Soft Clustering Approach Research Articles

Inferring ancestry with the hierarchical soft clustering approach tangleGen.

Understanding the genetic ancestry of populations is central to numerous scientific and societal fields. It contributes to a better understanding of human evolutionary history, advances personalized medicine, aids in forensic identification, and allows individuals to connect to their genealogical roots. Existing methods, such as ADMIXTURE, have significantly improved our ability to infer ancestries. However, these methods typically work with a fixed number of independent ancestral populations. As a result, they provide insight into genetic admixture, but do not include a hierarchical interpretation. In particular, the intricate ancestral population structures remain difficult to unravel. Alternative methods with a consistent inheritance structure, such as hierarchical clustering, may offer benefits in terms of interpreting the inferred ancestries. Here, we present tangleGen, a soft clustering tool that transfers the hierarchical machine learning framework Tangles, which leverages graph theoretical concepts, to the field of population genetics. The hierarchical perspective of tangleGen on the composition and structure of populations improves the interpretability of the inferred ancestral relationships. Moreover, tangleGen adds a new layer of explainability, as it allows identifying the SNPs that are responsible for the clustering structure. We demonstrate the capabilities and benefits of tangleGen for the inference of ancestral relationships, using both simulated data and data from the 1000 Genomes Project.

Clustering and Interpretation of time-series trajectories of chronic pain using evidential c-means

The most well-known unsupervised classification algorithms allow for the identification of hard or probabilistic partitions. However, when working with complex datasets such as those found in the healthcare domain, it is important to note that these partitions may have limitations, particularly when considering the uncertainty associated with medical data. Indeed, these algorithms do not account for modeling outliers, imprecise, or uncertain observations. In this study, we focused on analyzing time series of barometers attributes related to pain intensity in individuals with chronic pain. Our goal was to identify different typologies of care trajectories, highlighting distinct trajectories of chronic pain through clustering techniques. We propose a soft clustering approach based on feature extraction and selection from sequential data, aiming to improve interpretability and enhance the performance of the clustering procedure. Time series feature extraction and selection effectively handles complex and noisy data, improving interpretability and clustering performance. Evidential soft clustering models data uncertainty and imprecision, providing a better representation of nuances and variations in complex, raw data. The first step involves extracting features from the available time series and selecting the most important attributes. For this purpose, we use a time series feature extractor (Tsfresh) and unsupervised feature selection methods, including unsupervised Random Forest, Laplacian Score, and unsupervised Spectral Feature Selection. The second step involves using the evidential c-means (ECM) clustering algorithm on the extracted attributes. The ECM method, based on belief functions, allows for generating a credal partition that has the ability to model various forms of uncertainty. This partition can then be transformed into a hard partition to study individuals based on this uncertainty criterion. The results reveal the existence of two clusters of chronic pain related to discomfort and well-being, exhibiting excellent separability and compactness. Additionally, an uncertain cluster emerges grouping patients with intermediate characteristics. The interpretability of the determined partitions through descriptive analysis, statistical tests, and repeated measures multinomial regression on clinical and demographic data allowed us to determine the profile of patients in the identified trajectories.

Cluster-based Graph Collaborative Filtering

Graph Convolution Networks (GCNs) have significantly succeeded in learning user and item representations for recommendation systems. The core of their efficacy is the ability to explicitly exploit the collaborative signals from both the first- and high-order neighboring nodes. However, most existing GCN-based methods overlook the multiple interests of users while performing high-order graph convolution. Thus, the noisy information from unreliable neighbor nodes ( \(e.g.\) , users with dissimilar interests) negatively impacts the representation learning of the target node. Additionally, conducting graph convolution operations without differentiating high-order neighbors suffers the over-smoothing issue when stacking more layers, resulting in performance degradation. In this paper, we aim to capture more valuable information from high-order neighboring nodes while avoiding noise for better representation learning of the target node. To achieve this goal, we propose a novel GCN-based recommendation model, termed Cluster-based Graph Collaborative Filtering (ClusterGCF). This model performs high-order graph convolution on cluster-specific graphs, which are constructed by capturing the multiple interests of users and identifying the common interests among them. Specifically, we design an unsupervised and optimizable soft node clustering approach to classify user and item nodes into multiple clusters. Based on the soft node clustering results and the topology of the user-item interaction graph, we assign the nodes with probabilities for different clusters to construct the cluster-specific graphs. To evaluate the effectiveness of ClusterGCF, we conducted extensive experiments on four publicly available datasets. Experimental results demonstrate that our model can significantly improve recommendation performance.

