Directed Graph Model Research Articles

Software and Dependencies in Research Citation Graphs

Following the widespread digitalization of scholarship, software has become essential for research, but the current sociotechnical system of citation does not reflect this sufficiently. Citation provides context for research, but the current model for the respective research citation graphs does not integrate software. In this paper, I develop a directed graph model to alleviate this, describe challenges for its instantiation, and give an outlook of useful applications of research citation graphs, including transitive credit.

Unknown Vulnerability Risk Assessment Based on Directed Graph Models: A Survey

Nowadays, vulnerability attacks occur frequently. Due to the information asymmetry between attackers and defenders, vulnerabilities can be divided into known and unknown. Existing researches mainly focus on the risk assessment of known vulnerabilities. However, unknown vulnerabilities are more threatening and harder to detect. Therefore, unknown vulnerability risk assessment deserves the widespread attention. To model the exploit process, directed graph models are applied to vulnerability risk assessment. And security metrics are used to quantify the exploitability of vulnerabilities. In this paper, according to the data source of nodes, related works of unknown vulnerability risk assessment based on directed graph models are divided into two types. One is based on network-level data, the other is based on system-level data. The former is to visualize the network status, while the latter is to reflect the running process of the system. The concept and purpose of these directed graph models are given at first. Then, these models are analyzed from three aspects, including advantages, flaws and solutions. After that, challenges and solutions of unknown vulnerability risk assessment based on directed graph models are given. Meantime, security metrics for unknown vulnerability risk assessment based on directed graph models are summarized and classified. Finally, future work directions of unknown vulnerability risk assessment are discussed from the perspective of techniques and application trends. Consequently, this paper can fill in the lack of current survey on unknown vulnerability risk assessment based on directed graph models.

Solving the Observing and Downloading Integrated Scheduling Problem of Earth Observation Satellite with a Quantum Genetic Algorithm

Abstract This paper addresses the integrated Earth observation satellite scheduling problem. It is a complicated problem because observing and downloading operations are both involved. We use an acyclic directed graph model to describe the observing and downloading integrated scheduling problem. Based on the model which considering energy constraints and storage capacity constraints, we develop an efficient solving method using a novel quantum genetic algorithm. We design a new encoding and decoding scheme that can generate feasible solution and increase the diversity of the population. The results of the simulation experiments show that the proposed method solves the integrated Earth observation satellite scheduling problem with good performance and outperforms the genetic algorithm and greedy algorithm on all instances.

Trusted Device-to-Device Based Heterogeneous Cellular Networks: A New Framework for Connectivity Optimization

Device-to-device-based heterogeneous cellular networks (HCNs) have received considerable attention recently. However, connectivity optimization is a challenging problem for the existing network architecture that remains open in the literature because it is affected by many uncertain real-life factors, such as locations of femtocell base stations (FemtoBSs), the trust relationships among user equipments (UEs), and the dynamic characteristics of spectrum. In this paper, we propose a connectivity optimization method that jointly considers the influence of these factors. In this method, we first build a new framework called the trusted device-to-device-based HCN to capture the stochastic characteristics of the above factors through three models: the dynamic FemtoBS employment architecture, the trust-based directed graph model, and the dynamic spectrum graph model. Based on this framework, we then formulate the connectivity optimization problem as the maximization of the average number of connected UEs in a given time by selecting the optimal FemtoBSs. Since the maximization problem is proven to be NP-hard, we propose three heuristic algorithms, namely, the simple greedy algorithm, the submodular greedy algorithm, and the particle swarm optimization based FemtoBSs selection algorithm to obtain the suboptimal solutions. The effectiveness of the proposed algorithms is justified through extensive simulations with differently parameterized factors.

Products Ranking Through Aspect-Based Sentiment Analysis of Online Heterogeneous Reviews

With the rapid growth of online shopping platforms, more and more customers intend to share their shopping experience and product reviews on the Internet. Both large quantity and various forms of online reviews bring difficulties for potential consumers to summary all the heterogenous reviews for reference. This paper proposes a new ranking method through online reviews based on different aspects of the alternative products, which combines both objective and subjective sentiment values. Firstly, weights of these aspects are determined with LDA topic model to calculate the objective sentiment value of the product. During this process, the realistic meaning of each aspect is also summarized. Then, consumers’ personalized preferences are taken into consideration while calculating total scores of alternative products. Meanwhile, comparative superiority between every two products also contributes to their final scores. Therefore, a directed graph model is constructed and the final score of each product is computed by improved PageRank algorithm. Finally, a case study is given to illustrate the feasibility and effectiveness of the proposed method. The result demonstrates that while considering only objective sentiment values of the product, the ranking result obtained by our proposed method has a strong correlation with the actual sales orders. On the other hand, if consumers express subjective preferences towards a certain aspect, the final ranking is also consistent with the actual performance of alternative products. It provides a new research idea for online customer review mining and personalized recommendation.

