Proximity Matrix Research Articles

Modern Hopfield Networks for graph embedding.

The network embedding task is to represent a node in a network as a low-dimensional vector while incorporating the topological and structural information. Most existing approaches solve this problem by factorizing a proximity matrix, either directly or implicitly. In this work, we introduce a network embedding method from a new perspective, which leverages Modern Hopfield Networks (MHN) for associative learning. Our network learns associations between the content of each node and that node's neighbors. These associations serve as memories in the MHN. The recurrent dynamics of the network make it possible to recover the masked node, given that node's neighbors. Our proposed method is evaluated on different benchmark datasets for downstream tasks such as node classification, link prediction, and graph coarsening. The results show competitive performance compared to the common matrix factorization techniques and deep learning based methods.

Proximity matrix indicates heterogeneity in the ability to face child malnutrition and pandemics in Brazil: An ecological study.

Among the social inequalities that continue to still surpasses the basic rights of several citizens, political and environmental organizations decisively "drag" the "ghost" of hunger between different countries of the world, including Brazil. The COVID-19 pandemic has increased the difficulties encountered in fighting poverty, which has led Brazil to a worrying situation regarding its fragility in the fight against new pandemics. The present study aims to estimate, compare, and report the prevalence of mortality due to child malnutrition among the macro-regions of Brazil and verify possible associations with the outcome of death by COVID-19. This would identify the most fragile macro-regions in the country with the greatest need for care and investments. The prevalence of mortality was determined using data from the federal government database (DataSus). Child malnutrition was evaluated for the period from 1996 to 2017 and COVID-19 was evaluated from February to December 2020. The (dis)similarity between deaths from malnutrition and COVID-19 was evaluated by proximity matrix. The North and Northeast regions have above average number of deaths than expected for Brazil (p < 0.05). A prospective analysis reveals that the distribution of the North and Northeast macro-regions exceeds the upper limit of the CI in Brazil for up to the year 2024 (p < 0.05). The proximity matrix demonstrated the close relationship between deaths from COVID-19 and malnutrition for the Northern region followed by the Northeast region. There are discrepancies in frequencies between macro-regions. Prospective data indicate serious problems for the North and Northeast regions for the coming years. Therefore, strategies to contain the outcome of health hazards must be intensified in the macro-regions North and Northeast of the country.

Advanced setup for safe breath sampling and patient monitoring under highly infectious conditions in the clinical environment

Being the proximal matrix, breath offers immediate metabolic outlook of respiratory infections. However, high viral load in exhalations imposes higher transmission risk that needs improved methods for safe and repeatable analysis. Here, we have advanced the state-of-the-art methods for real-time and offline mass-spectrometry based analysis of exhaled volatile organic compounds (VOCs) under SARS-CoV-2 and/or similar respiratory conditions. To reduce infection risk, the general experimental setups for direct and offline breath sampling are modified. Certain mainstream and side-stream viral filters are examined for direct and lab-based applications. Confounders/contributions from filters and optimum operational conditions are assessed. We observed immediate effects of infection safety mandates on breath biomarker profiles. Main-stream filters induced physiological and analytical effects. Side-stream filters caused only systematic analytical effects. Observed substance specific effects partly depended on compound’s origin and properties, sampling flow and respiratory rate. For offline samples, storage time, -conditions and -temperature were crucial. Our methods provided repeatable conditions for point-of-care and lab-based breath analysis with low risk of disease transmission. Besides breath VOCs profiling in spontaneously breathing subjects at the screening scenario of COVID-19/similar test centres, our methods and protocols are applicable for moderately/severely ill (even mechanically-ventilated) and highly contagious patients at the intensive care.

Graph Distances for Determining Entities Relationships: A Topological Approach to Fraud Detection

A new model for the control of financial processes based on metric graphs is presented. Our motivation has its roots in the current interest in finding effective algorithms to detect and classify relations among elements of a social network. For example, the analysis of a set of companies working for a given public administration or other figures in which automatic fraud detection systems are needed. Given a set [Formula: see text] and a proximity function [Formula: see text], we define a new metric for [Formula: see text] by considering a path distance in [Formula: see text] that is considered as a graph. We analyze the properties of such a distance, and several procedures for defining the initial proximity matrix [Formula: see text]. Using this formalism, we state our main idea regarding fraud detection: financial fraud can be detected because it produces a meaningful local change of density in the metric space defined in this way.

