strategy to refine tail node embeddings in multi-relational graph. Eventually, we conduct the detection of social events according to the hierarchical clustering algorithm. Extensive experiments have been carried out to evaluate MRME-Net on the MAVEN and Twitter dataset, revealing a notable improvement of 3 %–13 %, 4 %–20 % and 6 %–30 % increases on NMI, AMI and ARI in the social event detection task.

Abstract 3528: Constructing knowledge graph of N6-methyladnosion regulations in cancer from literature using ChatGPT-4

Abstract N6-methyladenosine (m6A) is the most abundant mRNA modification with a crucial role in cellular processes. Its involvement in cancers is well-established. Dysregulation of m6A is linked to cancer initiation, progression, and therapy resistance, impacting the tumor microenvironment (TME) and promoting metastasis. Understanding m6A's regulatory role in cancer is essential for identifying new therapeutic targets. We considered the construction of a knowledge graph (KG) representing molecular regulatory mechanisms (MRMs) of m6A in healthy conditions and different cancers from PubMed papers. Such a KG consists of interconnected triplets of (head node, regulatory relationship, tail node), each denoting regulation between genes or a gene and a phenotype/process. Advances in natural language processing (NLP) have provided tools such as SemRep and GNBR to automate the extraction of these triplets from the literature. However, these methods don’t extract contextual information, such as the type of cancers or cells in which these MRMs occur, leading to contradictory regulations in the constructed KG. In addition, existing approaches were designed to extract generic regulatory relations and therefore struggled in properly capturing unique concepts associated with m6A regulatory mechanisms. To tackle this challenge, we designed a novel ChatGPT-4 prompt to extract relational graphs from papers tailored for m6A regulatory mechanisms. An annotated dataset of 400 titles on m6A-related MRMs were created to evaluate the performance. We then applied the prompt to 1023 papers to create a m6A Molecular Regulatory Mechanism Knowledge Graph (m6A-MRM-KG). This graph illuminates m6A's roles in gene expression regulation, especially in cancer and immunity. Anticipated to enhance our understanding of cancer development, it provides insights into potential therapeutic strategies. The m6A-MRM-KG not only organizes information but also empowers researchers to uncover novel insights in the dynamic relationships of m6A modifications in diverse biological contexts. Citation Format: Sumin Jo, Xidong Wu, Yiming Zeng, Arun Das, Ting-He Zhang, David Alexander Spellman, Adam Ferris, Shou-Jiang Gao, Jianqiu (Michelle) Zhang, Yu-Chiao Chiu, Yufei Huang. Constructing knowledge graph of N6-methyladnosion regulations in cancer from literature using ChatGPT-4 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3528.

Asymmetric Learning for Graph Neural Network based Link Prediction

Link prediction is a fundamental problem in many graph-based applications, such as protein-protein interaction prediction. Recently, graph neural network (GNN) has been widely used for link prediction. However, existing GNN-based link prediction (GNN-LP) methods suffer from scalability problem during training for large-scale graphs, which has received little attention from researchers. In this paper, we first analyze the computation complexity of existing GNN-LP methods, revealing that one reason for the scalability problem stems from their symmetric learning strategy in applying the same class of GNN models to learn representation for both head nodes and tail nodes. We then propose a novel method, called a sym m etric l earning (AML), for GNN-LP. More specifically, AML applies a GNN model to learn head node representation while applying a multi-layer perceptron (MLP) model to learn tail node representation. To the best of our knowledge, AML is the first GNN-LP method to adopt an asymmetric learning strategy for node representation learning. Furthermore, we design a novel model architecture and apply a row-wise mini-batch sampling strategy to ensure promising model accuracy and training efficiency for AML. Experiments on three real large-scale datasets show that AML is 1.7×∼7.3× faster in training than baselines with a symmetric learning strategy while having almost no accuracy loss.

DAHGN: Degree-Aware Heterogeneous Graph Neural Network

In recent years, Graph Neural Networks (GNNs), an emerging technology for learning from graph-structured data, have attracted much attention. Despite the widespread applications of GNNs across diverse domains, recent studies have shown that insufficient information aggregation for low-degree nodes (i.e., tail nodes) leads to inferior performance of GNNs. While numerous recent studies have aimed to address the issue of degree bias in homogeneous graphs, there are limited studies on analyzing such unfairness for heterogeneous information networks. In this article, we propose a novel semi-supervised contrastive learning framework named Degree-Aware Heterogeneous Graph Neural Network (DAHGN), which achieves promising performance by alleviating the degree bias. In brief, we first construct two contrastive graph views, one from a heterogeneous graph that focuses on capturing complementary information to enhance the node representations and one from an extracted homogeneous subgraph that focuses on applying different strategies to low- and high- degree nodes to eliminate the degree bias. We then leverage the label learned from the heterogeneous view to reconstruct the positive node pairs in the contrastive learning framework. Finally, we combine the task loss (semi-supervised) and the contrastive loss to optimize the model performance. We conduct extensive experiments on benchmark datasets to validate the effectiveness of our model, and the results demonstrate that the DAHGN outperforms a wide range of state-of-the-art methods in the semi-supervised node classification problem.

