Neighbor Weight Research Articles

Neighbor Weighting and Distance Metrics in Nearest Neighbor Nowcasting of Swedish GDP

Neighbor Weighting and Distance Metrics in Nearest Neighbor Nowcasting of Swedish GDP

Exploring menopausal dynamics: a cross-sectional analysis of age, symptomatology, and sociodemographic influences in a developing population of women aged 40-60

Objectives: Menopause, a biological milestone, marks a pivotal phase in women's lives characterized by ovarian function cessation and age-related changes. Our objective was to investigate menopausal symptoms and knowledge among women aged 40-60 years. Methods: This cross-sectional epidemiological study was conducted between June 1 and September 30, 2005, in the Nilufer Public Health Education and Research Area (NPHERA) region, aimed to assess menopausal symptoms and their correlates among 1013 women aged 40-60. The individuals included in the study were selected through a systematic sampling method, stratified by neighborhood weights and age groups based on the NPHERA 2004 Work Report and regional data, as well as information from the Health Centers Information System (HCIS), where the Electronic Health Records (EHR) are registered. Results: The mean age of natural menopause was found to be 46.7±4.8 years, showcasing sociodemographic factors' influence. Postmenopausal women experienced higher rates of symptoms, with physical and mental exhaustion (82.8%), irritability (78.4%), and depressive mood (76.4%) prevailing. Logistic regression revealed that employment status significantly influenced menopausal status. Moreover, the age at menopause correlated positively with the age of the woman's mother. Conclusion: This study contributes insights into menopausal experiences in developing countries, emphasizing the need for tailored healthcare approaches. Longitudinal investigations are warranted to comprehensively understand these associations and enhance women's quality of life during menopause.

A Study on Utility Analysis and Risk Prediction of Economic Management of Property Enterprises in Digital Economy

Abstract In the era of the digital economy, there is a widespread trend for property enterprise economic management to undergo digital transformation. The article examines how the digital economy propels the economic management of property enterprises, scrutinizing its risk components and the digital ecological model of management services. The evaluation index system for the digital economy and the economic management level of property enterprises has been designed, and its comprehensive score is solved using the entropy weight method. Based on the comprehensive score, a spatial Durbin model was established, and spatial autocorrelation tests and econometric regression analyses were performed. Additionally, the risk prediction model for the economic management of property enterprises was established using BP neural networks and risk evaluation indexes. The global Moran’s mean values for the economic management of property enterprises in the digital economy were 0.267 and 0.081, respectively, and both passed the 1% significance test. The development level of the digital economy significantly promotes the economic management of property enterprises at the 1% level, as indicated by the influence coefficients obtained from the neighbor weight matrix, geographic weight matrix, and economic distance weight matrix, which are 2.236, 2.132, and 2.078, respectively. The BP neural network predicts economic management risk for property enterprises with an accuracy of 91.95%. Relying on digital economy-related technology can effectively promote the improvement of the economic management level of property enterprises and can also effectively enable risk prediction in economic management.

Adaptive Label Propagation for Group Anomaly Detection in Large-Scale Networks

This paper concentrates on group anomalies in general large-scale networks. Existing algorithms on group anomalies mainly focus on homogeneous or bipartite networks, and thus are difficult to apply to heterogeneous networks directly. Moreover, these algorithms follow the non-overlapping hypothesis of groups implicitly, which is improper in many scenarios. In this paper, we introduce a novel algorithm called Adaptive Label Propagation (ALP) to solve these problems. ALP is designed based on label propagation (LP) frameworks, for the reason that LP-based frameworks are simple in thought and easy to scale. ALP is able to find overlapping groups by label propagation with belonging coefficients, and can be applied to heterogeneous networks for its design of adaptive neighbor weighting. Assigning different weights to neighbors in label propagation is a challenging task. Inspired by the combinatorial multi-armed bandit mechanism, ALP views the neighbors of each node as arms to be selected, and iteratively updates their weights by evaluating their expected rewards in following iterations. Experiments are conducted on four real-world networks. The results show that LP-based methods are effective for detecting group anomalies, and the comparison results with several state-of-the-art label propagation based community detection methods show the effectiveness of the proposed method

