Graph-based Clustering Techniques Research Articles

Minimizing of Forecasting Error in Fuzzy Time Series Model Using Graph-Based Clustering Method

In recent years, numerous fuzzy time series (FTS) forecasting models have been developed to address complex and incomplete problems. However, the accuracy of these models is specific to the problem at hand and varies across datasets. Despite claims of superiority over traditional statistical and single machine learning-based models, achieving improved forecasting accuracy remains a formidable challenge. In FTS models, the lengths of intervals and fuzzy relationship groups are considered crucial factors influencing forecasting accuracy. Hence, this study introduces an FTS forecasting model based on the graph-based clustering technique. The clustering algorithm, utilized during the fuzzification stage, enables the derivation of unequal interval lengths. The proposed model is applied to forecast two numerical datasets: enrollment data from the University of Alabama and the datasets of Gas prices RON95 in Vietnam. Comparisons of forecasting results between the proposed model and others are conducted for enrollment forecasts at the University of Alabama. The findings reveal that the proposed model achieves higher forecasting accuracy across all orders of fuzzy relationships when compared to its counterparts

Generalized framework for image and video object segmentation using affinity learning and message passing GNNS

Despite significant amount of work reported in the computer vision literature, segmenting images or videos based on multiple cues such as objectness, texture and motion, is still a challenge. This is particularly true when the number of objects to be segmented is not known or there are objects that are not classified in the training data (unknown objects). A possible remedy to this problem is to utilize graph-based clustering techniques such as Correlation Clustering. It is known that using long range affinities (Lifted multicut), makes correlation clustering more accurate than using only adjacent affinities (Multicut). However, the former is computationally expensive and hard to use. In this paper, we introduce a new framework to perform image/motion segmentation using an affinity learning module and a Message Passing Graph Neural Network (MPGNN). The affinity learning module uses a permutation invariant affinity representation to overcome the multi-object problem. The paper shows, both theoretically and empirically, that the proposed MPGNN aggregates higher order information and thereby converts the Lifted Multicut Problem (LMP) to a Multicut Problem (MP), which is easier and faster to solve. Importantly, the proposed method can be generalized to deal with different clustering problems with the same MPGNN architecture. For instance, our method produces competitive results for single image segmentation (on BSDS dataset) as well as unsupervised video object segmentation (on DAVIS17 dataset), by only changing the feature extraction part. In addition, using an ablation study on the proposed MPGNN architecture, we show that the way we update the parameterized affinities directly contributes to the accuracy of the results.

A fast [formula omitted] time hybrid clustering algorithm using the circumference proximity based merging technique for diversified datasets

Clustering has been widely employed for extracting intrinsic groups because of its low reliance on domain knowledge. Though several clustering techniques have been developed in the literature, the majority of them become inefficient due to their dependence on user-defined parameters or inability to cluster multi-scale datasets. To handle these issues, several hybrid clustering algorithms which combine the advantages of partitional, hierarchical, and graph-based clustering techniques have been developed. Hybrid clustering algorithms first partition the data into smaller clusters and then merge them into genuine clusters. However, most of them incur a quadratic time complexity which limits their application for large datasets. The proposed work presents a two-step fast hybrid clustering algorithm based on partitions and efficient merging to reduce the computational cost by maintaining the clustering quality. In the first step, the dispersion of the data is considered to produce balanced partitions. Then graph circumference proximity-based merging technique is proposed to merge the sub-clusters. The overall computational complexity of the algorithm is O(NlgN) where N is the number of data points, ignoring the dimension. To the best of our knowledge, this is the fastest known graph neighborhood-based hybrid clustering algorithm. Experimental results on various diversified datasets exhibit a significant improvement in the running time as well as cluster quality and robustness against noise.

Establishing the Forecasting Model with Time Series Data Based on Graph and Particle Swarm Optimization

In recent years, a wide variety of fuzzy time series (FTS) forecasting models have been created and recommended to handle the complicated and ambiguous challenges relating to time series data from real-world sources. However, the accuracy of a model is problem-specific and varies across data sets. But a model's precision varies between different data sets and depends on the situation at hand. Even though many models assert that they are better than statistics and a single machine learning-based model, increasing forecasting accuracy is still a challenging task. In the fuzzy time series models, the size of the intervals and the fuzzy relationship groups are thought to be crucial variables that affect the model's forecasting abilities. This study offers a hybrid FTS forecasting model that makes use of both the graph-based clustering technique (GBC) and particle swarm optimization (PSO) for adjusting interval lengths in the universe of discourse (UoD). The suggested model's forecasting results have been compared to those provided by other current models on a dataset of enrollments at the University of Alabama. For all orders of fuzzy relationships, the suggested model outperforms its counterparts in terms of forecasting accuracy.

