Clusters In Dataset Research Articles

Variational Graph Autoencoder with Adversarial Mutual Information Learning for Network Representation Learning

With the success of Graph Neural Network (GNN) in network data, some GNN-based representation learning methods for networks have emerged recently. Variational Graph Autoencoder (VGAE) is a basic GNN framework for network representation. Its purpose is to well preserve the topology and node attribute information of the network to learn node representation, but it only reconstructs network topology, and does not consider the reconstruction of node features. This strategy will make node representation can not well reserve node features information, impairing the ability of the VGAE method to learn higher quality representations. To solve this problem, we arise a new network representation method to improve the VGAE method for well retaining both node features and network structure information. The method utilizes adversarial mutual information learning to maximize the mutual information (MI) of node features and node representations during the encoding process of the variational autoencoder, which forces the variational encoder to get the representation containing the most informative node features. The method consists of three parts: a variational graph autoencoder includes a variational encoder (MI generator (G)) and a decoder, a positive MI sample module (maximizing MI module), and an MI discriminator (D). Furthermore, we explain why maximizing MI between node features and node representation can reconstruct node attributes. Finally, we conduct experiments on seven public representative datasets for nodes classification, nodes clustering, and graph visualization tasks. Experimental results demonstrate that the proposed algorithm significantly outperforms current popular network representation algorithms on these tasks. The best improvement is 17.13% than the VGAE method.

Flight Delay Prediction Model Based on Lightweight Network ECA-MobileNetV3

In exploring the flight delay problem, traditional deep learning algorithms suffer from low accuracy and extreme computational complexity; therefore, the deep flight delay prediction algorithm is difficult to directly deploy to the mobile terminal. In this paper, a flight delay prediction model based on the lightweight network ECA-MobileNetV3 algorithm is proposed. The algorithm first preprocesses the data with real flight information and weather information. Then, in order to increase the accuracy of the model without increasing the computational complexity too much, feature extraction is performed using the lightweight ECA-MobileNetV3 algorithm with the addition of the Efficient Channel Attention mechanism. Finally, the flight delay classification prediction level is output via a Softmax classifier. In the experiments of single airport and airport cluster datasets, the optimal accuracy of the ECA-MobileNetV3 algorithm is 98.97% and 96.81%, the number of parameters is 0.33 million and 0.55 million, and the computational volume is 32.80 million and 60.44 million, respectively, which are better than the performance of the MobileNetV3 algorithm under the same conditions. The improved model can achieve a better balance between accuracy and computational complexity, which is more conducive mobility.

Learning a restricted Boltzmann machine using biased Monte Carlo sampling

Restricted Boltzmann Machines are simple and powerful generative models that can encode any complex dataset. Despite all their advantages, in practice the trainings are often unstable and it is difficult to assess their quality because the dynamics are affected by extremely slow time dependencies. This situation becomes critical when dealing with low-dimensional clustered datasets, where the time required to sample ergodically the trained models becomes computationally prohibitive. In this work, we show that this divergence of Monte Carlo mixing times is related to a phenomenon of phase coexistence, similar to that which occurs in physics near a first-order phase transition. We show that sampling the equilibrium distribution using the Markov chain Monte Carlo method can be dramatically accelerated when using biased sampling techniques, in particular the Tethered Monte Carlo (TMC) method. This sampling technique efficiently solves the problem of evaluating the quality of a given trained model and generating new samples in a reasonable amount of time. Moreover, we show that this sampling technique can also be used to improve the computation of the log-likelihood gradient during training, leading to dramatic improvements in training RBMs with artificial clustered datasets. On real low-dimensional datasets, this new training method fits RBM models with significantly faster relaxation dynamics than those obtained with standard PCD recipes. We also show that TMC sampling can be used to recover the free-energy profile of the RBM. This proves to be extremely useful to compute the probability distribution of a given model and to improve the generation of new decorrelated samples in slow PCD-trained models. The main limitations of this method are, first, the restriction to effective low-dimensional datasets and, second, the fact that the Tethered MC method breaks the possibility of performing parallel alternative Monte Carlo updates, which limits the size of the systems we can consider in practice.