Friend Recommendation Method of Weighted Networks Based on Value and Match

To improve the accuracy of the recommendation in the knowledge social network, the paper proposes a friend recommendation algorithm based on user information to measure the value and the match between the users. On the one hand, it establishes the directed graph model of the value by ‘the number of answers’ and ‘the number of likes’, to calculate the users’ value scores. And it uses the material diffusion principle to predict the value scores which are zero in the model. On the other hand, it calculates the users’ similarity through ‘topic of concern’ and ‘industry’ simulates the similarity of user pairs by using probability distribution and establishes likelihood function. After using ‘EM’ algorithm to estimate the function parameters, it gets the users' matched and mismatched probability to calculate their match scores and establish undirected graph model. Finally, it builds the comprehensive score formula between the value and the match. Combined with the directed and undirected graph model, the Top-N friend recommendation algorithm is recommended. Through the experiments, it shows that the method is better than the recommendation algorithm based on collaborative filtering in Precision, Recall and F1-measure on a large scale recommendation.

Optimization of cluster resource indexing of Internet of Things based on improved ant colony algorithm

In Internet of Things, the resource distribution is random in space, which leads to the poor precision ratio of the cluster resource indexing of Internet of Things, so in order to improve the information fusion and dispatching ability of Internet of Things, it is necessary to optimize the resource indexing of Internet of Things. Therefore, an algorithm for cluster resource indexing of Internet of Things based on improved ant colony algorithm is proposed in this paper. Directed graph models are used to construct a distribution structure model of cluster resource indexing nodes of Internet of Things, carry out semantic association feature extraction in the cluster resource storage information flow of Internet of Things. And the improved ant colony algorithm is used to crawl and capture cluster information in Internet of Things. According to the ant colony trajectory information, the velocity and position of the cluster resource indexing of Internet of Things are updated, and the balanced ant colony algorithm is used to carry out the global search and local search to resources and initialize the clustering center, and the target function of the cluster resource indexing of Internet of Things is constructed and the optimization parameter is solved with the constraint condition of the minimum variance of the whole fitness. The strong ability of global optimization of the ant colony algorithm is used to realize resource indexing optimization. Simulation results show that the improved algorithm can quickly realize resource index convergence, effectively escape local minimum points, and has strong global search ability and relatively high resource indexing precision ratio.

Fault diagnosis strategy of CNC machine tools based on cascading failure

To ensure the safe operation of CNC machines, a fault diagnosis strategy based on cascading failure is proposed. According to fault mechanism analysis, a directed graph model of fault propagation between components in machine tool systems is established. In this study, the interpretative structural model method is used to realize the hierarchical structure of fault propagation model by matrix transformation and decomposition. Subsequently, the PageRank algorithm is introduced to evaluate the failure effects of the machine tool system components. The Johnson method is then applied to correct the component fault sequence and establish the model of rate of occurrence of failures that is based on time correlation. Finally, the fault diagnosis strategy is formulated through the component rate of the occurrence of failure, fault influence and fault propagation model, to identify the main cause of the fault and provide the basis for fault diagnosis. In the end, a machine tool equipment is used as an example for application to verify the validity of the method.

Fuzzy conditional random fields for temporal data mining

Temporal data mining is one of the interesting problems in computer science and its application has been performed in a wide variety of fields. The difference between the temporal data mining and data mining is the use of variable time. Therefore, the method used must be capable of processing variables of time. Compared with other methods, conditional random field has advantages in the processing variables of time. The method is a directed graph models that has been widely applied for segmenting and labelling sequence data that appears in various domains. In this study, we proposed use of Fuzzy Logic to be applied in Conditional Random Fields to overcome the problems of uncertainty. The experiment is compared Fuzzy Conditional Random Fields, Conditional Random Fields, and Hidden Markov Models. The result showed that accuracy of Fuzzy Conditional Random Fields is the best.

Big Data and Graph Theoretic Models: Simulating the Impact of Collateralization on a Financial System

In this paper we represent a financial system using a weighted directed graph model. We simulate and analyze the impact of financial regulations regarding the collateralization of derivative trades on systemic risk, employing a novel open source risk engine. The analysis finds that introducing collateralization does reduce the costs of resolving a financial system in crisis. It does not, however, change the distribution of risk in the system. The implications of the analysis highlight the importance of scenario based testing using hands on metrics to quantify the notion of system risk.

On strongly connected networks with excitable-refractory dynamics and delayed coupling.