Synchronous distributed ADMM for consensus convex optimization problems with self-loops

In this paper, a distributed algorithm based on the alternating direction method of multipliers is proposed to solve an unconstrained consensus optimization problem. The network topology among the agents is assumed to be connected and self-loops of each agent are considered. This algorithm takes less time than sequential algorithms because the optimization variables of the agents can be updated synchronously. In addition, it is shown that the algorithm is actually a Jacobi-proximal ADMM algorithm and the requirement of the proximal matrix is relaxed. A convergence analysis of the algorithm is provided, and its convergence rate is found to be O(1/k). Finally, the characteristics of the proposed algorithm are illustrated by numerical experiments, and comparisons are made with other ADMM algorithms in solving a least-squares problem.

GHOC: A generative model for hybrid‐order community detection

Recently, hybrid-order community detection has been proposed for addressing the hypergraph fragmentation issue suffered by the motif-based higher-order community detection. However, the existing attempts of hybrid-order community detection inadvertently damage the lower-order connectivity pattern and the higher-order connectivity pattern when constructing the fusion model. Additionally, like the higher-order community detection approaches, they also adopt a two-phase strategy that separately applies the existing graph node clustering methods to the proximity matrix derived from the lower-order connectivity pattern and the higher-order connectivity pattern. Therefore, the higher-order connectivity pattern is only utilized for constructing proximity matrix and hence has no direct effect on the final community results. In this paper, to address the above issues, we propose a Generative model for Hybrid-Order Community detection (GHOC). The main idea lies in defining a likelihood function of a generative model that finds the optimal community membership strength vectors of nodes, based on which the original lower-order connectivity pattern and the higher-order connectivity pattern can be directly reconstructed simultaneously. From the community membership strength vectors, the final community structure can be derived. Extensive experiments have been conducted on several data sets, and the results have confirmed the superiority of the proposed GHOC method.

Exploring clusters of rare events using unsupervised random forests

Given highly imbalanced data, most learning algorithms face the challenge of accurately predicting rare events, while such cases are the ones that carry importance and useful knowledge. In a binary class label dataset, the rare events are the ones in the minority class. This study used a stroke dataset with a binary class label and the class imbalance ratio was 54:1. In addition to that, the dataset contained missing values and mixed data types. To identify the intrinsic structures in the minority class (the stroke group), Random Forest Clustering was used to produce the proximity matrix and fed to Partition around Medoid (PAM) clustering method to identify the optimal number of clusters. The proximity plot seems to show there could be cluster tendency as Hopkins’s statistics test value was H = 0.8735 and k=2 was identified to be the optimal number of clusters. Based on the internal cluster validation, however, the silhouette coefficient width was small (0.1), indicating that many of the data objects were within the other boundary of the other class. We have suggested a further investigation plan in this paper for the next action.

Bridging the gap between soil spectroscopy and traditional laboratory: Insights for routine implementation

We need innovative solutions to help laboratories to rapidly characterize the soils and adopt internal quality controls with lower costs and ecosystem impacts. Soil spectroscopy emerged as an alternative to wet chemistry. However, soil laboratories still do not widely use this technology in their routines, mainly due to the lack of: i) standards and protocols, ii) spectral libraries, iii) capacity in spectral methods, and iv) professionals with expertise in chemometrics. Therefore, we aimed to provide basic guidelines for the use of soil spectra in laboratory routines regarding internal quality control and samples selection for analysis and prediction. We used 350–2500 nm spectra and unsupervised random forests to compute proximity matrices and cluster spectrally similar soil samples for outlier detection using an adjusted method for skewed distribution, which allowed us to establish an internal quality control. Then, we used the most important wavelengths from unsupervised random forest to order soil samples by laboratory and subset them into different set sizes for training and testing prediction models for clay, sand, organic matter cation exchange capacity and base saturation. Our results indicated that internal quality control based on soil spectra and unsupervised analysis can be implemented for laboratory routines. Spectra-based soil samples selection and attributes predictions can reduce the need for laboratory analysis by half. Soil spectroscopy can become an important alternative to improve traditional analytical results, save time, reduce costs, and mitigate environmental impact with acceptable accuracy. Soil data and full codes from this study are publicly available at https://github.com/raulpoppiel/urssa.