Establishing the Forecasting Model with Time Series Data Based on Graph and Particle Swarm Optimization

In recent years, a wide variety of fuzzy time series (FTS) forecasting models have been created and recommended to handle the complicated and ambiguous challenges relating to time series data from real-world sources. However, the accuracy of a model is problem-specific and varies across data sets. But a model’s precision varies between different data sets and depends on the situation at hand. Even though many models assert that they are better than statistics and a single machine learning-based model, increasing forecasting accuracy is still a challenging task. In the fuzzy time series models, the size of the intervals and the fuzzy relationship groups are thought to be crucial variables that affect the model’s forecasting abilities. This study offers a hybrid FTS forecasting model that makes use of both the graph-based clustering technique (GBC) and particle swarm optimization (PSO) for adjusting interval lengths in the universe of discourse (UoD). The suggested model’s forecasting results have been compared to those provided by other current models on a dataset of enrollments at the University of Alabama. For all orders of fuzzy relationships, the suggested model outperforms its counterparts in terms of forecasting accuracy.

Progressive Self-Supervised Clustering With Novel Category Discovery

These days, clustering is one of the most classical themes to analyze data structures in machine learning and pattern recognition. Recently, the anchor-based graph has been widely adopted to promote the clustering accuracy of plentiful graph-based clustering techniques. In order to achieve more satisfying clustering performance, we propose a novel clustering approach referred to as the progressive self-supervised clustering method with novel category discovery (PSSCNCD), which consists of three separate procedures specifically. First, we propose a new semisupervised framework with novel category discovery to guide label propagation processing, which is reinforced by the parameter-insensitive anchor-based graph obtained from balanced K -means and hierarchical K -means (BKHK). Second, we design a novel representative point selected strategy based on our semisupervised framework to discover each representative point and endow pseudolabel progressively, where every pseudolabel hypothetically corresponds to a real category in each self-supervised label propagation. Third, when sufficient representative points have been found, the labels of all samples will be finally predicted to obtain terminal clustering results. In addition, the experimental results on several toy examples and benchmark data sets comprehensively demonstrate that our method outperforms other clustering approaches.

Qualitative Regionalization of Bay of Bengal

Clustering is an essential statistical technique to find subgroups within a dataset. Clustering helps to uncover hidden relationships among the observational variables in the data. Many researchers have applied clustering techniques across a number of knowledge domains with success. Here we focus on application to ocean science, more particularly to the north region of the Indian Ocean, i.e., the Bay of Bengal. In this article, Regionalization is carried out for the Bay of Bengal to group the regions with similar characteristics using a graph-based clustering technique called SKATER (Spatial ‘K’luster Analysis by Tree Edge Removal). These subgroups are consistent with observations that characterize the near-surface features of the Bay of Bengal.

Regionalization for urban air mobility application in metropolitan areas: case studies in San Francisco and New York

In a new era of mobility where the transportation of persons or goods via flying vehicles over urban areas has garnered great interest in its application in urban space. With the anticipated utilization of sUAS in urban airspace, a multi-dimensional understanding of urban space is essential. As a first step to assess the feasibility of Urban Air Mobility (UAM) in urban areas, we conduct regionalization and correspondence analysis in highly urbanized areas – San Francisco, CA and Manhattan, NY – by incorporating population dataset and urban 3D airspace to delineate the regional boundaries. Regionalization is carried out using graph-based clustering technique called SKATER (Spatial ‘K’luster Analysis by Tree Edge Removal) to group the regions having similar characteristics and address the compound effect of both population and spatial information. By classifying the regions into five categories through correspondence analysis, the operational and economic feasibility of each region is evaluated. The results provide the region maps of each city with the most and least attractive regions for UAM application with the temporal notion, whether the clusters are daytime-intensive or nighttime-intensive areas. The outcomes have several unique information that can benefit drone delivery target area identification, landing location identification, demand prediction. Our approach can contribute to providing a useful basis for management for UAM in urban areas as well as the process of regulating airspace use.

On the Robustness of Graph-Based Clustering to Random Network Alterations

AbstractBiological functions emerge from complex and dynamic networks of protein–protein interactions. Because these protein–protein interaction networks, or interactomes, represent pairwise connections within a hierarchically organized system, it is often useful to identify higher-order associations embedded within them, such as multimember protein complexes. Graph-based clustering techniques are widely used to accomplish this goal, and dozens of field-specific and general clustering algorithms exist. However, interactomes can be prone to errors, especially when inferred from high-throughput biochemical assays. Therefore, robustness to network-level noise is an important criterion. Here, we tested the robustness of a range of graph-based clustering algorithms in the presence of noise, including algorithms common across domains and those specific to protein networks. Strikingly, we found that all of the clustering algorithms tested here markedly amplified network-level noise. Randomly rewiring only 1% of network edges yielded more than a 50% change in clustering results. Moreover, we found the impact of network noise on individual clusters was not uniform: some clusters were consistently robust to injected noise, whereas others were not. Therefore we developed the clust.perturb R package and Shiny web application to measure the reproducibility of clusters by randomly perturbing the network. We show that clust.perturb results are predictive of real-world cluster stability: poorly reproducible clusters as identified by clust.perturb are significantly less likely to be reclustered across experiments. We conclude that graph-based clustering amplifies noise in protein interaction networks, but quantifying the robustness of a cluster to network noise can separate stable protein complexes from spurious associations.