Dealing with complete separation and quasi-complete separation in logistic regression for linguistic data

Logistic regression is a powerful and widely used analytical tool in linguistics for modelling a binary outcome variable against a set of explanatory variables. One challenge that can arise when applying logistic regression to linguistics data is complete or quasi-complete separation, phenomena that occur when (paradoxically) the model has too much explanatory power, resulting in effectively infinite coefficient estimates and standard errors. Instead of seeing this as a drawback of the method, or naïvely removing covariates that cause separation, we demonstrate a straightforward and user-friendly modification of logistic regression, based on penalising the coefficient estimates, that is capable of systematically handling separation. We illustrate the use of penalised, multi-level logistic regression on two clustered datasets relating to second language acquisition and corpus data, showing in both cases how penalisation remedies the problem of separation and thus facilitates sensible and valid statistical conclusions to be drawn. We also show via simulation that results are not overly sensitive to the amount of penalisation employed for handling separation.

Rank Based Iterative Clustering (RBIC) for indoor localization

In the current era, technological advancements have resulted in a spike in various sensor data-intensive applications including indoor localization services. WiFi fingerprint-based localization for large indoor areas is primarily contingent on laborious site surveys. To ease the flow of research, grossly labeled data collected using a crowd-sourced method could be clustered. However, for a real-time localization approach, the challenge is to generate the clusters containing data samples that are really physically close. In this paper, a two-phase semi-supervised localization approach has been proposed, which is general enough to be applied to indoor localization datasets. In the offline phase, a Rank-Based Iterative Clustering (RBIC) algorithm has been proposed that generates a clustered dataset with a negligible chance of containing physically apart location points within a common cluster. RBIC can be viewed as a clustering ensemble model. Different clustering algorithms are selected as baseline algorithms and assigned unique ranks depending on well-known clustering scores to be fed as input to RBIC. In the online phase, the users’ location is estimated using machine learning (ML) classifiers based on the dynamic received signal strength indicator (RSSI) vector received through its handheld device transceiver. The system is evaluated with three benchmark datasets for WiFi based indoor localization. For the first dataset, JUIndoorLoc, 94% to 99% localization accuracy is achieved for individual supervised classifiers. For the second and third datasets, the ranges of obtained localization accuracies are 96% to 99% and 95% to 98% respectively.

Scalable and Efficient Clustering for Fingerprint-Based Positioning

Indoor positioning based on IEEE 802.11 wireless LAN (Wi-Fi) fingerprinting needs a reference data set, also known as a radio map, in order to match the incoming fingerprint in the operational phase with the most similar fingerprint in the data set and then estimate the device position indoors. Scalability problems may arise when the radio map is large, e.g., providing positioning in large geographical areas or involving crowdsourced data collection. Some researchers divide the radio map into smaller independent clusters, such that the search area is reduced to less dense groups than the initial database with similar features. Thus, the computational load in the operational stage is reduced both at the user devices and on servers. Nevertheless, the clustering models are machine-learning algorithms without specific domain knowledge on indoor positioning or signal propagation. This work proposes several clustering variants to optimize the coarse and fine-grained search and evaluates them over different clustering models and data sets. Moreover, we provide guidelines to obtain efficient and accurate positioning depending on the data set features. Finally, we show that the proposed new clustering variants reduce the execution time by half and the positioning error by <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\approx 7$ </tex-math></inline-formula> % with respect to fingerprinting with the traditional clustering models.

SupCAM: Chromosome cluster types identification using supervised contrastive learning with category-variant augmentation and self-margin loss.