A digital myxoid pseudocyst with novel dermoscopic signs

Sir, A 61-year-old female presented to us with a slowly progressive, asymptomatic, localized swelling in the proximal nail fold (PNF) of the left ring finger present for nine months. An examination revealed a discrete, well marginated, skin-colored, firm, non-tender nodule, 8 × 8 mm in size, between the distal interphalangeal (DIP) joint and the PNF of the left ring finger. Besides this, the concomitant nail plate also had longitudinal guttering originating from underneath the PNF and extending until the free edge of the nail plate, corresponding to the possibility of pressure effects of the lesion on the nail matrix (Fig. 1a). Polarized light dermoscopy revealed thick, white, linear structures intermingled with dotted vessels and short, linear, curved, comma-shaped vessels in the center of the lesion (Fig. 2a). High-resolution ultrasonography with power Doppler of the lesion revealed a hypoechoic area without any vascular component (Fig. 2b) and normal appearing interphalangeal joints. Under aseptic precautions, triamcinolone acetonide (TA) at 10 mg/mL was injected intralesionally from its lateral margin until the blanching of the lesion. A transparent, jelly-like material extruded through the entrance of the needle, thus confirming the clinical diagnosis of a digital myxoid pseudocyst (DMP) (Fig. 1b). No recurrence was observed within six months of follow-up (Fig. 1c). A DMP, also called a digital synovial cyst or ganglion cyst, is a benign tumor of the digit. It is not a true cyst as it lacks a cyst wall lined with epithelial cells. It is filled with mucoid materials overproduced by fibroblasts [1]. They arise due to repetitive trauma causing either focal mucinous degeneration of the connective tissue or the leakage of synovial fluid from the DIP joint. Two types of digital mucous cysts are distinguished: the myxomatous, or superficial, type and the ganglion, or deep, type. A DMP may sometimes compress the nail matrix, creating longitudinal depressions in the nail plate, as seen in this case. Nail changes may precede cyst formation by up to six months. Occasionally, the cyst may exert pressure on the nail matrix giving rise to a bluish-red lunula [2]. Salerni et al. described the dermoscopic features of a DMP as bright white areas with a linear, branched, serpentine vascular pattern [3]. Dermoscopy of our case, however, revealed white, linear structures intermingled with dotted vessels and short, linear, curved, comma-shaped vessels. These non-identical dermoscopic patterns could be attributed to the different depth of the DMP in our case to these observed by Salerni et al. The pseudocyst in our case also demonstrated compression effects on the proximal nail matrix indicated by longitudinal guttering in the nail plate, while no such finding was present in the former. In addition, arboriform and polymorphic vascular patterns have also been described. Repetitive puncturing, aspiration, sclerotherapy, and intralesional steroids have been attempted; however, recurrences may develop [4]. Although surgical excision of a DMP with synovectomy shows a lower recurrence rate, it may cause complications such as nail deformities, post-operative pain, and limited post-operative functional use. To conclude, the dermoscopic features of the DMP in the current case may hold an addition to the existing dermoscopic signs, hence aid in the diagnosis of the DMP.

Be Nice to Thy Neighbors: Spatial Impact of Foreign Direct Investment on Poverty in Africa

This study examines the spatial impact of FDI on the poverty of 44 African countries. In achieving this, the study uses the Driscoll–Kraay fixed effect instrumental variable regression, the instrumental variable generalized method of moments estimator (IV-GMM), and the spatial Durbin model. The empirical investigation of this study yielded four significant findings: (1) neighboring countries’ FDIs have a positive and significant impact on the incidence and intensity of the host country’s poverty, (2) improved institutional quality in neighboring countries has a significant impact on the FDI–poverty reduction nexus of the host country, (3) the empirical results lend support for a significant spatial spillover of poverty in the region, (4) the marginal effect results indicate that countries within the region are no longer in isolation or independent, i.e., the level of poverty in a particular country is influenced by its determinants in the neighboring country. This result is robust to the alternative proximity matrix, which is the inverse distance. Since there is spatial interdependence among African countries, we recommend that African governments, through the African Union (AU), should not only champion the institutional reform in the region, but also establish a binding mechanism to ensure reform implementation.