Multi-Label Classification With Dual Tail-Node Augmentation for Drug Repositioning.

Due to the lengthy and costly process of new drug discovery, increasing attention has been paid to drug repositioning, i.e., identifying new drug-disease associations. Current machine learning methods for drug repositioning mainly leverage matrix factorization or graph neural networks, and have achieved impressive performance. However, they often suffer from insufficient training labels of inter-domain associations, while ignore the intra-domain associations. Moreover, they often neglect the importance of tail nodes that have few known associations, which limits their effectiveness in drug repositioning. In this paper, we propose a novel multi-label classification model with dual Tail-Node Augmentation for Drug Repositioning (TNA-DR). We incorporate disease-disease similarity and drug-drug similarity information into k-nearest neighbor ( kNN) augmentation module and contrastive augmentation module, respectively, which effectively complements the weak supervision of drug-disease associations. Furthermore, before employing the two augmentation modules, we filter the nodes by their degrees, so that the two modules are only applied to tail nodes. We conduct 10-fold cross validation experiments on four different real-world datasets, and our model achieves the state-of-the-art performance on all the four datasets. We also demonstrate our model's capability of identifying drug candidates for new diseases and discovering potential new links between existing drugs and diseases.

HeadTailTransfer: An efficient sampling method to improve the performance of graph neural network method in predicting sparse ncRNA–protein interactions

Noncoding RNA (ncRNA) is a functional RNA derived from DNA transcription, and most transcribed genes are transcribed into ncRNA. ncRNA is not directly involved in the translation of proteins, but it can participate in gene expression in cells and affect protein synthesis, thus playing an important role in biological processes such as growth, proliferation, metabolism, and information transmission. Therefore, understanding the interaction between ncRNA and protein is the basis for studying ncRNA regulation of protein-related biological activities. However, it is very expensive and time-consuming to verify ncRNA–protein interaction through biological experiments, and prediction methods based on machine learning have been developed rapidly. Recently, the graph neural network model (GNN) stands out for its excellent performance, but lacks a general framework for predicting ncRNA–protein interactions. We propose a GNN-based framework to predict ncRNA–protein interactions, which can utilize topological structure information to complete prediction tasks faster and more accurately. Meanwhile, for some smaller datasets, many ncRNA nodes lack neighbor information, resulting in lower prediction accuracy. For some larger datasets, the long-tail distribution causes the prediction of the tail nodes (sparse nodes linking few neighbors) to be affected. Therefore, we propose a new sampling method named HeadTailTransfer to mitigate these effects. Experimental results illustrate the effectiveness of this method. Especially for task-specific prediction on the RPI369 dataset in the Graphsage-based neural network framework, the AUC and ACC values increased from 56.8% and 52.2% to 80.2% and 71.8%, respectively. Our data and codes are available: https://github.com/kkkayle/HeadTailTransfer.

Modified Pulat’s algorithm for the maximum outflow problem in undirected generalized networks

A generalized network is characterized by arc multipliers that specify which portion of the flow entering an arc at its tail node reaches its head node. In this paper, the goal is to maximize the flow excess at the sink in undirected generalized networks. We show that the results of the directed networks are not satis ed here. We have speci cally stated and proved the optimality conditions of the problem. We have also designed an algorithm according to Pulat's algorithm and achieved an efficient version.

Energy-Efficient and Privacy-Preserving Data Aggregation Algorithm for Wireless Sensor Networks

Privacy-preserving data aggregation is a kind of fundamental and essential algorithm for wireless sensor networks. However, the existing aggregation algorithms consume a large amount of energy to assure sensory data security. In this article, we propose an energy-efficient and privacy-preserving data aggregation algorithm (EPDA). We organize a sensor network into a tree and connect the leaf nodes of the tree to form many chains. EPDA requires only the data sensed by the tail nodes of the chains to be sliced to ensure privacy. Also, EPDA significantly decreases energy consumption and prolongs the lifetime of the network. We compare our scheme with the existing schemes through theoretical analysis and simulations. The analysis and simulation results show that EPDA outperforms the existing schemes.