Chromosome segmentation is a crucial analyzing task in karyotyping, a technique used in experiments to discover chromosomal abnormalities. Chromosomes often touch and occlude with each other in images, forming various chromosome clusters. The majority of chromosome segmentation methods only work on a single type of chromosome cluster. Therefore, the pre-task of chromosome segmentation, the identification of chromosome cluster types, requires more focus. Unfortunately, the previous method used for this task is limited by the small-scale chromosome cluster dataset, ChrCluster, and needs the help of large-scale natural image datasets, such as ImageNet. We realized that semantic differences between chromosomes and natural objects should not be ignored, and thus developed a novel two-step method called SupCAM, which could avoid overfitting only using ChrCluster and achieve a better performance. In the first step, we pre-trained the backbone network on ChrCluster following the supervised contrastive learning framework. We introduced two improvements to the model. One is called the category-variant image composition method, which augments samples by synthesizing valid images and proper labels. The other introduces angular margin into large-scale instance contrastive loss, namely self-margin loss, to increase the intraclass consistency and decrease interclass similarity. In the second step, we fine-tuned the network and obtained the final classification model. We validated the effectiveness of modules through massive ablation studies. Finally, SupCAM achieved an accuracy of 94.99% with the ChrCluster dataset, which outperformed the method used previously for this task. In summary, SupCAM significantly supports the chromosome cluster type identification task to achieve better automatic chromosome segmentation.

A burst storm from the repeating FRB 20200120E in an M81 globular cluster Dataset

A burst storm from the repeating FRB 20200120E in an M81 globular cluster Dataset

Cloud-Cluster: An uncertainty clustering algorithm based on cloud model

As a cornerstone of the world, uncertainty embodies the nature of data and knowledge. Existing uncertainty theory-based clustering algorithms learn fuzziness, i.e., the uncertainty of clustering objects belonging to different clusters. However, these algorithms do not refer to the fuzziness of objects themselves, i.e., the randomness of data. Here, we propose a clustering algorithm named Cloud-Cluster, which simultaneously characterizes the fuzziness and randomness of objects to reserve uncertain information, and to describe clusters into concepts. It embeds random uncertainty of concepts to extend the data distribution range for better data partitions and gradually constructs accurate concepts by an improved backward cloud transformation algorithm (MBCT-SR-Ex). Moreover, to ensure that the concept clustering process gradually converges, Cloud-Cluster introduces the Cluster Concept Drift Degree to evaluate the uncertainty of concepts during the clustering process. Experiments on UCI and OpenML clustering datasets show that Cloud-Cluster improves the average clustering accuracy by over 14% compared to K-Means and uncertainty theory-based clustering algorithms. Extensive experimental results on the evaluation of uncertainty show that Cloud-Cluster can handle the uncertainty of datasets in the clustering process well, in addition to exhibiting robustness with unclear clusters.

AntiRef: reference clusters of human antibody sequences.

Genetic biases in the human antibody repertoire result in publicly available antibody sequence datasets that contain many duplicate or highly similar sequences. Available datasets are further skewed by the predominance of studies focused on specific disease states, primarily cancer, autoimmunity, and a small number of infectious diseases that includes HIV, influenza, and SARS-CoV-2. These biases and redundancies are a barrier to rapid similarity searches and reduce the efficiency with which these datasets can be used to train statistical or machine-learning models. Identity-based clustering provides a solution; however, the extremely large size of available antibody sequence datasets makes such clustering operations computationally intensive and potentially out of reach for many scientists and researchers who would benefit from such data. Antibody Reference Clusters (AntiRef), which is modeled after UniRef, provides clustered datasets of filtered human antibody sequences. Due to the modular nature of recombined antibody genes, the clustering thresholds used by UniRef for general protein sequences are suboptimal for antibody clustering. Starting with an input dataset of ∼451M full-length, productive human antibody sequences, AntiRef provides reference datasets clustered at a range of antibody-optimized identity thresholds. AntiRef90 is one-third the size of the input dataset and less than half the size of the non-redundant AntiRef100. AntiRef datasets are available on Zenodo (zenodo.org/record/7474336). All code used to generate AntiRef is available on GitHub (github.com/briney/antiref). The AntiRef versioning scheme (current version: v2022.12.14) refers to the date on which sequences were retrieved from OAS.

Tree-Structured Data Clustering-Driven Neural Network for Intra Prediction in Video Coding.