Chemotherapy casted shadow: Mees' and Beau's lines.

A 28-year-old female received six cycles of rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone chemotherapy to treat diffuse large B-cell lymphoma. She completed her treatment (see Figure 1) 3 weeks before this photograph. Examination showed multiple white lines on the fingernails, known as Mees’ lines. Mees’ lines are transverse, nonblanching white bands that run parallel to the lunula around the entire nail bed. The lines matched the start of each cycle, with the distance between the lines representing the time interval between cycles. In this patient, four lines can be identified with consistent intervals on the body of her nails, while the fifth and sixth lines are seen at a more significant distance at the nail-free edge. This might be related to treatment delay after cycle 2 (40 days) due to intensive care unit admission with sepsis and coronavirus disease (COVID-19) pneumonia. In the third and fourth cycles (the outer lines on the nail bed), we can see roughness and indentation (on the left-hand thumb and index finger), representing Beau's lines. Beau's lines are transverse depressions in the nail plate caused by transient cell division interruption in the proximal nail matrix. Both lines are commonly seen in patients with a history of chemotherapy exposure. The authors declare they have no conflicts of interest. The photograph was taken with the patient's informed consent and in accordance with the declaration of Helsinki The authors received no specific funding for this work. Informed consent was obtained. The authors declare that the data supporting the findings of this report are available within the article.

DGCU: A new deep directed method based on Gaussian embedding for clustering uncertain graphs

Clustering uncertain graphs, with the aim of discovering the most reliable partitions, has many applications in real-world networks. In literature, uncertainty is applied somehow to the traditional graph clustering algorithms, so that probabilistic extensions of them are achieved. Due to the difficulty of this technique per algorithm, besides the increasing noise and sparsity followed by more graph dimensions, we propose a deep learning based method to overcome these challenges. As a preprocessing, we first generate a Probabilistic Proximity Matrix (PPM) and a Probabilistic Similarity Matrix (PSM) using the topological structures of the uncertain graph. Afterward, we benefit from Gaussian embedding to capture uncertainty beyond the efficient representation of nodes in a low-dimensional space. PPM is used to embed nodes in a Gaussian space, and PSM is utilized for embedding directed clustering in which a mutually objective function is employed to learn node embedding and clustering simultaneously. Finally, the representations can be partitioned by any deterministic graph clustering algorithm. Various experiments were accomplished on four real-world datasets. The results show more effectiveness of the proposed method than the state-of-the-art approaches in detecting co-cluster objects and well-separated clusters, also indicating high reliability with regard to the pairwise connection probabilities within the clusters.

Исследование применимости PERT метода к процессу управления проектами дизайн-центра микроэлектроники

This paper presents a study of the applicability of the PERT method to the project management process of the microelectronics design center in comparison with the optimization mathematical model, which formed the basis of the software developed earlier. The study was based on corporate and cluster analyses of the projected results of project implementation by microelectronics design centers. The predicted values were obtained using a network model based on the PERT method and a mathematical model. To study the degree of similarity of the results of the network and mathematical models, we used a visualized proximity matrix constructed by the Ward method based on the criteria of the square of the Euclidean distance. The results are graphically represented by a similarity dendrogram. The paper illustrates the obtained dependencies of the implementation time on the totality of the number of simultaneously executed projects, the number of works and performers. The conducted studies have shown a significant correlation between the projected time of project implementation through network and mathematical models. That is, there is a direct relationship between the calculated time of implementation of mathematical and network model projects. Comparative analysis has shown that from the point of view of the minimum project execution time, the forecast of the PERT method network model is more flexible and relevant for managing small groups of projects.

A General Iterative Clustering Algorithm.