Spatial Spillover Effect of Water Environment Pollution Control in Basins—Based on Environmental Regulations

The basin economy is essential in China’s high-quality development, which brings prodigious economic and social benefits. However, high industrialization and urbanization led to a significant escalation in water pollution within basins. The achievement of synergistic control of water environment pollution at the basin scale has emerged as a primary goal for governmental departments in developing environmental policies. This study constructs a “four-in-one” basin’s water environment regulation system comprising four categories: total, quality, project, and governance. Further, the spatial spillover effects of water environment pollution control in 27 Chinese provincial-level administrative regions spanning key river basins were analyzed. This study aims to provide evidence from China on the transfer of water pollution in basins worldwide. The results indicate the following conclusions: (1) A consistent upward trend in the environmental regulations and water environment pollution indexes across all provinces in the basin. Compared to 2006, in 2018, the mean value of the basin environmental regulation index was 0.21, which is an increase of 23.5%, and the mean value of the water environmental pollution index was 0.83, which is an increase of 29.7%. (2) Whether in the basin’s upper, middle, or lower reaches, the relationship between environmental regulations and water pollution follows a “U”-shaped pattern, with thresholds of 0.318, 0.331, and 0.390, respectively. (3) Under the neighbor weight matrix and water flow distance weight matrix, the impact coefficient of the water environment pollution index on the surrounding areas is significantly positive, implying that the implementation of local environmental policies will radiate to the neighboring areas in the short term and bring positive governance effects. (4) Regarding the time-lag effect, it is observed that the lag term associated with the water environment pollution index exhibits a statistically significant positive relationship. This finding suggests that the pollution of the water environment within the basin follows a cumulative and continuous pattern. (5) It is noted that long-term environmental regulation measures do not contribute favorably to the amelioration of water pollution in the neighboring regions. This implies the presence of a characteristic neighboring avoidance effect.

Time Image De-Noising Method Based on Sparse Regularization

The blurring of texture edges often occurs during image data transmission and acquisition. To ensure the detailed clarity of the drag-time images, we propose a time image de-noising method based on sparse regularization. First, the image pixel sparsity index is set, and then an image de-noising model is established based on sparse regularization processing to obtain the neighborhood weights of similar image blocks. Second, a time image de-noising algorithm is designed to determine whether the coding coefficient reaches the standard value, and a new image de-noising method is obtained. Finally, the images of electronic clocks and mechanical clocks are used as two kinds of time images to compare different image de-noising methods, respectively. The results show that the sparsity regularization method has the highest peak signal-to-noise ratio among the six compared methods for different noise standard deviations and two time images. The image structure similarity is always above which shows that the proposed method is better than the other five image de-noising methods.

Use of weighted local constant method to short-term forecasting of electric load in cities at weekends

To forecast accurately the electric load of a city at the weekend, aiming at the problems of nonlinearity and model interpretability in power behavior, a local constant first-order weighted forecasting model is developed. First of all, the values of the electric load are reconstructed in phase space, stability analysis can effectively describe the operation mechanism and law of weekend power. Then, the weights of neighbors are used in the first-order local forecasting model, and the nearest neighbors are used in grid optimization. The forecasted values are then separated to be the phase forecasting points in terms of the spatial dimension of the electric loads and the lags of the time delay obtained in the training stage. Numerical experimental results of samples at different scales show the following: (1)The forecasting performance of the proposed model is better than other models in the forecasting errors in this work (among three types of measurements: root mean square error (RMSE), mean absolute percentage error (MAPE), and mean absolute error (MAE)); (2) the significance test (KSPA) also confirmed the universality and consistency of the method, and the effective diagnosis and statistical analysis of the weekend power mechanism are helpful to promote weekend power management and commercial design.