Intra prediction is a crucial part of video compression, which utilizes local information in images to eliminate spatial redundancy. As the state-of-the-art video coding standard, Versatile Video Coding (H.266/VVC) employs multiple directional prediction modes in intra prediction to find the texture trend of local areas. Then the prediction is made based on reference samples in the selected direction. Recently, neural network-based intra prediction has achieved great success. Deep network models are trained and applied to assist the HEVC and VVC intra modes. In this paper, we propose a novel tree-structured data clustering-driven neural network (dubbed TreeNet) for intra prediction, which builds the networks and clusters the training data in a tree-structured manner. Specifically, in each network split and training process of TreeNet, every parent network on a leaf node is split into two child networks by adding or subtracting Gaussian random noise. Then data clustering-driven training is applied to train the two derived child networks using the clustered training data of their parent. On the one hand, the networks at the same level in TreeNet are trained with non-overlapping clustered datasets, and thus they can learn different prediction abilities. On the other hand, the networks at different levels are trained with hierarchically clustered datasets, and thus they will have different generalization abilities. TreeNet is integrated into VVC to assist or replace intra prediction modes to test its performance. In addition, a fast termination strategy is proposed to accelerate the search of TreeNet. The experimental results demonstrate that when TreeNet is used to assist the VVC Intra modes, TreeNet with depth = 3 can bring an average of 3.78% bitrate saving (up to 8.12%) over VTM-17.0. If TreeNet with the same depth replaces all VVC intra modes, an average of 1.59% bitrate saving can be reached.

Self-Supervised Video-Centralised Transformer for Video Face Clustering.

This article presents a novel method for face clustering in videos using a video-centralised transformer. Previous works often employed contrastive learning to learn frame-level representation and used average pooling to aggregate the features along the temporal dimension. This approach may not fully capture the complicated video dynamics. In addition, despite the recent progress in video-based contrastive learning, few have attempted to learn a self-supervised clustering-friendly face representation that benefits the video face clustering task. To overcome these limitations, our method employs a transformer to directly learn video-level representations that can better reflect the temporally-varying property of faces in videos, while we also propose a video-centralised self-supervised framework to train the transformer model. We also investigate face clustering in egocentric videos, a fast-emerging field that has not been studied yet in works related to face clustering. To this end, we present and release the first large-scale egocentric video face clustering dataset named EasyCom-Clustering. We evaluate our proposed method on both the widely used Big Bang Theory (BBT) dataset and the new EasyCom-Clustering dataset. Results show the performance of our video-centralised transformer has surpassed all previous state-of-the-art methods on both benchmarks, exhibiting a self-attentive understanding of face videos.

Load-Aware VM Migration Using Hypergraph Based CDB-LSTM

Live Virtual Machine (VM) migration is one of the foremost techniques for progressing Cloud Data Centers’ (CDC) proficiency as it leads to better resource usage. The workload of CDC is often dynamic in nature, it is better to envisage the upcoming workload for early detection of overload status, underload status and to trigger the migration at an appropriate point wherein enough number of resources are available. Though various statistical and machine learning approaches are widely applied for resource usage prediction, they often failed to handle the increase of non-linear CDC data. To overcome this issue, a novel Hypergrah based Convolutional Deep Bi-Directional-Long Short Term Memory (CDB-LSTM) model is proposed. The CDB-LSTM adopts Helly property of Hypergraph and Savitzky–Golay (SG) filter to select informative samples and exclude noisy inference & outliers. The proposed approach optimizes resource usage prediction and reduces the number of migrations with minimal computational complexity during live VM migration. Further, the proposed prediction approach implements the correlation co-efficient measure to select the appropriate destination server for VM migration. A Hypergraph based CDB-LSTM was validated using Google cluster dataset and compared with state-of-the-art approaches in terms of various evaluation metrics.