We consider a directed graph model for the human brain’s neural architecture that is based on small scale, directed, strongly connected sub-graphs (SCGs) of neurons, that are connected together by a sparser mesoscopic network. We assume transmission delays within neuron-to-neuron stimulation, and that individual neurons have an excitable-refractory dynamic, with single firing ‘spikes’ occurring on a much faster time scale than that of the transmission delays. We demonstrate numerically that the SCGs typically have attractors that are equivalent to continual winding maps over relatively low-dimensional tori, thus representing a limit on the range of distinct behaviour. For a discrete formulation, we conduct a large-scale survey of SCGs of varying size, but with the same local structure. We demonstrate that there may be benefits (increased processing capacity and efficiency) in brains having evolved to have a larger number of small irreducible sub-graphs, rather than few, large irreducible sub-graphs. The network of SCGs could be thought of as an architecture that has evolved to create decisions in the light of partial or early incoming information. Hence the applicability of the proposed paradigm to underpinning human cognition.

Prioritization and Ranking of Phases for the Management of the Construction Project of the LNG Plant Production Line

Purpose: the main purpose of this article is to explore the feasibility of using the ISM-Interpretive Structural Modeling method to prioritize the phases in the management of the LNG plant construction project. To achieve this goal, the following tasks must be solved in the article: identify the phases of the project at the planning stage that affect the Final Investment Decision (FID); draw up a reachability matrix that defines the interrelations between the phases of the project; split the phases of the project into hierarchical levels; construct a directed graph model and a diagram of the degree of influence and coherence of the project phases facilitating the adoption of strategic management decisions and the coordination of internal and external stakeholders. Methods: this article is based on the concept of project management and complex systems, which can be investigated using the Interpretive Structural Modeling method, with reference to the specifics of LNG plants. Results: the phases of the project were identified and the relationships between them in the planning phase influencing the FID were established. As a result of modeling, the phases of the project were aligned in a hierarchical structure of the directed graph and positioned on the influence and connectivity diagram, which allows the Project Office of the company to correctly determine priorities in coordination of project participants. Conclusions and Relevance: the present studies show that the method of Interpretive Structural Modeling is effective in identifying and modeling complex relationships between the phases of the project, as a result of which it is possible to build a strategy for the adoption of the Final Investment Decision. This methodology can be extended to subsequent stages of the project. Project managers can prioritize and rank the phases of the project by levels to make effective management decisions to improve the coordination of internal and external stakeholders involved in the implementation of this project. The practical application of the results will improve the management processes between the Project Office and company departments and external stakeholders.

Life Cycle Characteristics of Warm-Season Severe Thunderstorms in Central United States from 2010 to 2014

Weather monitoring systems, such as Doppler radars, collect a high volume of measurements with fine spatial and temporal resolutions that provide opportunities to study many convective weather events. This study examines the spatial and temporal characteristics of severe thunderstorm life cycles in central United States mainly covering Kansas, Oklahoma, and northern Texas during the warm seasons from 2010 to 2014. Thunderstorms are identified using radar reflectivity and cloud-to-ground lightning data and are tracked using a directed graph model that can represent the whole life cycle of a thunderstorm. Thunderstorms were stored in a GIS database with a number of additional thunderstorm attributes. Spatial and temporal characteristics of the thunderstorms were analyzed, including the yearly total number of thunderstorms, their monthly distribution, durations, initiation time, termination time, movement speed and direction, and the spatial distributions of thunderstorm tracks, initiations, and terminations. Results revealed that thunderstorms were most frequent across the eastern part of the study area, especially at the borders between Kansas, Missouri, Oklahoma, and Arkansas. Finally, thunderstorm occurrence is linked to land cover, including a comparison of thunderstorms between urban and surrounding rural areas. Results demonstrated that thunderstorms would favor forests and urban areas. This study demonstrates that advanced GIS representations and analyses for spatiotemporal events provide effective research tools to meteorological studies.

Goodness of fit for log-linear network models: dynamic Markov bases using hypergraphs

Social networks and other sparse data sets pose significant challenges for statistical inference, since many standard statistical methods for testing model/data fit are not applicable in such settings. Algebraic statistics offers a theoretically justified approach to goodness-of-fit testing that relies on the theory of Markov bases. Most current practices require the computation of the entire basis, which is infeasible in many practical settings. We present a dynamic approach to explore the fiber of a model, which bypasses this issue, and is based on the combinatorics of hypergraphs arising from the toric algebra structure of log-linear models. We demonstrate the approach on the Holland–Leinhardt $$p_1$$ model for random directed graphs that allows for reciprocation effects.

Monocular Depth-Ordering Reasoning with Occlusion Edge Detection and Couple Layers Inference

A depth-ordering reasoning approach first provides novel occlusion edge detection, generating precise same-layer relationship judgment and producing reliable region proposals for the depth-ordering inference. Specifically, a novel sparsity-induced regression model learns a discriminative feature subspace. In addition, kernel ridge regression assigns the occlusion label for each edge. The kernel trick guarantees linearly separable edges in a rich, high-dimensional feature space. Secondly, a couple layers inference approach infers the final depth order. In the semilocal layer, a novel triple descriptor judges the foreground relationship. In the global layer, the inference is executed by finding a valid path on a directed graph model. The proposed approach is validated on the Cornell depth-order and NYU 2 datasets.