Topological Pooling on Graphs

Graph neural networks (GNNs) have demonstrated a significant success in various graph learning tasks, from graph classification to anomaly detection. There recently has emerged a number of approaches adopting a graph pooling operation within GNNs, with a goal to preserve graph attributive and structural features during the graph representation learning. However, most existing graph pooling operations suffer from the limitations of relying on node-wise neighbor weighting and embedding, which leads to insufficient encoding of rich topological structures and node attributes exhibited by real-world networks. By invoking the machinery of persistent homology and the concept of landmarks, we propose a novel topological pooling layer and witness complex-based topological embedding mechanism that allow us to systematically integrate hidden topological information at both local and global levels. Specifically, we design new learnable local and global topological representations Wit-TopoPool which allow us to simultaneously extract rich discriminative topological information from graphs. Experiments on 11 diverse benchmark datasets against 18 baseline models in conjunction with graph classification tasks indicate that Wit-TopoPool significantly outperforms all competitors across all datasets.

Sequential recommendation model integrating micro-behaviors and attribute enhancement

Sequential recommendation predicts the items that the user may interact with next based on the time-series information of the user-item interactions, and learns the users’ dynamic preferences. However, most existing sequential recommendation models ignore the micro-behaviors and the importance of attribute information. Thus, we propose a new model named Sequential Recommendation Model Integrating Micro-behaviors and Attribute Enhancement (SRMA). First, we build a user-item interaction graph and a user-item-attribute interaction graph by introducing user and item attributes. In addition, we perform the temporal attention embedding propagation in the user-item interaction graph, in which the multi-head attention is used to learn the temporal neighborhood weights under micro-behaviors. Simultaneously, we perform the attribute attention embedding propagation in the user-item-attribute interaction graph, which learns the high-hop interactions among users, items and attributes under micro-behaviors, and assigns different weights to attributes through the attribute attention. Finally, the prediction is made by combining the embedding of each layer in the two graphs. Experiments on two real datasets show that the model has good performance.

Supervised Machine Learning and Bayesian Regression Kriging with Application to COVID-19 Incidences in Sub-Saharan Africa

Since COVID-19 invasion of the World, human life has been affected greatly. Several studies have shown a positive correlation between COVID-19 infections and pre-existing conditions such as Diabetes, Cancer, Tuberculosis, and Hypertension. In this study, we would like to determine whether demographic variables have a contribution to the spread of COVID-19 infections. We will apply a machine language method to select the demographic variables which are impactful in the spread of COVID-19 cases in Sub-Saharan Africa. Then we shall determine the nature of COVID-19 cases patterns applying the K-Nearest Neighbor (KNN) in calculating the neighborhood weights between locations/countries. The weights would then be tested for significance to conclude whether the cases patterns are either random, sparsely or clustered. We would then perform simulations to estimate the social demographic/covariates/fixed effects parameters and the random effects parameters. The Bayesian Kriging would be applied to predict Covid-19 cases based on the estimated social demographical variables coefficients/parameters and the random effects parameters in unknown/new locations in Sub Saharan Africa with a known uncertainty. The results showed that Children aged (0-14) years living with HIV AIDS, Prevalence of HIV Total (percentage of population ages 15-49) and Access to electricity (as a percentage of the population) was estimated to contribute to the increase of COVID-19 cases. Prediction of the COVID-19 cases in unknown locations showed that most of the cases were predicted in the elevated locations/areas than in the lower/flatter locations. This could mean that high elevated areas are associated with lower temperatures which increases the spread of COVID-19 cases as opposed to lower/flatter areas which are associated with higher temperatures which reduces the spread of COVID-19 cases.

DMGF-Net: An Efficient Dynamic Multi-Graph Fusion Network for Traffic Prediction

Traffic prediction is the core task of intelligent transportation system (ITS) and accurate traffic prediction can greatly improve the utilization of public resources. Dynamic interaction of multiple spatial relationships will influence the accuracy of traffic prediction. However, many existing methods only consider static spatial relationships, which restricts the accuracy of the prediction. To address the above problem, in this article, we propose the Dynamic Multi-Graph Fusion Network (DMGF-Net) to model the spatial-temporal correlations in traffic network. In the DMGF-Net, the fusion graph is designed to leverage and extract the various spatial correlations between different regions by fusing spatial graph, semantic graph, and spatial-semantic graph. Further, to dynamically learn the importance of different neighbors, we design the Dynamic Spatial-Temporal Unit (DSTU), which can adjust the aggregation weights of different neighbors by combining the convolution operation and the attention mechanism. It can selectively aggregate spatial-temporal features from different neighbors. Extensive experiments on three datasets demonstrate that effectiveness of our model, especially on PEMS08, our model achieves an increase of about 8.55% and 7.55% in terms of MAE and RMSE than the static model STGCN.