Improved Method for Enhanced Quality of Service in IoHT Task Dependency Optimization

Keeping a proper level of task dependency throughout the scheduling process is critical to achieving the aim of decreasing the make-span rate in Internet of Health Things (IoHT) projects. We provide a smart model strategy for effective task scheduling in the IoHT environment for e-healthcare systems by merging hybrid moth flame optimisation (HMFO) with cloud computing. The HMFO algorithm guarantees that all available resources are distributed evenly, resulting in improved quality of service (QoS). We study the Google cluster dataset to learn about the scheduling behaviours of cloud-based jobs in order to train our model. After training, an HMFO model may be used to plan activities in real time. To assess the success of our strategy, we run simulations in the CloudSim environment, taking into account crucial parameters such as resource utilisation, reaction time, and energy consumption. According to a comparative analysis, our hybrid HMFO system surpasses the alternatives in terms of reaction time, average run duration, and cost savings. Our method has proven to be effective due to the favourable effects it has had on response rates, prices, and run times. Combining IoT and cloud computing has the potential to improve healthcare delivery in a variety of ways. One unique strategy we offer for scheduling IOHT jobs is to combine a deep neural network (DNN) algorithm with the MFO technique. Job scheduling in electronic healthcare systems can be optimised with the help of our hybrid MFO-DNN algorithm by taking into account a variety of different objectives, the most important of which are lowering response times while improving resource utilisation and maintaining consistent load balances. The MFO approach searches the search space and provides early solutions, while the DNN algorithm refines and improves those first findings. In comprehensive simulations conducted in a real-world hospital setting, the hybrid MFO-DNN technique outperformed existing scheduling algorithms in terms of reaction time, resource utilisation, and load balancing. The simulated healthcare environments were as true to life as was feasible. The suggested technique has been demonstrated to be both dependable and scalable, making it appropriate for use in large-scale IOHT deployments. This study considerably enhances the state of the art in IOHT task scheduling in E healthcare systems by developing a hybrid optimisation technique that takes advantage of the strengths of both MFO and DNN. The findings indicate that this strategy has the potential to improve the quality and efficiency of healthcare delivery, which helps patients receive care that is both effective and timely.

Audio-Visual Activity Guided Cross-Modal Identity Association for Active Speaker Detection

Active speaker detection in videos addresses associating a source face, visible in the video frames, with the underlying speech in the audio modality. The two primary sources of information to derive such a speech-face relationship are i) visual activity and its interaction with the speech signal and ii) co-occurrences of speakers' identities across modalities in the form of face and speech. The two approaches have their limitations: the audio-visual activity models get confused with other frequently occurring vocal activities, such as laughing and chewing, while the speakers' identity-based methods are limited to videos having enough disambiguating information to establish a speech-face association. Since the two approaches are independent, we investigate their complementary nature in this work. We propose a novel unsupervised framework to guide the speakers' cross-modal identity association with the audio-visual activity for active speaker detection. Through experiments on entertainment media videos from two benchmark datasets–the AVA active speaker (movies) and Visual Person Clustering Dataset (TV shows)–we show that a simple late fusion of the two approaches enhances the active speaker detection performance.

Ability Grouping of Elderly Individuals Based on an Improved K‐Prototypes Algorithm

Elderly individuals with similar abilities are likely to have similar care needs. Abilities are the basis for analysing the need patterns of elderly individuals. Abilities of elderly individuals are both numerical and categorical; therefore, an improved K‐prototypes cluster algorithm for mixed attributes was proposed to group elderly individuals based on abilities. In this algorithm, the dissimilarity measure of categorical attributes is designed to consider not only the difference between the object to be allocated and the prototype of the cluster but also the differ6ence between the object to be allocated and other objects in the cluster and can measure the difference in sequential values of ordinal categorical attributes. Experimental results show that the improved K‐prototypes algorithm performs well for clustering datasets containing mixed attributes. Taking the CHARLS dataset as an example, distinct groups of elderly individuals based on the improved K‐prototypes algorithm showed significant ability differences.