A model for generation of directed graphs of information by the directed graph of metainformation for a two-level model of information units with a complex structure

A model for generation of directed graphs of information by the directed graph of metainformation for a two-level connected directed graph model of information units is described. It can be used as a basis in computer tools for editing information with a complex structure with different levels of abstraction and invariance to technological spaces (subject domains) in a common conceptual framework for carriers of this information, making it unnecessary to organize specialized training for these carriers or to involve professional intermediaries.

Automatic HAZOP analysis method for unsteady operation in chemical based on qualitative simulation and inference

Comparing with continuous production process, unsteady operation process, such as startup and shutdown, tends to abnormal situations due to a large number of operations of operators and dynamic state changes involved. To guarantee a safe operation, process hazard analysis (PHA) is very important to proactively identify the potential safety problems. In the chemical process industry, hazard and operability (HAZOP) analysis is the most widely used method. In this paper, based on proposed qualitative simulation and inference method, an automatic HAZOP analysis method for unsteady operation processes is proposed. Mass transfer and relationships among process variables are expressed by Petri net–directed graph model based fuzzy logic. Operating procedure is expressed according to a formal expression. Possible operation deviations from normal operating procedure are identified by using a group of guidewords. Hazards are identified automatically by qualitative simulation and inference when wrong operation process is performed. The method is validated by a rectification column system.

Optimal Sensor Placement for Fault Diagnosis Using Magnitude Ratio

Major performance losses occur in process industries due to failures that are not identified at the incipient stage. Early detection of such faults is also critical for the safety of the equipment, operating personnel, and other resources. When a fault occurs in a system, it can propagate and affect several process variables. Variables that need to be measured in order to detect and diagnose the faults have to be identified and chosen economically. An algorithmic approach for identifying the optimal number, type, and location of the sensors for fault detection and diagnosis is useful, particularly for large-scale, chemical process plants. In this work, previous algorithms for sensor placement that use signed directed graph (SDG) models for the process are enhanced to include magnitude ratio (MR) information to identify more promising sensor locations. Further, we also study the combination of fault evolution sequences (FES) already introduced in the literature and the MR information for effective fault diagnosis. This is achieved by including the idea of artificial sensors that represent pairwise sensors from the original list of possible sensors. Based on the MR and FES, the artificial sensors can assume discrete values, much like the SDG approach. A symmetric difference operator is used on both the original sensors (whose behaviors are modeled as before using SDG) and the artificial sensors to identify sensor placements. This approach elegantly incorporates the new MR and FES information in the original well-accepted SDG based sensor placement algorithms. Several case studies are presented to demonstrate the usefulness of the proposed approach.

Saliency generation from complex scene via digraph and Bayesian inference

Making a close inspection of the recent progress made in computer vision we will find that more and more advancement can be attributed to the introduction of bio-inspired algorithms, which is a flourishing area of computing. Biological patterns can be generated from salient attention principle of the nervous system, which pops out the most important information from large backgrounds. New evidences of the underlying neuronal mechanisms indicate that biological motion phenomena happen on higher cognition level instead of low-level pixels in sensing environment, especially in a complex scene. Inspired by it, this paper proposes a novel saliency detection method based on a directed graph model and multi-scale Bayesian inference. We first create a directed graph with superpixels as its nodes and introduce a baseline node whose saliency is considered to be zero. The saliency of each node is defined as the shortest distance from the baseline node to it and Dijkstra׳s algorithm is adjusted to solve this optimization problem with great efficiency. Furthermore, we extend this model to a multi-scale version to cope with salient regions of different sizes and a Bayesian inference strategy via 3D color histogram is developed to achieve pixel-level saliency. Experimental results on some benchmark dataset demonstrate the superiority of our method with respect to 18 state-of-the-art saliency detection methods and our method achieves the highest recall in MSRA-1000. Additional experiments on ship detection of SAR images show that the proposed method can overcome the shortcomings of traditional CFAR detector and has much fewer false alarms in cluttered background.

Exact estimation of multiple directed acyclic graphs

This paper considers structure learning for multiple related directed acyclic graph (DAG) models. Building on recent developments in exact estimation of DAGs using integer linear programming (ILP), we present an ILP approach for joint estimation over multiple DAGs. Unlike previous work, we do not require that the vertices in each DAG share a common ordering. Furthermore, we allow for (potentially unknown) dependency structure between the DAGs. Results are presented on both simulated data and fMRI data obtained from multiple subjects.