Information Data Research Articles

Research of data generation options based on multi-scenario modelling for decision-making under uncertainty in the development of maritime passenger terminal

The topic of the paper is the study of decision-making options under uncertainty, as well as in practical situations of working with limited data in conditions of stochastic, variable influence of the external environment in the sphere of maritime passenger transportation. Attention is drawn to the fact that in this case, making a decision on modernization only on the basis of the experience of the port manager or some industry experience, on the one hand, is quite limited in the choice of alternatives, and on the other hand, may lead to incorrect decisions in the field of development forecasting, formation of measures to change the position of the marine passenger port in the sea region in relation to other terminals and evaluation of the target function of the port. It is noted that one of the tools for data collection is continuous monitoring of changes in route ferry and cruise networks and analyzing changes in the infrastructure of other passenger ports in the region. For completeness of data it is proposed to use multi-scenario modeling in the digital twin of the sea passenger port developed in AnyLogic environment, as well as to perform optimization experiments to obtain a set of values for the number of cruise and ferry vessels for the upcoming navigation. It is noted that the use of a simulation model makes it possible to include such an influence of the external environment as the trend of increasing size of cruise and ferry ships, which forms new infrastructure requirements for the creation of new or modernization of existing berths. In this case, the introduction of prioritization in the model of the incoming flow of ships is reasonably proposed. The object of the study is the incoming flow of cruise and ferry ships with prioritization in the queue. A comparative analysis of applicability of different mathematical distributions of mass service systems for incoming ship flows into the sea passenger port system with prioritization is presented. The developed simulation model is based on the available real data on the passenger port infrastructure of marine “Passenger Port of St. Petersburg ‘Sea Facade’ and ship calls for the past years. A multi-scenario random input flow of cruise and ferry ships with several priority characteristics is considered. As a result of modeling a set of data is formed, which for the port manager is a necessary field of utility and a basis for the use of decisionmaking models under uncertainty. As a result of multi-scenario simulation, the obtained throughput simulation data under different flow distributions were analyzed. Based on the performed experiments, the efficiency of using the beta distribution was determined. Based on the simulation results, the applicability boundaries for each distribution were determined and a confidence interval of maximum intensity was obtained, which can be used as a reference for making decisions on port infrastructure modernization. The methodology of data formation on the basis of multiscenario modeling allows to form the necessary set of data taking into account the influence of the external environment for decision-making on the model of marine passenger terminal development. The presented methodology can be extended to the study of systems: “sea ferry / cruise lines — sea passenger port — land infrastructure”.

ARMENIAN-CHINESE RELATIONS: HISTORICAL FOUNDATIONS OF DEVELOPMENT AND ARMENIANS IN CHINA / ՀԱՅ- ՉԻՆԱԿԱՆ ՀԱՐԱԲԵՐՈՒԹՅՈՒՆՆԵՐԻ ԶԱՐԳԱՑՄԱՆ ՊԱՏՄԱԿԱՆ ՀԻՄՔԵՐԸ ԵՎ ՉԻՆԱՍՏԱՆԻ ՀԱՅԵՐԸ

The multifaceted and diverse relations between Armenia and China have developed over a long historical period, and today they are manifested in new forms and actions, including the development of trade, economic, cultural, and diplomatic ties. In this article, the author attempts to present the reasons for the presence of Armenians in China, the formation of Armenian communities, their activities, and historical data on their existence and development, based on information and publications from various sources. The relevance of the topic lies in the fact that the approaches and opportunities for the development of Armenian-Chinese relations, discussed on various platforms, have been presented in the historical context, and an attempt has been made to identify the key directions and areas that remain relevant today.