Clustering Longitudinal Data for Growth Curve Modelling by Gibbs Sampler and Information Criterion

Clustering longitudinal data for growth curve modelling is considered in this paper, where we aim to optimally estimate the underpinning unknown group partition matrix. Instead of following the conventional soft clustering approach, which assumes the columns of the partition matrix to have i.i.d. multinomial or categorical prior distributions and uses a regression model with the response following a finite mixture distribution to estimate the posterior distribution of the partition matrix, we propose an iterative partition and regression procedure to find the best partition matrix and the associated best growth curve regression model for each identified cluster. We show that the best partition matrix is the one minimizing a recently developed empirical Bayes information criterion (eBIC), which, due to the involved combinatorial explosion, is difficult to compute via enumerating all candidate partition matrices. Thus, we develop a Gibbs sampling method to generate a Markov chain of candidate partition matrices that has its equilibrium probability distribution equal the one induced from eBIC. We further show that the best partition matrix, given a priori the number of latent clusters, can be consistently estimated and is computationally scalable based on this Markov chain. The number of latent clusters is also best estimated by minimizing eBIC. The proposed iterative clustering and regression method is assessed by a comprehensive simulation study before being applied to two real-world growth curve modelling examples involving longitudinal data clustering.

An entropy-based membership approach on type-II fuzzy set (EMT2FCM) for biomedical image segmentation

Medical image segmentation is a challenging task owing to its aliasing artifacts, existence of mixed pixels, interference of noise and blurring effect, etc. The brain and the spinal cord together constitute the Central Nervous System (CNS), with the brain being the most complex component. Brain segmentation is crucial for surgical planning, disease detection, and therapeutic interventions. A unique entropy based soft clustering approach (EMT2FCM) for the brain MRI imaging on both real and simulated datasets is proposed in this paper. This paper presents a refined version of the type-2 Fuzzy C-Means (FCM) technique (EMT2FCM) for isolating diverse tissues in Magnetic Resonance (MR) brain images, achieved through an enhanced entropy-based membership function. Difficulty in edge detection occurs due to the vagueness caused by mixed pixels. To overcome the problem with mixed pixels a new membership function proposal has been given. EMT2FCM, i.e., fuzzy membership function with Shannon's entropy-based variation, is able to handle uncertainty and vagueness in images offered by mixed pixels, by implementing new distance criteria modification for the calculation of optimum cluster. The approach has been validated using datasets from BRAINWEB (simulated) and IBSR (real-life). Following the deployment of our new algorithm, we assessed its performance using six observation criteria. The results demonstrated significant improvements in clustering performance, as evidenced by enhanced Mean Squared Error (MSE), Dice Index, and Jaccard Index, Accuracy, Sensitivity and Partition Indices validation measures. The evaluation of EMT2FCM in comparison to other state-of-the-art methods is established to reduce segmentation errors significantly.

Withdrawal: He, Y., Zhang, J., Bio‐Inspired Algorithms For Industrial Robot Control Using A Deep Learning Method. Agronomy Journal. 2022; Accepted Articles. https://doi.org/10.1002/agj2.21214.

AbstractBio‐inspired computing is a research method designed to address problems through computational methods based on biology and the natural environment. Bio‐inspired computing often grows and expands on a tiny foundation of certain laws through unattained deep learning. Importantly there is a dramatic increase in the research about bio‐inspired algorithms and techniques, from machine learning algorithms to other techniques based on values or modelling techniques of biological processes. The challenges of presenting and implementing complex strategies in practice generate new and innovative solutions in typically complicated areas. The areas and problems in which machine learning normally manages enable everyone to explore new challenges in bio‐inspired calculation methodologies. To manage these kinds of complexity, the implementation of bio‐inspired algorithms using the deep soft clustering (BiADSC) approach has been proposed to analyze complicated dynamics demands for human intelligence, which will take time and create a trade‐off between the dimension of the state and controllability. Deep novel framework focus on classification tasks through a combination of domain‐based extracting features with revolutionary neural networks (RNN) method. Furthermore, one of the deep learning models includes high‐performance neural networks using soft clustering algorithms. The clustering algorithm provides the data of basic units by local action that can communicate either direct or indirect way locally. Experimental results show that the proposed architecture enables in‐time data access and can successfully serve a range of simultaneous applications with a performance of 98.83%.This article is protected by copyright. All rights reserved