Regionalization for urban air mobility application with analyses of 3D urban space and geodemography in San Francisco and New York

Abstract In a new era of mobility where the transportation of persons or goods via flying vehicles over urban areas has garnered great interest. As a first step to assess the feasibility of Urban Air Mobility (UAM) in urban areas, we conduct regionalization and correspondence analysis in highly urbanized areas – San Francisco, CA and Manhattan, NY – by incorporating population dataset and urban 3D airspace to delineate the regional boundaries. Regionalization is carried out using graph-based clustering technique called SKATER (Spatial ‘K’luster Analysis by Tree Edge Removal) to group the regions having similar characteristics and address the compound effect of both population and spatial information. By classifying the regions into five categories through correspondence analysis, the operational and economic feasibility of each region is evaluated. The results provide the region maps of each city with the most and least attractive regions for UAM application with the temporal notion, whether the clusters are daytime-intensive or nighttime-intensive areas. Our approach can contribute to providing a useful basis for management for UAM in urban areas as well as the process of regulating airspace use.

First Results in Leak Localization in Water Distribution Networks using Graph-Based Clustering and Deep Learning

This paper presents a methodology for the localization of leaks in water distribution networks (WDNs) by means of the combination of a deep learning (DL) approach and a graph-based clustering technique. A data set for all possible leak locations is generated from pressure measurements and utilized to feed an image encoding process based on the Gramian Angular Field (GAF) technique, hence producing an equivalent data set of images. The pressure measurements are generated through the WDN simulation engine EPANET. To accomplish the training stage, the network is iteratively segmented into clusters using the Graph Agglomerative Clustering (GAC) method, and a deep learning neural network (DLNN) is trained to correctly indicate the leak location at one of the created clusters. The achieved neural networks tree can be traversed through its different branches depending on each classification result, until the final cluster is reached. Consequently, leaks can be located with a success rate that grows inversely to the size of the clusters. Due to the dependency of the latter on the number of clusters, which can be settled, the presented method is adaptable to the considered network features ( as e.g. dimensions, sensors placement and accuracy) and requisites (as e.g. localization area size).

Co-Regularized Discriminative Spectral Clustering With Adaptive Similarity Measure in Dual-Kernel Space

Spectral clustering is a very popular graph-based clustering technique that partitions data groups based on the input data similarity matrix. Many past studies based on spectral clustering, however, do not consider the global discriminative structure of the dataset. Also, the benefits of using more than one kernel have not been fully exploited with respect to spectral clustering, although it has been established by these past studies that using more than one kernel in clustering can result in a more accurate clustering than those obtained with a single kernel. Multi-kernel approaches, however, tend to be more time consuming compared to single kernel methods. To compensate these drawbacks, we integrate a global discriminative term into the clustering with an adaptive neighbor framework. This is done to preserve both the global geometric information and global discriminative information in a dual kernel space, in an attempt to optimize clustering performance. Via co-regularization, we utilize more than one kernel space to take advantage of the benefits of multiple kernels. We, however, use two heterogeneous kernels to help us reduce clustering time, since the ability to quickly process data is as equally important as its accuracy in this era of information explosion. Since these different kernel spaces admit the same underlying clustering of the data, we approach the problem looking for clustering consistent across the two kernel views. Hence we are able to detect the non-linear intrinsic geometrical information of the dataset. We perform clustering using the obtained indicator matrix from our modified Laplacian utilizing k-means. Our Experimental outcomes show that our approach gives satisfactory results in terms of accuracy and NMI, with time-to-cluster savings in comparison to other state-of-the-art clustering methods using both synthetic and public datasets.