The quality of a cluster analysis of unlabeled units depends on the quality of the between units dissimilarity measures. Data‐dependent dissimilarity is more objective than data independent geometric measures such as Euclidean distance. As suggested by Breiman, many data driven approaches are based on decision tree ensembles, such as a random forest (RF), that produce a proximity matrix that can easily be transformed into a dissimilarity matrix. An RF can be obtained using labels that distinguish units with real data from units with synthetic data. The resulting dissimilarity matrix is input to a clustering program and units are assigned labels corresponding to cluster membership. We introduce a general iterative cluster (GIC) algorithm that improves the proximity matrix and clusters of the base RF. The cluster labels are used to grow a new RF yielding an updated proximity matrix, which is entered into the clustering program. The process is repeated until convergence. The same procedure can be used with many base procedures such as the extremely randomized tree ensemble. We evaluate the performance of the GIC algorithm using benchmark and simulated data sets. The properties measured by the Silhouette score are substantially superior to the base clustering algorithm. The GIC package has been released in R: https://cran.r‐project.org/web/packages/GIC/index.html.

ИССЛЕДОВАНИЕ НАБОРА ТЕХНОЛОГИЧЕСКИХ ОПЕРАЦИЙ ПОДГОТОВКИ СЕМЕННОГО МАТЕРИАЛА ХВОЙНЫХ ПОРОД ДЛЯ ЛЕСОВОССТАНОВЛЕНИЯ

Preliminary studies have shown the need to develop an integrated approach to the problem of reforestation, which would take into account current trends in the field of ecophysiology, technology and technology of forestry, as well as the efficiency of technological processes. At the first stage, a reforestation algorithm was developed to implement this approach. The algorithm consists of 31 enlarged technological operations and 268 implementations of the algorithm. The algorithm was decomposed for further detailed study of technological operations and the development of a mathematical model aimed at finding the optimal set of operations for reforestation. Six groups were identified, each of which should be considered separately. The work considers the 10th operator of the III-th and V-th groups. An algorithm for preparing seeds and an algorithm for separating seeds have been developed. The matrices for the implementation of the algorithms were constructed (a 13 × 16-dimensional matrix for the preparation of forest seeds and a 14×35-dimensional matrix for the separation of forest seeds, containing binary values). A visualized proximity matrix based on the squared Euclidean distance and the Ward method was used to study the operators of algorithms for similarity. As a result of data processing, the operators of the algorithm are determined, which are far enough from the center of the groups, which can be ignored in the future when developing an optimization mathematical model of the reforestation process

Adaptive-order proximity learning for graph-based clustering

Recently, structured proximity matrix learning, which aims to learn a structured proximity matrix with explicit clustering structures from the first-order proximity matrix, has become the mainstream of graph-based clustering. However, the first-order proximity matrix always lacks several must-links compared to the groundtruth in real-world data, which results in a mismatched problem and affects the clustering performance. To alleviate this problem, this work introduces the high-order proximity to structured proximity matrix learning, and explores a novel framework named Adaptive-Order Proximity Learning (AOPL) to learn a consensus structured proximity matrix from the proximities of multiple orders. To be specific, AOPL selects the appropriate orders first, then assigns weights to these selected orders adaptively. In this way, a consensus structured proximity matrix is learned from the proximity matrices of appropriate orders. Based on AOPL framework, two practical models with different properties are derived, namely AOPL-Root and AOPL-Log. Besides, AOPL and the derived models are regarded as the same optimization problem subjected to some slightly different constraints. An efficient algorithm is proposed to solve them and the corresponding theoretical analyses are provided. Extensive experiments on several real-world datasets demonstrate superb performance of our model.