Overcoming weaknesses of density peak clustering using a data-dependent similarity measure

Density Peak Clustering (DPC) is a popular state-of-the-art clustering algorithm, which requires pairwise (dis)similarity of data objects to detect arbitrary shaped clusters. While it is shown to perform well for many applications, DPC remains: (i) not robust for datasets with clusters having different densities, and (ii) sensitive to the change in the units/scales used to represent data. These drawbacks are mainly due to the use of the data-independent similarity measure based on the Euclidean distance. In this paper, we address these issues by proposing an effective data-dependent similarity measure based on Probability Mass, which we call MP-Similarity, and by incorporating it in DPC to create MP-DPC, a data-dependent variant of DPC. We evaluate and compare MP-DPC against diverse baselines using several clustering metrics and datasets. Our experiments demonstrate that: (a) MP-DPC produces better clustering results than DPC using the Euclidean distance and existing data-dependent similarity measures; (b) MP-Similarity coupled with Shared-Nearest-Neighbor-based density metric in DPC further enhances the quality of clustering results; and (c) unlike DPC with existing data-independent and data-dependent similarity measures, MP-DPC is robust to the change in the units/scales used to represent data. Our findings suggest that MP-Similarity provides a more viable solution for DPC in datasets with unknown distribution or units/scales of features, which is often the case in many real-world applications.

Robust Testing of Paired Outcomes Incorporating Covariate Effects in Clustered Data with Informative Cluster Size

Paired outcomes are common in correlated clustered data where the main aim is to compare the distributions of the outcomes in a pair. In such clustered paired data, informative cluster sizes can occur when the number of pairs in a cluster (i.e., a cluster size) is correlated to the paired outcomes or the paired differences. There have been some attempts to develop robust rank-based tests for comparing paired outcomes in such complex clustered data. Most of these existing rank tests developed for paired outcomes in clustered data compare the marginal distributions in a pair and ignore any covariate effect on the outcomes. However, when potentially important covariate data is available in observational studies, ignoring these covariate effects on the outcomes can result in a flawed inference. In this article, using rank based weighted estimating equations, we propose a robust procedure for covariate effect adjusted comparison of paired outcomes in a clustered data that can also address the issue of informative cluster size. Through simulated scenarios and real-life neuroimaging data, we demonstrate the importance of considering covariate effects during paired testing and robust performances of our proposed method in covariate adjusted paired comparisons in complex clustered data settings.

ADC-DL: Communication-efficient distributed learning with hierarchical clustering and adaptive dataset condensation

The rapid growth of modern mobile devices leads to a large number of distributed data, which is extremely valuable for learning models. Unfortunately, model training by collecting all these original data to a centralized cloud server is not applicable due to data privacy and communication costs concerns, hindering artificial intelligence from empowering mobile devices. Moreover, these data are not identically and independently distributed (Non-IID) caused by their different context, which will deteriorate the performance of the model. To address these issues, we propose a novel Distributed Learning algorithm based on hierarchical clustering and Adaptive Dataset Condensation, named ADC-DL, which learns a shared model by collecting the synthetic samples generated on each device. To tackle the heterogeneity of data distribution, we propose an entropy topsis comprehensive tiering model for hierarchical clustering, which distinguishes clients in terms of their data characteristics. Subsequently, synthetic dummy samples are generated based on the hierarchical structure utilizing adaptive dataset condensation. The procedure of dataset condensation can be adjusted adaptively according to the tier of the client. Extensive experiments demonstrate that the performance of our ADC-DL is more outstanding in prediction accuracy and communication costs compared with existing algorithms.

Efficiency rectification of predictive machine learning models by data centric optimization

There are two ways to optimize the performance of machine learning models, one is model centric, and another is data centric. Model centric approaches sometimes cannot yield expected results, so researchers and developers are now focusing on data centric optimization. Main steps of data centric approaches are – noise detection and correction, effective feature selection, feature transformation and data augmentation in case of limited datasets. Be it image, audio, text or numerical data, the mentioned approaches are always beneficial irrespective of model architecture. This paper is devoted to developing new preprocessing techniques for numerical data. The developed algorithm is tested with different artificial and real datasets for linear regression, k-means clustering, SVM classifier and deep learning networks. The algorithm is easy to implement and computationally less expensive. Experimental results show that proposed methods work well to improve model efficiency.