Graph-based clustering of extracted paraphrases for labelling crime reports

Paraphrases are well-known as synonyms that express the same context in different articulations. Extracting paraphrases from a large text corpus is a challenging task in Natural Language Processing applications. The present work proposes a graph based clustering technique for discovering labels of crime reports based on extracted paraphrases from large untagged crime corpora. Initially, the entity pairs are represented as shallow parse trees where the headword in each tree reflects the actual meaning of the phrase between the entities. Though the phrases having similar headwords have been collected together, there exist many phrases between the entities that express similar context without sharing the same headword. Therefore, clustering is done to create a group of similar meaning phrases termed as paraphrases. A complete weighted graph is constructed with the phrases as nodes and cosine similarity between pair of phrases as the weight of an edge with the phrases as terminal nodes. The graph is made sparse by removing edges with weights less than a threshold value and clustering coefficient has been calculated for each node. The subgraph(s) comprising node(s) with the highest clustering coefficient has been extracted with their adjacent edges. The remaining nodes with their adjacent edges in the graph are added one at a time to an extracted subgraph, if and only if the average clustering coefficient of the resultant subgraph increases and an agglomerative merging technique is applied to merge the extracted subgraphs until no merging takes place. Finally, each subgraph represents a cluster of phrases, yields one aspect of crime. Based on the extracted paraphrases, the reports can be easily labelled. The proposed work deals with crime reports for United States of America (USA), United Arab Emirates (UAE) and India and the evaluation is performed in terms of various supervised and unsupervised techniques.

A Graph Based Clustering Approach for Relation Extraction From Crime Data

Application of natural language processing techniques based on crime data can prove to be beneficial in several processes of the criminal justice industry. The availability of massive crime reports helps law enforcement agencies when a criminal investigation is launched. While investigating a crime, questions like what type of crime, who committed the crime, what happened at which place, on what time, and what actions are taken, keep arising. Now, it is not feasible for the law enforcement agencies to get into the detail of these available massive crime reports and get the answers. To tackle these problems associated with criminal justice industry, the proposed work considers a textual corpus containing information of crime against women in India and extracts substantial relations between the named entities present in the corpus by a hierarchical graph-based clustering technique. For extracting the relations, different types of entity pairs have been chosen and similarities among them have been measured based on the intermediate context words. Depending on the similarity score, a weighted graph has been formed and a similarity threshold is set to partition the graph based on the edge weights. With the iterative application of the clustering algorithm, all the named entity pairs are grouped into clusters, each of which signifies different crime aspects. Each cluster is characterized using the most frequent context word present in it. The proposed relation extraction scheme helps in crime pattern analysis that can aid in various criminal investigation requirements. The results with optimal cluster validation indices depict the effectiveness of this method.

Brain state flexibility accompanies motor-skill acquisition

Learning requires the traversal of inherently distinct cognitive states to produce behavioral adaptation. Yet, tools to explicitly measure these states with non-invasive imaging – and to assess their dynamics during learning – remain limited. Here, we describe an approach based on a distinct application of graph theory in which points in time are represented by network nodes, and similarities in brain states between two different time points are represented as network edges. We use a graph-based clustering technique to identify clusters of time points representing canonical brain states, and to assess the manner in which the brain moves from one state to another as learning progresses. We observe the presence of two primary states characterized by either high activation in sensorimotor cortex or high activation in a frontal-subcortical system. Flexible switching among these primary states and other less common states becomes more frequent as learning progresses, and is inversely correlated with individual differences in learning rate. These results are consistent with the notion that the development of automaticity is associated with a greater freedom to use cognitive resources for other processes. Taken together, our work offers new insights into the constrained, low dimensional nature of brain dynamics characteristic of early learning, which give way to less constrained, high-dimensional dynamics in later learning.

A Novel Graph Based Fuzzy Clustering Technique For Unsupervised Classification Of Remote Sensing Images

Abstract. This paper addresses the problem of unsupervised land-cover classification of multi-spectral remotely sensed images in the context of self-learning by exploring different graph based clustering techniques hierarchically. The only assumption used here is that the number of land-cover classes is known a priori. Object based image analysis paradigm which processes a given image at different levels, has emerged as a popular alternative to the pixel based approaches for remote sensing image segmentation considering the high spatial resolution of the images. A graph based fuzzy clustering technique is proposed here to obtain a better merging of an initially oversegmented image in the spectral domain compared to conventional clustering techniques. Instead of using Euclidean distance measure, the cumulative graph edge weight is used to find the distance between a pair of points to better cope with the topology of the feature space. In order to handle uncertainty in assigning class labels to pixels, which is not always a crisp allocation for remote sensing data, fuzzy set theoretic technique is incorporated to the graph based clustering. Minimum Spanning Tree (MST) based clustering technique is used to over-segment the image at the first level. Furthermore, considering that the spectral signature of different land-cover classes may overlap significantly, a self-learning based Maximum Likelihood (ML) classifier coupled with the Expectation Maximization (EM) based iterative unsupervised parameter retraining scheme is used to generate the final land-cover classification map. Results on two medium resolution images establish the superior performance of the proposed technique in comparison to the traditional fuzzy c-means clustering technique.

An effective graph-based clustering technique to identify coherent patterns from gene expression data

This paper presents an effective parameter-less graph based clustering technique (GCEPD). GCEPD produces highly coherent clusters in terms of various cluster validity measures. The technique finds highly coherent patterns containing genes with high biological relevance. Experiments with real life datasets establish that the method produces clusters that are significantly better than other similar algorithms in terms of various quality measures.