A Novel Energy-Efficient and Privacy-Preserving Data Aggregation for WSNs

Data aggregation is a fundamental and efficient algorithm to reduce the communication overhead and energy consumption in wireless sensor networks (WSNs). However, WSNs need both energy-efficient and privacy-preserving when being deployed in an unprotected area. In this paper, we propose an energy-efficient and privacy-preserving data aggregation algorithm CBDA (the chain-based data aggregation). In the CBDA, sensor nodes will be organized as a tree topology. The leaf nodes of the tree sequentially reconnect with each other to form many chain topologies. For assuring data privacy, after data gathering, (1) the tail nodes (the nodes which on the tail of chain) need to slice their sensing data into J fragments. One fragment is kept by themselves, and they distribute the J -1 data fragments to their neighbor nodes. (2) Each tail node will inject some fake fragments into its J -1 fragments to interfere with adversaries. The CBDA can achieve less energy consumption and higher aggregation accuracy during data aggregation. We perform a comprehensive simulation to make a comparison among the CBDA with existing algorithms. The experimental results demonstrate that the CBDA outperforms the existing algorithms.

Social Influence Maximization in Hypergraph in Social Networks

Crowd psychology plays an important role in determining the kind of activities that a person performs. In reality, in a social network, crowd influence has been observed and it cannot be ignored when considering information diffusion problems. In this paper, we model crowd influence as a hyperedge e = (H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e</sub> , v) with weight 0 ≤ P <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e</sub> ≤ 1, where H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e</sub> is the head node set and v is the tail node, means v will be activated by H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e</sub> with probability P <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e</sub> only after each node in H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e</sub> is activated. Then, the Social Influence Maximization Problem in Hypergraph (SIMPH) aims to select k initially-influenced seed users in a directed hypergraph G = (V, E, P). The objective is to maximize the expected number of eventually-influenced users. We show that SIMPH is NP-hard and the objective function is neither submodular nor supermodular. We develop a lower bound and an upper bound that are submodular. We prove that maximizing these two bounds are still NP-hard under IC model. Then, we present a D-SSA algorithm for general weighted social influence maximization problem preserving (1 - 1=e - ε)-approximation. We formulate a sandwich approximation framework, which preserves a theoretical analysis result. Finally, we evaluate our algorithm on real world data sets. The results show the effectiveness and the efficiency of the proposed algorithm.

Generalized Dynamic Flow on Lossy Network

Despite of implicit flow conservation on every arc of traditional network flow model, the generalized network flow model assumes the proportional and symmetric loss factor on each arc. This paper considers two terminal lossy networks specifying the portion of flow entering an arc at its tail node that reaches to its head node. We studied the generalized maximum continuous dynamic contraflow (GMCDCF) and generalized continuous earliest arrival contraflow (GCEACF) problems. The problems are efficiently solved with pseudo-polynomial time algorithms. Moreover, a fully polynomial time approximation scheme (FPTAS) is proposed in polynomial time.

Security Analysis of a Distributed Networked System Under Eavesdropping Attacks

In this brief, we consider the security problem for a network of linear time-invariant dynamical systems under eavesdropping attacks, where an eavesdropper can measure the states of some nodes in the network. We provide a necessary and sufficient condition for the eavesdropper to reveal the states of the networked system. We also present some necessary conditions for the case when the network has tail nodes or at least one node not monitored by the eavesdropper. In addition, we investigate several typical network structures and show that the networked system can be completely observed provided that the eavesdropper can monitor one node in each cycle. Further, we characterize the minimal set of necessarily monitored nodes with the upper bound set of such nodes. Finally, we verify the theoretical results by numerical simulations.

Near-Optimal Convergent Approach for Composed Influence Maximization Problem in Social Networks

Crowd psychology is a critical factor when considering information diffusion, which has been modeled as composed influence. The composed influence is represented as a hyperedge in a graph model. A hyperedge e = (He, v) contains the head node set He and tail node v. Then a social network is modeled as a hypergraph. e can only propagate this influence when all nodes in He become active first. In this paper, the Composed Influence Maximization (CIM) also aims to select k initially-influenced seed users in such a social network. The objective is to maximize the expected number of eventually-influenced users. We present an approximating method for this objective function by formulating a series of submodular functions and these functions are convergent. Then, we develop a lower bound and an upper bound problems whose objective functions are submodular. We design a greedy strategy based on the lower bound maximization for solving CIM. We formulate a sandwich approximation framework, which preserves the theoretical analysis result. Finally, we evaluate our algorithm on real world data sets. The results show the effectiveness and the efficiency of the proposed algorithm.