Impact of cultural and genetic structure on food choices along the Silk Road

The complex interplay between genetics, culture, and environment forms an individual's biology, influencing their behavior, choices, and health. However, to what extent information derived from this intertwined network could be quantitatively summarized to provide a glance at an individual's lifestyle is difficult to say. Here, we focused on dietary preferences as cultural proxies and genome-wide data of 543 individuals from six historical Silk Road countries: Georgia, Armenia, Azerbaijan, Uzbekistan, Kazakhstan, and Tajikistan. These lands favored the dispersal of innovations, foods, and DNA halfway across Eurasia, thus representing an ideal subject to explore interactions of cultural factors and genetic ancestry. We used discriminant analysis of principal components to infer cultural clusters, where mixed memberships are allowed. Five different clusters emerged. Of these, clusters 1 and 3, driven by aversion to pork and alcoholic beverages, mirrored genetic admixture patterns with the exception of Azerbaijan, which shares preferences supported by Islamic culture with Eastern countries. Cluster 3 was driven by protein-rich foods, whose preference was significantly related to steppe pastoralist ancestry. Sex and age were secondary clustering factors, with clusters formed by male and young individuals being related to alcohol preference and a reduced liking for vegetables. The soft clustering approach enabled us to model and summarize the individual's dietary information in short and informative vectors, which show meaningful interaction with other nondietary attributes of the studied individuals. Encoding other cultural variables would help summarize an individual's culture quantitatively, thus ultimately supporting its inclusion as a covariate in future association studies.

A privacy preserving diagnostic collaboration framework for facial paralysis using federated learning

Most of the machine learning and artificial intelligence applications are data driven. When it comes to sensitive data, maintaining the data privacy principles is a big challenge. Building a machine learning model for classifying sensitive data is discussed in this paper. Focus is given for medical field where the patient data comes under sensitive or private information category. There are restrictions to share patient data for research purposes or collaboration among doctors in different hospitals due to the privacy concerns. In this work, we take facial paralysis as an example and discuss how to build a model for facial paralysis detection. Here, the data used for training the model is face images which implicitly reveals identity of the patient. We analyse how the facial paralysis images from multiple hospitals can be combined together for building an efficient facial paralysis detection system without compromising privacy of patients. Support Vector Machine based federated learning is applied for the purpose. Hospitals are considered as clients which are the data sources where the local training happens and there is a server performing federated averaging. Unlike in traditional federated learning, soft clustering approach is considered at server side and the update to each client is different. The federated averaging algorithm at server takes care of the distribution of data each client holds and customises the update sent to each client. This approach improves the local test accuracy and the convergence speed. To validate the findings, experiments are conducted with MNIST and covid pneumonia datasets as well.

Corpus Statistics Empowered Document Classification

In natural language processing (NLP), document classification is an important task that relies on the proper thematic representation of the documents. Gaussian mixture-based clustering is widespread for capturing rich thematic semantics but ignores emphasizing potential terms in the corpus. Moreover, the soft clustering approach causes long-tail noise by putting every word into every cluster, which affects the natural thematic representation of documents and their proper classification. It is more challenging to capture semantic insights when dealing with short-length documents where word co-occurrence information is limited. In this context, for long texts, we proposed Weighted Sparse Document Vector (WSDV), which performs clustering on the weighted data that emphasizes vital terms and moderates the soft clustering by removing outliers from the converged clusters. Besides the removal of outliers, WSDV utilizes corpus statistics in different steps for the vectorial representation of the document. For short texts, we proposed Weighted Compact Document Vector (WCDV), which captures better semantic insights in building document vectors by emphasizing potential terms and capturing uncertainty information while measuring the affinity between distributions of words. Using available corpus statistics, WCDV sufficiently handles the data sparsity of short texts without depending on external knowledge sources. To evaluate the proposed models, we performed a multiclass document classification using standard performance measures (precision, recall, f1-score, and accuracy) on three long- and two short-text benchmark datasets that outperform some state-of-the-art models. The experimental results demonstrate that in the long-text classification, WSDV reached 97.83% accuracy on the AgNews dataset, 86.05% accuracy on the 20Newsgroup dataset, and 98.67% accuracy on the R8 dataset. In the short-text classification, WCDV reached 72.7% accuracy on the SearchSnippets dataset and 89.4% accuracy on the Twitter dataset.