Design of general LQR-ANN-controller of „crane-load” system. Part 1

The first part of the article describes the research concept and presents the results, which in the future allow the development of a LQR-neurocontroller of the movement of the dynamic "crane-load" system. For this purpose, the the problem of optimal control was stated. It uses a mathematical model in which the control function is considered as the rate of driving force change. This increases the order of the system by one. For individual components of the integral criterion, the weight coefficients were chosen and the values of the initial conditions were substantiated. The original problem of the synthesis of a LQR-controller is reduced to the Riccati equation. For one case, a solution to the Riccati equation was obtained and graphic dependencies corresponding to the obtained optimal control were built. The analysis of graphical dependencies made it possible to establish the disadvantages and advantages of the obtained optimal control. Among the advantages are the smoothness of the movement of the system and the provision of a zero value of the driving force at the beginning of the movement. This makes it possible to reduce the dynamic forces of the drive of the crane movement mechanism and its metal structure. Among the disadvantages of optimal control is a significant rate of increase of the driving force at the beginning of the movement, which can cause difficulties in the implementation of optimal control in practice, as well as an overshoot of the crane velocity. Multiple solutions of the Riccati equation made it possible to obtain datasets for training, validation, and testing of an artificial neural network, which is considered as a universal approximator of Riccati equation solutions. The process of data normalization and formation of training pairs is described. All data regarding the optimal values of the controller coefficients were obtained for load masses that varied within 60...25,000 kg and lengths of flexible suspension that varied within 1.2...12 m. In addition, the power of the weight coefficient varied within -5...-30, which corresponds to the minimization of the driving force rate.

Modification of Cellulose Isolated from Coconut Dregs Using Glycidyl Methacrylate (GMA)

The main objective of this work was to produce grafted cellulose from coconut dregs. In this work, cellulose isolated from coconut dregs was grafted with various concentrations of glycidyl methacrylate (GMA). Cellulose-GMA prepared using 20% of ​​GMA produced the highest grafting percentage, where the grafting percentage reached 299.2%. The obtained grafted cellulose was characterized by FT-IR, XRD, and TGA. FT-IR spectra confirmed the formation of cellulose-GMA and XRD data showed a slight decrease in the crystallinity of cellulose after GMA addition from 6.98% to 6.02%, 5.57%, and 5.83% for cellulose-GMA prepared using 15%, 20%, and 25% of GMA, respectively. Cellulose-GMA showed higher thermal stability than cellulose, that potentially used in some applications at higher temperatures.

СИСТЕМА ОПАЦЮВАННЯ МНОЖИНИ СЕНСОРІВ ЗА ДОПОМОГОЮ МІКРОКОНТРОЛЕРІВ ТИПУ RASPBERRY PI PICO

There were investigated alternative methods for receiving and analyzing multiple signals from different sources, and conducted a literature review in relevant areas. In particular, we analyzed an expansion board of the ReSpeaker 6-Mic Circular Array Kit type based on a Raspberry Pi minicomputer, which is equipped with six microphones and is intended for applications in the field of artificial intelligence and voice interfaces. It is shown that the ReSpeaker board has limited application, it is intended for stationary use with limited scalability. In this work, the authors created and tested an experimental electronic system for processing signals from twelve sensors using four Raspberry Pi Pico microcontrollers, as well as software compiled using the micropython language, which allows us to significantly expand the scope of its application. Condenser microphones complete with amplifiers of the KY-037 module type were used as highly sensitive sensors. The conversion of an analog signal into a digital one and its primary processing is carried out using Pico microcontrollers for the implementation of high-precision embedded systems. To ensure data exchange between microcontrollers and the central node and peripheral devices, a one-way I2C serial bus was used, which provides synchronous, multi-controller network communication. Its use is associated with setting a short distance between the boards, on the order of tens of centimeters. The main functions of microcontrollers are analog-to-digital conversion of signals at the output of amplifiers, calculation of the average intensity of signals from each microphone, comparison of averaged signals to detect the maximum intensity, formation and transmission of data to a buffer or central node. The effectiveness of the experimental system was checked using compiled software tools that provide real-time processing of sound signals. Thanks to the successfully selected hardware and software components, a prototype of the system was created, capable of operating effectively in field conditions.

Computer Modelling of Logistic Regression for Binary Classification

This article discusses the practical aspects of applying logistic regression for binary data classification. Logistic regression determines the probability of an object belonging to one of two classes. This probability is calculated with the help of a sigmoid function, the argument of which is a linear convolution of the feature vector of the object with the weighting coefficients obtained during the minimization of the logarithmic loss function. Predicted class labels are determined by comparing the calculated probability with a given threshold value. The logistic regression study was performed using the computer simulation method. For this, a software complex was developed, the work of which reproduces the main stages of logistic regression: preparation of input data, training, testing with determination of quality metrics of binary classification, application of the logistic regression method for data classification in practice. The paper examines the effect of overlapping and imbalance of classes in the input data set on the efficiency of binary classification. The overlapping of classes is modeled by the formation of input data based on two shifted relative to each other density functions of the normal distribution of random variables. Class imbalance is simulated by the probability of switching between these features. It is shown that when the distance between the mathematical expectations of the density functions of the normal distribution decreases or when the dispersion of random variables increases, the overlapping of relevant classes increases, which leads to an increase in the number of objects that the classifier can assign to one or another class. Approaching the probability of switching between the distribution functions of random variables to the extreme values of the unit interval leads to an increase in class imbalance, which is manifested in an increase in the number of elements of the input data set labeled with the label of the same class. It has been experimentally confirmed that the AUC ROC metric, popular in binary classification problems, is dependent on the degree of class overlap and relatively resistant to class imbalance.