Path-based capacity-restrained dynamic traffic assignment algorithm

ABSTRACTThis study proposes a new fast convergence path-based dynamic traffic assignment algorithm. The algorithm is based on an analytical capacity-restrained dynamic user equilibrium model, in which link travel times are temporal functions of link flows. In previous models the length of time intervals has to be much longer (about five times) than the link travel times in each time interval using the simplifying assumption that each link will be fully traversed within the time interval that a path reaches its tail node. Otherwise, these models will make large errors and will exhibit difficulties in attaining optimality. This paper tackles the problem by proposing a systematic way of link segmentation. The numerical and case study results of the proposed algorithm show its excellent performance. The algorithm rapidly converges to optimality and provides path information over time intervals, which make it suitable for evaluating the intelligent transportation systems in strategic transport planning.

Synchrony and Elementary Operations on Coupled Cell Networks

Given a finite graph (network), let every node (cell) represent an individual dynamics given by a system of ordinary differential equations, and every arrow (edge) encode the dynamical influence of the tail node on the head node. We have then defined a coupled cell system that is associated with the given network structure. Subspaces that are defined by equalities of cell coordinates and left invariant under every coupled cell system respecting the network structure are called synchrony subspaces. They are completely determined by the network structure and form a complete lattice under set inclusions. We analyze the transition of the lattice of synchrony subspaces of a network that is caused by structural changes in the network topology, such as deletion and addition of cells or edges, and rewirings of edges. We give sufficient, and in some cases both sufficient and necessary, conditions under which lattice elements persist or disappear.

Communication Strategies of Multi-subjects for Micro Public Welfare from the Perspective of Social Network Relationship

From the research perspective of social network relationship, this paper analyzes the roles that are respectively played by information bridge node, core node and long tail node. Then, it probes into the fact that media professionals, public welfare organizations and MicroBlog citizen individuals in communities of network public welfare respectively take advantage of the relationship of the social network they are located in and exploit opportunities of new media empowerment to realize group mobilization and resource integration. Especially, the paper focuses on making a thorough inquiry into the communication strategies used by multi-subjects in micro public welfare based on social network relationship.

A 52 GHz Frequency Synthesizer Featuring a 2nd Harmonic Extraction Technique That Preserves VCO Performance

This paper introduces a 2nd harmonic extraction technique and its implementation in a 46.4-58.1 GHz frequency synthesizer. The frequency doubling approach is based on tapping second harmonic signals at the VCO supply and tail nodes and amplifying them to provide a differential output. Since the amplifiers do not load the VCO outputs, the proposed technique does not affect either the tuning range or the frequency of the VCO. Moreover, a novel noise bypass technique is utilized to ensure that the amplifiers do not degrade the VCO phase noise. As a result, the frequency synthesizer achieves 22.4% tuning range (46.4-58.1 GHz) and phase noise below -118 dBc/Hz while consuming 66 mW from a 1 V supply. The stacked common gate amplifier can also be utilized for voltage regulation, providing a relatively constant FOM performance over a 2X power dissipation range. The synthesizer occupies 0.6 mm ×1 mm in IBM 32 nm SOI CMOS.

Generalized max flow in series–parallel graphs

In the generalized max flow problem, the aim is to find a maximum flow in a generalized network, i.e., a network with multipliers on the arcs that specify which portion of the flow entering an arc at its tail node reaches its head node. We consider this problem for the class of series–parallel graphs. First, we study the continuous case of the problem and prove that it can be solved using a greedy approach. Based on this result, we present a combinatorial algorithm that runs in O(nm+mlogm) time, where m is the number of arcs, and a dynamic programming algorithm with running time O(mlogm). For the integral version of the problem, which is known to be NP-complete, we present a pseudo-polynomial algorithm.

AT-Mine: An Efficient Algorithm of Frequent Itemset Mining on Uncertain Dataset

Frequent itemset/pattern mining (FIM) over uncertain transaction dataset is a fundamental task in data mining. In this paper, we study the problem of FIM over uncertain datasets. There are two main approaches for FIM: the level-wise approach and the pattern-growth approach. The level-wise approach requires multiple scans of dataset and generates candidate itemsets. The pattern-growth approach requires a large amount of memory and computation time to process tree nodes because the current algorithms for uncertain datasets cannot create a tree as compact as the original FP-Tree. In this paper, we propose an array based tail node tree structure (namely AT-Tree) to maintain transaction itemsets, and a pattern-growth based algorithm named AT-Mine for FIM over uncertain dataset. AT-Tree is created by two scans of dataset and it is as compact as the original FP-Tree. AT-Mine mines frequent itemsets from AT-Tree without additional scan of dataset. We evaluate our algorithm using sparse and dense datasets; the experimental results show that our algorithm has achieved better performance than the state-of-the-art FIM algorithms on uncertain transaction datasets, especially for small minimum expected support number.