137-OR: A Genetic Approach to Detangling the Heterogeneity of Fasting Insulin Levels

Fasting insulin (FI) is a heterogeneous trait with high variability between individuals and populations. Increased FI is associated with an increased risk of type 2 diabetes (T2D) and its complications, including coronary artery disease (CAD) and hypertension (HTN) . We hypothesize that we could use a genetic approach to cluster FI-associated variants, characterize the clusters in terms of insulin processes within the Mexican American Cardiovascular Disease Study (MACAD; N=821) , and test for association of polygenic scores (PS) created from these clusters with T2D, CAD and HTN in three cohorts: Mass General Brigham Biobank (MGBB; N=18,127) , Aspirin in Reducing Events in the Elderly (ASPREE; N=13,349) , and Framingham Heart Study (FHS; N=7,145) . We identified six clusters using a soft clustering approach with 446 FI-increasing genetic alleles and 27 FI-related traits. Associations in two of the clusters, allowed us to characterize them and show their impact on T2D and T2D-associated complications. The first cluster had alleles associated with increased subcutaneous and visceral adiposity and was characterized by reduced insulin sensitivity (from euglycemic clamp; P=0.02) , increased insulin secretion (from oral glucose tolerance test [OGTT]; P=0.04) , and increased insulin clearance (from euglycemic clamp; P=0.02) in MACAD. This cluster’s PS was associated with HTN (MGBB Odds ratio (OR) = 1.04, P=0.013) , T2D (MGBB OR = 1.04, P =0.001) and CAD (MGB OR=1.07, P =0.008; ASPREE OR=1.04, P =0.036) The second cluster had alleles associated with a lipodystrophy phenotype and was characterized by increased insulin secretion (from OGTT P =0.019) in MACAD. We observed associations of the PS with T2D (MGBB OR= 1.13, P =2.3x10-10) and HTN (MGBB OR = 1.06, P =0.001) which were replicated in ASPREE and FHS. Our results suggest the presence of two different FI-associated genetic domains and elucidate the deleterious mechanisms by which increased FI is associated with T2D and T2D complications. Disclosure M.Sevilla: None. M.O.Goodarzi: None. D.A.Dicorpo: None. J.I.Rotter: None. J.B.Meigs: Consultant; Quest Diagnostics. M.Udler: None. A.Manning: None. Funding UM1DK078616

The Genetics of Type 2 Diabetes in Youth: Where We Are and the Road Ahead

The Genetics of Type 2 Diabetes in Youth: Where We Are and the Road Ahead

A Novel Soft Clustering Approach for Gene Expression Data

Gene expression data represents a condition matrix where each row represents the gene and the column shows the condition. Micro array used to detect gene expression in lab for thousands of gene at a time. Genes encode proteins which in turn will dictate the cell function. The production of messenger RNA along with processing the same are the two main stages involved in the process of gene expression. The biological networks complexity added with the volume of data containing imprecision and outliers increases the challenges in dealing with them. Clustering methods are hence essential to identify the patterns present in massive gene data. Many techniques involve hierarchical, partitioning, grid based, density based, model based and soft clustering approaches for dealing with the gene expression data. Understanding the gene regulation and other useful information from this data can be possible only through effective clustering algorithms. Though many methods are discussed in the literature, we concentrate on providing a soft clustering approach for analyzing the gene expression data. The population elements are grouped based on the fuzziness principle and a degree of membership is assigned to all the elements. An improved Fuzzy clustering by Local Approximation of Memberships (FLAME) is proposed in this work which overcomes the limitations of the other approaches while dealing with the non-linear relationships and provide better segregation of biological functions.