Dynamic Process Fault Detection and Diagnosis Method Based on Factor Analysis: Application on the Three‐Tank System Process

ABSTRACTTo address the issue of underreporting faults in the detection of tiny faults by dynamic factor analysis (DFA), a novel fault detection and diagnosis method based on DFA‐sliding window combined with mean square error (DFA‐SWMSE) is proposed. Firstly, the data matrix is augmented by introducing time lag shifts. Secondly, factor analysis (FA) is applied to the augmented data matrix, achieving dimensionality reduction and feature extraction while retaining most of the original data's information. Then, the sliding window technique is applied to calculate the mean square error of the dimensionally reduced data, allowing for the monitoring of the system's current state and the detection of tiny faults. Finally, effective fault diagnosis is achieved through the analysis of fault factors and variable contributions. The proposed method is validated using a complex dynamic numerical example and a three‐tank system process named Sim3Tanks. This system has gained widespread application in the field of process fault detection due to its ability to simulate and generate various types of faults. The proposed method is compared with principal component analysis (PCA), dynamic principal component analysis (DPCA), PCA similarity factor (SPCA), FA, and DFA. The experimental results thoroughly validate the effectiveness of the proposed method in detecting and diagnosing tiny faults in dynamic processes.

Aspects of Developing an Innovative, Energy-Efficient, Low-Emission Co-Generator

Purpose. To develop a model and construct an experimental, innovative co-generator with high energy efficiency and low emissions. This involves developing a system that can efficiently generate both electricity and heat simultaneously while minimizing emissions, contributing to sustainability efforts, and addressing energy demands in a more environmentally friendly manner. Methodology. To achieve the goal, a system approach is utilized, enabling the selection of modelling types for the development of an experimental, cutting-edge co-generation system capable of efficiently producing both electricity and heat with superior energy efficiency and minimal emissions. For this purpose, the following steps were completed: processing and summarizing available literature and patent sources, analysing scientific and technical papers on the selection and application of modelling types in co-generation systems, and considering principles and individual approaches to input data formation for mathematical modelling. This process enables the selection of a mechanism and the creation of simulation models for effective energy production at various enterprises. Findings. Assessment is performed of the energy efficiency of the co-generator system under various operating conditions, comparing it with existing conventional methods of electricity and heat generation. Results are presented of performance testing to determine the system’s capability to simultaneously generate electricity and heat efficiently, considering factors such as output stability, load responsiveness, and overall reliability. Identification and evaluation are performed of innovative technologies and methodologies integrated into the co-generator design, highlighting their effectiveness in enhancing energy efficiency and reducing emissions. Insights into any operational challenges encountered during the construction, testing, and optimization phases, along with proposed solutions or improvements to solve these problems. The analysis of the overall environmental impact of deploying the co-generator showed potential benefits in terms of reduced greenhouse gas emissions and local air quality improvement. Originality. Using a combination of scientific approaches encompassing physics and heat transfer engineering, in accordance with the first law of thermodynamics, such as the conservation laws of energy and entropy, and principles of heat exchange employed to transfer heat between different mediums, gas kinetics have yielded values for the energy transformation coefficient, indicating qualitative characteristics of fuel thermolysis and power generation as the final product. Practical value. The results provide for developing a comprehensive model of an advanced co-generation system capable of efficiently producing both electricity and heat with superior energy efficiency and minimal emissions. It also entails determining the types of models for mathematical modelling at all management levels and establishing a new method for input data formation for both technologies and their subsystems, incorporating additional technological implementations.