Who are the evil backstage manipulators: Boosting graph attention networks against deep fraudsters

Telecommunications fraud causes vast economic losses all over the world every year. Although Graph Neural Networks (GNNs) brings new possibilities to the solution of fraud detection problem, their performance in telecom fraud detection is not satisfying. An important reason is that little prior work has noticed the deep fraudster phenomenon, which makes it difficult to be eradicated. Another reason may be the shortage of available datasets for researchers. After all, few operators are willing to publish telecom fraud detection benchmark datasets due to the sensitivity and privacy of subscriber data, resulting in less related research work. In this paper, we rearrange and publish two real-world telecom fraud detection datasets based on publicly available sources on the Internet. And we propose a GNN-based semi-supervised telecom fraud detection method by Boosting graph Attention Networks Against Deep frAudsters (BANADA). Specifically, we first learn different neighbor weights with the help of inner-layer attention, and then design simplified inter-layer attention to achieve inter-layer feature aggregation. Finally, we use GNNs with different depths as different base classifiers and perform ensemble learning through AdaBoost to discriminate deep fraudsters. Extensive experiments on two real-world telecom fraud datasets demonstrate the effectiveness of our proposed BANADA, which outperforms all state-of-the-art GNNs and GNN-based fraud detectors. The BANADA code and datasets are available at https://github.com/xxhu94/BANADA.

Confidence of a k-Nearest Neighbors Python Algorithm for the 3D Visualization of Sedimentary Porous Media

In a previous paper, the authors implemented a machine learning k-nearest neighbors (KNN) algorithm and Python libraries to create two 3D interactive models of the stratigraphic architecture of the Quaternary onshore Llobregat River Delta (NE Spain) for groundwater exploration purposes. The main limitation of this previous paper was its lack of routines for evaluating the confidence of the 3D models. Building from the previous paper, this paper refines the programming code and introduces an additional algorithm to evaluate the confidence of the KNN predictions. A variant of the Similarity Ratio method was used to quantify the KNN prediction confidence. This variant used weights that were inversely proportional to the distance between each grain-size class and the inferred point to work out a value that played the role of similarity. While the KNN algorithm and Python libraries demonstrated their efficacy for obtaining 3D models of the stratigraphic arrangement of sedimentary porous media, the KNN prediction confidence verified the certainty of the 3D models. In the Llobregat River Delta, the KNN prediction confidence at each prospecting depth was a function of the available data density at that depth. As expected, the KNN prediction confidence decreased according to the decreasing data density at lower depths. The obtained average-weighted confidence was in the 0.44−0.53 range for gravel bodies at prospecting depths in the 12.7−72.4 m b.s.l. range and was in the 0.42−0.55 range for coarse sand bodies at prospecting depths in the 4.6−83.9 m b.s.l. range. In a couple of cases, spurious average-weighted confidences of 0.29 in one gravel body and 0.30 in one coarse sand body were obtained. These figures were interpreted as the result of the quite different weights of neighbors from different grain-size classes at short distances. The KNN algorithm confidence has proven its suitability for identifying these anomalous results in the supposedly well-depurated grain-size database used in this study. The introduced KNN algorithm confidence quantifies the reliability of the 3D interactive models, which is a necessary stage to make decisions in economic and environmental geology. In the Llobregat River Delta, this quantification clearly improves groundwater exploration predictability.