What the Language You Tweet Says About Your Occupation

Many aspects of people's lives are proven to be deeply connected to their jobs. In this paper, we first investigate the distinct characteristics of major occupation categories based on tweets. From multiple social media platforms, we gather several types of user information. From users' LinkedIn webpages, we learn their proficiencies. To overcome the ambiguity of self-reported information, a soft clustering approach is applied to extract occupations from crowd-sourced data. Eight job categories are extracted, including Marketing, Administrator, Start-up, Editor, Software Engineer, Public Relation, Office Clerk, and Designer. Meanwhile, users' posts on Twitter provide cues for understanding their linguistic styles, interests, and personalities. Our results suggest that people of different jobs have unique tendencies in certain language styles and interests. Our results also clearly reveal distinctive levels in terms of Big Five Traits for different jobs. Finally, a classifier is built to predict job types based on the features extracted from tweets. A high accuracy indicates a strong discrimination power of language features for job prediction task.

Sentiment Analysis based on Soft Clustering through Dimensionality Reduction Technique

Clustering based sentiment analysis confers new directions to analyze real-world opinions without human participation and pre-tagged training data overhead. Clustering based techniques do not rely on linguistic information and more convenient as compared to other traditional machine learning techniques. Combining the dimensionality reduction techniques with clustering algorithms highly influence the computational cost and improve the performance of sentiment analysis. In this research, we applied Principal Component Analysis technique to reduce the size of features set. This reduced feature set improves binary K-means clustering results of sentiments analysis. In our experiments, we demonstrate the performance of the clustering system with a reduced feature set to provide high-quality sentiment analysis. However, K-mean clustering has its own limitations such as hard assignment and instability of results. To overcome the limitation of traditional K-means algorithm we applied soft clustering (Expectation maximization algorithm) approach which stabilizes clustering accuracy. This approach allows a soft assignment to cluster documents. Consequently, our experimental accuracy is 95% with standard deviation rate of 0.1% which is sufficient to apply the clustering technique in real-world applications.

A Soft Clustering Approach to Detect Socio-Ecological Landscape Boundaries Using Bayesian Networks

Detecting socio-ecological boundaries in traditional rural landscapes is very important for the planning and sustainability of these landscapes. Most of the traditional methods to detect ecological boundaries have two major shortcomings: they are unable to include uncertainty, and they often exclude socio-economic information. This paper presents a new approach, based on unsupervised Bayesian network classifiers, to find spatial clusters and their boundaries in socio-ecological systems. As a case study, a Mediterranean cultural landscape was used. As a result, six socio-ecological sectors, following both longitudinal and altitudinal gradients, were identified. In addition, different socio-ecological boundaries were detected using a probability threshold. Thanks to its probabilistic nature, the proposed method allows experts and stakeholders to distinguish between different levels of uncertainty in landscape management. The inherent complexity and heterogeneity of the natural landscape is easily handled by Bayesian networks. Moreover, variables from different sources and characteristics can be simultaneously included. These features confer an advantage over other traditional techniques.

Flexible Assignment of Loading Bays for Efficient Vehicle Routing in Urban Last Mile Delivery

Urban freight deliveries are often subject to many access restrictions which creates the need to establish a system of loading bays and to split the last mile delivery into driving and walking parts. A new model based on hard and soft clustering approach is developed to solve the loading bay assignment problem for efficient vehicle routing and walking in last mile delivery. The flexibility of the model is provided by the soft clustering approach based on different membership degrees of customers to loading bays. Especially for instances with large numbers of loading bays, soft clustering seems to give better results, it leads to higher flexibility of city logistics systems, minimal driving distances, and adequately short walking paths, which contribute to the goal of reaching sustainable urban freight deliveries.

Social image mining for fashion analysis and forecasting

Fashion industries need to be attentive towards the changing fashion and its upcoming market demands to grow their business, optimally. This paper describes research work involved in image mining for fashion analysis and forecasting using fashion-related images collected from the social network. A novel soft clustering technique is proposed for grouping the social fashion images. This technique is robust against uncertainty found in given images. The proposed clustering approach is compared with existing soft clustering approaches. It is found that the proposed approach performs well. Attributes of fashion items found in each cluster are analyzed through correlation, causal analysis, and fashion cycle visualization. Predictive models are applied to the clustered fashion items for style forecasting. A comparative study of predictive models is also done to find an optimal technique for various fashion items. As social visual perception is helpful for decision making, the proposed system is very useful in fashion industries to uplift their business.