Selection of aerodynamic characteristics and engine parameters of a maneuverable aircraft under epistemic uncertainty

At the preliminary stage of aircraft design, it is usually required to solve the problem of insufficient initial data for applying traditional algoristic-type mathematical programming models. In many cases, numerous input parameters cannot be accurately specified at the time of making design decisions. If the inaccuracy of parameters is not taken into account, the actual values of target functions may differ significantly from the calculated ones when solving optimization problems. In this regard, the issue of current interest is the development of algorithms to improve the reliability of design decisions under conditions of epistemological uncertainty when experts engage in the formation of initial data. The paper considers the problem of selecting aerodynamic characteristics and engine parameters of maneuverable aircraft under conditions of uncertainty associated with inaccuracy of expert data. The applied problem under consideration is further complicated by the necessity to include “black box” models in the algorithms being developed. The paper proposes algorithms that apply the uncertainty theory together with “black box” models that implement optimization calculation techniques derived from previous engineering practice of aircraft design. Using these algorithms, experts are able to set uncertain parameters whereby the lack of data is factored in using uncertainty distribution functions. In cases of monotonicity of uncertain parameters in target functions, application of the uncertainty theory provides a significant reduction in computational costs compared to the method of statistical modeling in optimization calculations. The paper presents the results of computational studies of the developed algorithms. A genetic algorithm (mathematical optimization solver) is used to optimize the search for design solutions. Pareto frontiers are obtained for different confidence levels enabling to make design decisions including values of aerodynamic characteristics and engine parameters of maneuverable aircraft.

Zonal shape reconstruction for Shack-Hartmann sensors and deflectometry

Zonal shape reconstruction for Shack-Hartmann sensors and deflectometry

АВТОМАТИЗАЦІЯ ЗБОРУ ТА АНАЛІЗУ ЗВІТНОСТІ ШКІЛ ІЗ ВИКОРИСТАННЯ КОШТІВ ДЕРЖАВНИХ ОСВІТНІХ СУБВЕНЦІЙ

The availability of reliable educational statistical, administrative, and financial information is a prerequisite for the development of effective educational policy. The importance of the reliability of educational management information has only increased since Russian military invasion of Ukraine in 2022. In these circumstances, a detailed analysis of the use of state educational subventions as the main budgetary instrument for financing education is needed. Increased transparency and automation of information gathering can be achieved by introducing an information system for collecting and analyzing school reports on the use of state educational subventions. Currently, such generalized information is collected by the State Treasury Service of Ukraine (STSU) starting from the level of the territorial community, but it is not available to educational managers. The purpose of this study is to formulate a methodology for developing information and analytical tools for automated collection and analysis of financial statements of general secondary education institutions using state educational subventions in order to increase transparency and efficiency of budget funds, create preconditions for the formation of reliable data, on the basis of which it will be possible to make informed financial and organisational decisions, evaluate their effectiveness at all levels of educational management. The article analyzes the existing reporting forms submitted by the general secondary education institutions (GSEI), and the regulatory framework for their submission. Improving the reliability and quality of the collected financial data can be achieved by developing and implementing an information subsystem (module) for automated gathering and analysis of this information from the level of GSEIs in the environment of the software and hardware complex "Automated Information Complex of Educational Management", which is already used by education managers. The requirements for such a module and its interaction with the already available information in the SHC AICOM and other state information systems are substantiated. It is shown that the optimal way is to organize automated collection of financial information from the level of general secondary schools. The implementation of the module will improve the information support of educational and financial managers, and will allow for a comprehensive solution to the issues of local education development.

The possibilities of the Teams information system in creating a digital footprint of first-year students

The article examines in detail the problem of assessing the involvement of first-year students in the educational process and the degree of assimilation of knowledge using digital technologies that make it possible to obtain a student’s digital footprint. The purpose of the article is to present the technology for obtaining digital footprint data when teaching first-year students at Yaroslavl State Technical University in the discipline “Information Systems Architecture” in the MS Teams environment and to show options for performing digital footprint analysis. Methodology. The formation of digital footprint data was carried out in the process of lecture surveys, work on laboratory assignments and passing an exam. To obtain a digital footprint, information systems MS Teams, MS Excel, MS Forms were used. Data obtained in the MS Teams environment was subjected to pre-processing, cleaning, merging tables, and adding (deleting) attributes. The generated tables were analyzed using classification and cluster analysis to search for non-trivial knowledge. Cluster analysis allows us to identify groups of students who successfully complete lecture and exam testing and those who have difficulties completing test tasks. The analysis allows us to identify the level of complexity of tasks and allows for their adjustment. A comparison of the success of completing tasks during the lecture course and in the exam gives an idea of the students’ ability to prepare for participation in testing and the ability to memorize large volumes of complex scientific and technical information.