Reference-free method for investigating classification uncertainty in large-scale land cover datasets

In the last decade, land cover products are produced at a global scale and updated with an unprecedent speed with the development of earth observation and mapping techniques. However, assessing large-scale land cover datasets is always a challenging task because of lack of ground truth and high dependency on manual inspection. To promote the efficiency of assessment, this study introduces a generic framework that identifies potential land cover classification errors without traditional reference data. Depending on rich features in remote sensing data, the overall procedure of can be regarded as a multi-class anomaly detection problem. To improve the performance in dealing with complex classification scheme, a pairing strategy is firstly proposed for pairwise analysis on each two classes, so that the multi-class problem is decomposed into multiple smaller binary-class problems. Secondly, the proximity matrix of each class pair is generated on the basis of extra trees model, which is applicable to mixed numerical and categorical data and provides robust proximity measurement by including multiple levels of randomness. Finally, an improved proximity-based anomaly detection algorithm is applied, and the detection results of each class pair are ensembled to obtain the final anomaly score of each land instance. Experiments show that the overall area under the receiver operating characteristic curve score reaches approximately 0.9 for synthetic datasets and 0.8 for real-life datasets. The proposed method is expected to provide users valuable data for a more efficient assessment in the absence of ground truth in practice.

Proximal Support Matrix Machine

In this paper, we have proposed a novel model called proximal support matrix machine (PSMM), which is mainly based on the models of proximal support vector machine (PSVM) and low rank support matrix machine (LRSMM). In design, the PSMM model has comprehensively considered both the relationship between samples of the same class and the structure of rows or columns of matrix data. To a certain extent, our novel model can be regarded as a synthesis of the PSVM model and the LRSMM model. Since the PSMM model is an unconstrained convex problem in essence, we have established an alternating direction method of multipliers algorithm to deal with the proposed model. Finally, since a great deal of experiments on the minst digital database show that the PSMM classifier has a good ability to distinguish two digits with little difference, it encourages us to conduct more complex experiments on MIT face database, INRIA person database, the students face database and Japan female facial expression database. Meanwhile, the final experimental results show that PSMM performs better than PSVM, twin support vector machine, LRSMM and linear twin multiple rank support matrix machine in the demanding image classification tasks.

Assessment of genome stability of pfleFB hybrid generations through molecular DNA markers

Using PEG-mediated protoplast fusion, a total of nine pfle somatic hybrids were developed between Pleurotus florida and Lentinula edodes. Only six could developed fruit bodies and the second generation was made of each through tissue culture isolate from their fruit bodies. The first generation were denoted as pfle1p, pfle1q, pfle1v, pfle 1s, pfle1o and the second generation were as pfle1pFB, pfle1qFB, pfle1vFB, pfle 1sFB, pfle1oFB. The genetic closeness, stability and variance is analyzed compared to their first parental strains using nine RAPD molecular markers. Amplified PCR bands were scored analyzed using SPSS software version 19. Polymorphisms were found calculating a total of 104 bands is 94.23% where the highest polymorphism was observed in RAPD-02, RAPD- 03, RAPD-08 and SS-11. Maximum bands (16) were generated by SRS-06 and minimum in SRS-05, ranged from 330 bp – 3000 bp and 450 bp – 2000 bp, respectively. Jaccards proximity matrix was generated from the scored data followed by dendogram including two generations with parents, using single linkage. The proximity matrix was ranged from 8.00-60.00. Two parental strains P. florida and L. edodes found to be most distantly related among all, where pfle 1pFB, pfle 1rFB were very close in genetic distance and another two hybrid lines, pfle 1vFB, pfle 1qFB were also showed the same. It was observed that the nearest neighbor showed a little variance genetically, but the hybrid lines are mostly similar with their second generation.

Multiple Kernel Representation Learning on Networks

Learning representations of nodes in a low dimensional space is a crucial task with numerous interesting applications in network analysis, including link prediction, node classification, and visualization. Two popular approaches for this problem are matrix factorization and random walk-based models. In this paper, we aim to bring together the best of both worlds, towards learning node representations. In particular, we propose a weighted matrix factorization model that encodes random walk-based information about nodes of the network. The benefit of this novel formulation is that it enables us to utilize kernel functions without realizing the exact proximity matrix so that it enhances the expressiveness of existing matrix decomposition methods with kernels and alleviates their computational complexities. We extend the approach with a multiple kernel learning formulation that provides the flexibility of learning the kernel as the linear combination of a dictionary of kernels in data-driven fashion. We perform an empirical evaluation on real-world networks, showing that the proposed model outperforms baseline node embedding algorithms in downstream machine learning tasks.