Identification of Important Nodes in Multilayer Heterogeneous Networks Incorporating Multirelational Information

Centrality is an effective method to identify important nodes in complex networks, but it is still a challenge to find influential nodes by making full use of multiple relationships and global network topological features in complex networks. To address these problems, this article proposes an importance identification method for multilayer heterogeneous network node by incorporating multirelational information (MLC). This method studies the relational characteristics of heterogeneous nodes in detail and divides the heterogeneous nodes into different layers according to the node types, which can be further divided into core and auxiliary layers. The importance of the auxiliary layer is quantified by designing the interlayer influence and determining the interlayer influence weights of different connectivity influences; the centrality score of heterogeneous nodes under multiconnectivity relationships is fused using the transmission characteristics of internode relationships in the auxiliary layer, which in turn measures the importance of nodes in the core layer. To evaluate the proposed algorithm, we conduct experiments on five real multilayer heterogeneous networks of different sizes. The results show that MLC can make full use of different types of internode association relationship information, effectively fuse network structure information such as the neighbor weights of core and auxiliary layer nodes, and outperform the existing techniques in identifying important nodes.

Eigenvalue distribution of large random matrices arising in deep neural networks: Orthogonal case

This paper deals with the distribution of singular values of the input–output Jacobian of deep untrained neural networks in the limit of their infinite width. The Jacobian is the product of random matrices where the independent weight matrices alternate with diagonal matrices whose entries depend on the corresponding column of the nearest neighbor weight matrix. The problem has been considered in the several recent studies of the field for the Gaussian weights and biases and also for the weights that are Haar distributed orthogonal matrices and Gaussian biases. Based on a free probability argument, it was claimed in those papers that, in the limit of infinite width (matrix size), the singular value distribution of the Jacobian coincides with that of the analog of the Jacobian with special random but weight independent diagonal matrices, the case well known in random matrix theory. In this paper, we justify the claim for random Haar distributed weight matrices and Gaussian biases. This, in particular, justifies the validity of the mean field approximation in the infinite width limit for the deep untrained neural networks and extends the macroscopic universality of random matrix theory to this new class of random matrices.

Image retrieval based on weighted nearest neighbor tag prediction

Abstract With the development of communication and computer technology, the application of big data technology has become increasingly widespread. Reasonable, effective, and fast retrieval methods for querying information from massive data have become an important content of current research. This article provides an image retrieval method based on the weighted nearest neighbor label prediction for the problem of automatic image annotation and keyword image retrieval. In order to improve the performance of the test method, scientific experimental verification was implemented. The nearest neighbor weights are determined by maximizing the training image annotation, and experiments are carried out from multiple angles based on the Mahalanobis metric learning integration model. The experimental results show that the proposed tag correlation prediction propagation model has obvious improvements in accuracy, recall rate, break-even point, and overall average accuracy performance compared with other widely used algorithm models.

Lyapunov exponent in the Vicsek model.

The well-known Vicsek model describes the dynamics of a flock of self-propelled particles (SPPs). Surprisingly, there is no direct measure of the chaotic behavior of such systems. Here we discuss the dynamical phase transition present in Vicsek systems in light of the largest Lyapunov exponent (LLE), which is numerically computed by following the dynamical evolution in tangent space for up to two million SPPs. As discontinuities in the neighbor weighting factor hinder the computations, we propose a smooth form of the Vicsek model. We find a chaotic regime for the collective behavior of the SPPs based on the LLE. The dependence of LLE with the applied noise, used as a control parameter, changes sensibly in the vicinity of the well-known transition points of the Vicsek model.

DETERMINING LEARNING SUCCESS of kNN ALGORITHM on ZOO DATASET

People learn by examining, observing and researching their environment. They actually gains experience from what they have learned. By using the experience they have gained, they can adapt to the new situation they encounter and make decisions. People always make decisions by comparing their previous knowledge while describing objects and classifying them. Similarities and differences to previously learned objects are very effective in decision making. It has been shown in the studies that the experiential learning method can also be used on machines. Intelligent machines and devices that use machine learning methods in their structure are widely used in many areas. Machine learning can be performed using different algorithms. These algorithms use the attributes of the objects in the data set when making decisions. Similarities and differences in the attributes of objects are obtained by comparing them with previous experiences. As a result of the comparison, a decision is made and predictions are made about the classes of the objects. In this study, kNN machine learning algorithm, which is a supervised learning method, was used on the Zoo dataset. In this data set, there are attributes of common living things. By using these attributes, the classes of living things in the data set are determined. The “k” neighbor value and weight parameter selected in the kNN algorithm affect the learning success. In this study, the effect of two parameters used in the kNN algorithm on learning success is shown. According to the results obtained, the "k=1" neighbor value and the "Distance Weight" parameter were selected and the highest success result was obtained.