OPTIMIZATION OF MANAGEMENT DECISIONS IN TAX RELATIONS: NEW TECHNOLOGIES AND STANDARDS IN ACCOUNTING PROCESSES

The article addresses the critical role of accounting information in the context of tax relations, particularly in the dynamic business environment influenced by globalization. It emphasizes the necessity for structured data organization and information flow to enhance managerial decision-making processes. A significant focus is placed on the quality of accounting data, which is foundational for effective tax reporting and compliance. The study reviews existing literature, highlighting the contributions of Ukrainian researchers who stress the importance of relevant and high-quality accounting information for decision-making. The paper aims to explore the evolving approaches to accounting data formation, driven by globalization and the challenges it presents in tax interactions. It discusses the implications of adopting International Financial Reporting Standards (IFRS), which have been developed to ensure transparency, comparability, and high-quality financial reporting across borders. The article notes that while IFRS adoption is mandatory for many entities in Europe, discrepancies in implementation exist among member states, leading to varied disclosure requirements. Furthermore, the introduction of the Standard Audit File for Tax (SAF-T) is examined as a modern solution for enhancing data exchange between businesses and tax authorities. This XML format enables tax officials to access comprehensive accounting information, facilitating tax audits and compliance checks. The integration of SAF-T with IFRS reporting is posited to improve data quality and operational efficiency while addressing the increasing demand for reliable accounting information. In conclusion, the article argues that the implementation of IFRS and SAF-T can significantly enhance financial transparency and comparability, thereby reducing risks and costs associated with investment evaluations in the global market. However, it also acknowledges potential challenges in managing accounting practices and the need for advanced measures to accurately reflect a company's financial status and performance. This comprehensive analysis underscores the importance of high-quality accounting data in navigating the complexities of modern tax relations

The Effect of CEO Narcissism, Company Size, and Free Cash Flow on Financial Performance through CSR in SOEs on the IDX 2018-2022

This study aimed to determine the effect of CEO Narcissism, Company Size, and Free Cash Flow through Corporate Social Responsibility (CSR) on Financial Performance in State-owned enterprises Companies. This study uses a quantitative approach with data sources from the Indonesia Stock Exchange (IDX) official website using the Warppls 7.0 application. The research sample data were from 16 state-owned enterprise companies. The test results show that CEO Narcissism has a negative and significant effect on Corporate Social Responsibility (CSR), Company Size has a negative and insignificant effect on Corporate Social Responsibility (CSR), Free Cash Flow has a positive and insignificant effect on Corporate Social Responsibility (CSR), CEO Narcissism is unable to mediate the effect of Corporate Social Responsibility (CSR) on financial performance. Company Size cannot mediate the effect of Corporate Social Responsibility (CSR) on financial performance. Free Cash Flow cannot mediate the influence of Corporate Social Responsibility (CSR) on financial performance. The Total Determination Coefficient in this study is 0.285 or 28.5%. This indicates the model can explain 28.5% of the data's information. Other factors not included in the study model account for the remainder.

A Review of Blockchain-Based Research on E-Health Data Sharing

With the development of medical data informatization, the data security, data privacy, and data controllability of electronic medical data with patient's personal privacy have gained social attention. Blockchain technology has decentralized characteristics, and the development of blockchain technology provides an effective solution for the secure sharing of e-medical data. Many scholars have proposed e-medical data security sharing solutions based on blockchain technology, and these solutions have attempted to solve the security problems encountered in sharing traditional e-medical data. By summarizing and analyzing the schemes in this field, this study analyzes the advantages, disadvantages, and challenges that the current mainstream schemes have, mainly in terms of blockchain data security storage, access control security sharing, and so on, in order to promote the further development of the e-medical data security sharing scheme based on blockchain technology. This study analyzes the blockchain-based e-medical data security sharing scheme proposed by scholars by analyzing the blockchain technology, cryptography technology, smart contract technology, P2P network technology and other related topics. Through systematic analysis and comparison, it explores the challenges of the current research and provides references for the future proposals of blockchain technology-based schemes.