Analysis of Purchase History Data Based on a New Latent Class Model for RFM Analysis

Recently, it has become easier to make use of various kinds of information on customers (e.g. customers’ purchase history), due to the development of information technology. Especially in the marketing field, in fact, many companies try to employ customer segmentation for the services customization which leads to increase customer loyalty and to keep high customer retention. One of the well-known approaches for the customer analysis based on purchase history data is the RFM analysis. The RFM analysis is usually used to segment customers into several groups by using three variables; how long it has been since their last purchase, how many times they purchased, and how much they spent. However, the conventional method of the RFM analysis did not assume a generative model. Therefore, when applying to an actual data set and scoring each index of R, F, M scores, several problems occur. The main problem is that an analyst should arbitrarily decide the threshold for the scores of RFM. On the other hand, in the field of machine learning, the probabilistic latent semantic analysis is widely used for soft clustering. The latent class model enables us to cluster customers into latent classes and to calculate the assignment probabilities of each customer to each latent class. In this paper, we propose a new latent class model for the RFM analysis based on the purchase history data. The proposed model enables to decide the scoring of RFM and segment customers automatically, and the soft clustering approach helps the interpretation of the result. Furthermore, the proposed model takes account of the generation model of RFM scores. From the result of actual data analysis, it became clear that it is possible to extract latent classes that express the statistical characteristics of data well. Given a generative model estimated from the given data, it is also possible to predict future purchase behaviors of customers or to generate virtual data for simulation analysis and make decisions based on the result. We verify the effectiveness of our model by analyzing a real purchase history data of a Japanese major retail company.

A New Recommendation Approach Based on Probabilistic Soft Clustering Methods: A Scientific Documentation Case Study

Recommender system (RS) clustering is an important issue, both for the improvement of the collaborative filtering (CF) accuracy and to obtain analytical information from their high sparse datasets. RS items and users usually share features belonging to different clusters, e.g., a musical-comedy movie. Soft clustering, therefore, is the CF clustering’s most natural approach. In this paper, we propose a new prediction approach for probabilistic soft clustering methods. In addition, we put to test a not traditional scientific documentation CF dataset: SD4AI, and we compare results with the MovieLens baseline. Not traditional CF datasets have challenging features, such as not regular rating frequency distributions, broad range of rating values, and a particularly high sparsity. The results show the suitability of using soft-clustering approaches, where their probabilistic overlapping parameters find optimum values when balanced hard/soft clustering is used. This paper opens some promising lines of research, such as RSs’ use in the scientific documentation field, the Internet of Things-based datasets processing, and design of new model-based soft clustering methods.

Meta-analysis of cell- specific transcriptomic data using fuzzy c-means clustering discovers versatile viral responsive genes

BackgroundDespite advances in the gene-set enrichment analysis methods; inadequate definitions of gene-sets cause a major limitation in the discovery of novel biological processes from the transcriptomic datasets. Typically, gene-sets are obtained from publicly available pathway databases, which contain generalized definitions frequently derived by manual curation. Recently unsupervised clustering algorithms have been proposed to identify gene-sets from transcriptomics datasets deposited in public domain. These data-driven definitions of the gene-sets can be context-specific revealing novel biological mechanisms. However, the previously proposed algorithms for identification of data-driven gene-sets are based on hard clustering which do not allow overlap across clusters, a characteristic that is predominantly observed across biological pathways.ResultsWe developed a pipeline using fuzzy-C-means (FCM) soft clustering approach to identify gene-sets which recapitulates topological characteristics of biological pathways. Specifically, we apply our pipeline to derive gene-sets from transcriptomic data measuring response of monocyte derived dendritic cells and A549 epithelial cells to influenza infections. Our approach apply Ward’s method for the selection of initial conditions, optimize parameters of FCM algorithm for human cell-specific transcriptomic data and identify robust gene-sets along with versatile viral responsive genes.ConclusionWe validate our gene-sets and demonstrate that by identifying genes associated with multiple gene-sets, FCM clustering algorithm significantly improves interpretation of transcriptomic data facilitating investigation of novel biological processes by leveraging on transcriptomic data available in the public domain. We develop an interactive ‘Fuzzy Inference of Gene-sets (FIGS)’ package (GitHub: https://github.com/Thakar-Lab/FIGS) to facilitate use of of pipeline. Future extension of FIGS across different immune cell-types will improve mechanistic investigation followed by high-throughput omics studies.