AN INTERPRETABLE MACHINE LEARNING MODEL FOR INDIVIDUALIZED PROTOCOL SELECTION AND GONADOTROPIN DOSE SELECTION DURING OVARIAN STIMULATION

To develop an interpretable machine learning model for individualized gonadotropin dose selection during controlled ovarian stimulation. Historical, de-identified electronic medical record (EMR) data was collected from 4 IVF clinics in the United States. Records were filtered for autologous, non-canceled IVF retrievals, resulting in 7,977 cycles started between 2014 and 2020. A multiple linear regression model was developed with cross validation and recursive feature elimination to predict the number of eggs and mature (MII) eggs retrieved using baseline parameters available prior to start of treatment. The predictor variables were then used to create a patient similarity model based on K nearest neighbors (KNN), an interpretable machine learning technique. After identifying the best performing distance metrics, neighbor weights, and number of neighbors, the model was used to predict the number of eggs and MII eggs retrieved by calculating the weighted average from the set of K neighbors most similar to the patient of interest. The performance of the KNN model was compared to linear regression in terms of R-squared (R2) and mean absolute error (MAE). The KNN model was then used to (a) query the K most similar patients, and (b) identify the optimal gonadotropin dose in terms of highest number of MII eggs retrieved. We developed linear regression and KNN models using patient age, BMI, diagnosis, AMH, AFC, number of previous IVF cycles, and parity. KNN achieved highest performance using the Manhattan distance, 50-80 similar patients, and distance-based neighbor weighting. The KNN model outperformed linear regression for eggs retrieved (R2: 0.43 vs. 0.39, MAE: 4.84 vs. 4.98) and for MII eggs retrieved (R2: 0.39 vs. 0.35, MAE: 4.01 vs. 4.11). We then investigated the application of these models for gonadotropin dose selection. Linear models indicated that gonadotropin dose is negatively correlated with MII eggs, which may in part reflect that poor-prognosis patients are prescribed higher doses. In contrast, the KNN model showed that 22% of patients had a concave dose response curve, in which there was an optimal dose that maximized the number of MII eggs. We developed a patient similarity model using K nearest neighbors. The model showed better accuracy than linear regression for predicting eggs and MII eggs retrieved, and allowed the evaluation of which starting dose maximized the number of MII eggs retrieved, which is not possible with a linear model. Future work will optimize techniques for matching similar patients and extend the modeling for protocol selection.

Phase lag index-based graph attention networks for detecting driving fatigue.

It is important to understand the changes in the characteristics of the brain network in the state of driving fatigue and to reveal the pattern of functional connectivity between brain regions when fatigue occurs. This paper proposes a method for the detection of driving fatigue based on electroencephalogram (EEG) signals using a phase lag indexgraph attention network (PLI-GAT). Phase synchronization between EEG signals is a key attribute for establishing communication links among different regions of the brain, and so, the PLI is used to construct a functional brain network reflecting the relationship between EEG signals from different channels. Multi-channel EEG time-frequency features are then modeled as graph data, and the driving fatigue monitoring model is trained using a GAT. Compared with traditional graph neural networks, the GAT applies an aggregation operation to adjacent EEG channel features through the attention mechanism. This enables the adaptive assignment of different neighbor weights, which greatly improves the expressiveness of the graph neural network model. The proposed method is validated on the publicly available SEED-VIG dataset, and the accuracy of fatigue state recognition is found to reach 85.53%. The results show that the functional connectivity among different channels is significantly enhanced in the fatigue state.