Methodology for Collecting Data on the Activity of Malware for Windows OS Based on MITRE ATT&CK

The digitalization of the modern economy has led to the emergence of information technologies in various areas of human activity. In addition to positive effects, this has enhanced the problem of countering cyber threats. The implementation of cyber threats often impacts serious consequences, especially when it comes to critical information infrastructure. Malware is an important part of the modern landscape of cyber threats; the most high-profile cybercrimes of recent years are associated with the use of malware. In this regard, the problem area of countering malware is actively developing, and one of the promising areas of research in this area is the creation of methods for detecting malware based on machine learning. However, the weak point of many well-known studies is the construction of reliable data sets for machine learning models, when the authors do not disclose the features of the formation, preprocessing and labeling of data on malware. This fact compromises the reproducibility a lot of studies. This paper proposes a methodology for collecting data on malware activity based on the MITRE ATT&CK matrix and Sigma rules and designed for Windows OS. The proposed methodology is aimed at improving the quality of datasets containing malware and legitimate processes behavior’s features, as well as at reducing the time of data label by an expert method. A software stand was prepared and experiments were carried out for testing the methodology. The results of experiments confirmed applicability of our methodology.

Transdisciplinary fine-grained citation content analysis: A multi-task learning perspective for citation aspect and sentiment classification

Transdisciplinary fine-grained citation content analysis: A multi-task learning perspective for citation aspect and sentiment classification

Neural Technologies for Objects Classification with Mobile Applications

This paper is related to the study of the features of the neural technologies’ application, in particular, ResNet neural networks for the classification of objects in photographs. The work aims to increase the accuracy of recognition and classification of objects in photographs by using various models of the ResNet neural network. The paper analyzes the features of the application of the corresponding models in comparison with other architectures of deep neural networks and evaluates their efficiency and accuracy in the classification of objects in photographs. The process of data formation for training neural networks, their processing and sorting is described. A web application and a mobile application for recognizing and classifying objects in a photo were also developed. A system for classifying objects, in particular airplanes in photographs, was developed using neural network technologies. It gives a recognition and classification accuracy of about 95%. Research results of ResNet models are of great practical importance, as they can improve the classification accuracy of various images. Features of ResNet, such as the use of skip connections or residual connections, make it effective in the relevant tasks. The results of the study will help to implement ResNet in various fields, including medicine, automatic pattern recognition and other areas where the classification of objects in photographs is an important task.

Bibliometric analysis of research challenges and trends in urban underground space

AbstractThe utilization and development of urban underground space play a crucial role in optimizing the layout of civic architecture and enhancing the urban ecological environment, which contributes toward increasing the overall carrying capacity and promoting sustainable development in megacities. To delve into the research progress of urban underground space, knowledge maps were created using the information visualization software VOSviewer. The literature was systematically extracted from three prominent databases, namely, Web of Science, Scopus, and China National Knowledge Infrastructure. According to the bibliometric analysis of the co‐citation and core words co‐occurrence, the trends and challenges of research on urban underground space were identified. As highlighted by the results obtained, it still remains highly challenging to achieve interdisciplinary collaboration in urban underground space research; the research trends of urban underground space consist of collaborative planning and whole life cycle sustainable development, multisource geological data informatization and resource evaluation, infrastructure design optimization, and intelligent construction. The knowledge map, drawn using bibliometric methods, offers a quantitative analysis of literature retrieval across various levels. It is recognized as an essential tool for exploring and identifying challenges and trends in urban underground space.

Disentangled Partial Label Learning

Partial label learning (PLL) induces a multi-class classifier from training examples each associated with a set of candidate labels, among which only one is valid. The formation of real-world data typically arises from heterogeneous entanglement of series latent explanatory factors, which are considered intrinsic properties for discriminating between different patterns. Though learning disentangled representation is expected to facilitate label disambiguation for partial-label (PL) examples, few existing works were dedicated to addressing this issue. In this paper, we make the first attempt towards disentangled PLL and propose a novel approach named TERIAL, which makes predictions according to derived disentangled representation of instances and label embeddings. The TERIAL approach formulates the PL examples as an undirected bipartite graph where instances are only connected with their candidate labels, and employs a tailored neighborhood routing mechanism to yield disentangled representation of nodes in the graph. Specifically, the proposed routing mechanism progressively infers the explanatory factors that contribute to the edge between adjacent nodes and augments the representation of the central node with factor-aware embedding information propagated from specific neighbors simultaneously via iteratively analyzing the promising subspace clusters formed by the node and its neighbors. The estimated labeling confidence matrix is also introduced to accommodate unreliable links owing to the inherent ambiguity of PLL. Moreover, we theoretically prove that the neighborhood routing mechanism will converge to the point estimate that maximizes the marginal likelihood of observed PL training examples. Comprehensive experiments over various datasets demonstrate that our approach outperforms the state-of-the-art counterparts.