Incremental Cluster Research Articles

Machine learning cluster analysis identifies increased 12-month mortality risk in transcatheter aortic valve replacement recipients

BackgroundLong-term mortality risk is seldom re-assessed in contemporary clinical practice following successful transcatheter aortic valve implantation (TAVR). Unsupervised machine learning permits pattern discovery within complex multidimensional patient data and may facilitate recognition of groups requiring closer post-TAVR surveillance.MethodsWe analysed and differentiated routinely collected demographic, biochemical, and cardiac imaging data into distinct clusters using unsupervised machine learning. k-means clustering was performed on data from 200 patients who underwent TAVR for severe aortic stenosis (AS). Input features were ranked according to their influence on cluster assignment. Survival analyses were performed with Kaplan–Meier and Cox proportional hazards models. Nested cox models were used to identify any incremental prognostic benefit cluster assignment achieved beyond conventional risk scores.ResultsAnalysis identified two distinct clusters. Compared to Cluster 1, Cluster 2 demonstrated significantly worse all-cause mortality at 12 months (HR 6.3, p < 0.01), and was characterised by more advanced cardiac remodelling with worse indices of multi-chamber cardiac function, as quantified by strain imaging. Cluster assignment demonstrated greater predictive power for 12-month mortality as compared with conventional risk and frailty calculators.Conclusionk-means clustering identified two prognostically distinct phenogroups of patients who had undergone TAVR with better discriminatory power than conventional risk and frailty calculators. Our results highlight the utility of machine learning applications for clinical risk prediction and scope to improve patient surveillance.

ICVI-ARTMAP: Using Incremental Cluster Validity Indices and Adaptive Resonance Theory Reset Mechanism to Accelerate Validation and Achieve Multiprototype Unsupervised Representations.

This article presents an adaptive resonance theory predictive mapping (ARTMAP) model, which uses incremental cluster validity indices (iCVIs) to perform unsupervised learning, namely, iCVI-ARTMAP. Incorporating iCVIs to the decision-making and many-to-one mapping capabilities of this adaptive resonance theory (ART)-based model can improve the choices of clusters to which samples are incrementally assigned. These improvements are accomplished by intelligently performing the operations of swapping sample assignments between clusters, splitting and merging clusters, and caching the values of variables when iCVI values need to be recomputed. Using recursive formulations enables iCVI-ARTMAP to considerably reduce the computational burden associated with cluster validity index (CVI)-based offline clustering. In this work, six iCVI-ARTMAP variants were realized via the integration of one information-theoretic and five sum-of-squares-based iCVIs into fuzzy ARTMAP. With proper choice of iCVI, iCVI-ARTMAP either outperformed or performed comparably to three ART-based and four non-ART-based clustering algorithms in experiments using benchmark datasets of different natures. Naturally, the performance of iCVI-ARTMAP is subject to the selected iCVI and its suitability to the data at hand; fortunately, it is a general model in which other iCVIs can be easily embedded.

Incremental Cluster Validity Index-Guided Online Learning for Performance and Robustness to Presentation Order.

In streaming data applications, the incoming samples are processed and discarded, and therefore, intelligent decision-making is crucial for the performance of lifelong learning systems. In addition, the order in which the samples arrive may heavily affect the performance of incremental learners. The recently introduced incremental cluster validity indices (iCVIs) provide valuable aid in addressing such class of problems. Their primary use case has been cluster quality monitoring; nonetheless, they have been recently integrated in a streaming clustering method. In this context, the work presented, here, introduces the first adaptive resonance theory (ART)-based model that uses iCVIs for unsupervised and semi-supervised online learning. Moreover, it shows how to use iCVIs to regulate ART vigilance via an iCVI-based match tracking mechanism. The model achieves improved accuracy and robustness to ordering effects by integrating an online iCVI module as module B of a topological ART predictive mapping (TopoARTMAP)-thereby being named iCVI-TopoARTMAP-and using iCVI-driven postprocessing heuristics at the end of each learning step. The online iCVI module provides assignments of input samples to clusters at each iteration in accordance to any of the several iCVIs. The iCVI-TopoARTMAP maintains useful properties shared by the ART predictive mapping (ARTMAP) models, such as stability, immunity to catastrophic forgetting, and the many-to-one mapping capability via the map field module. The performance and robustness to the presentation order of iCVI-TopoARTMAP were evaluated via experiments with synthetic and real-world datasets.

A Fully Online Clustering Approach for Enhanced Performance of Health Information System

Health Data clustering is a significant research direction aiming to extract knowledge from a continuous health data flow to support online health decisions. However, processing health data clusters is still a challenging task. Existing clustering approaches are subject to various limitations in terms of considering the neighbor clusters and conducting multiple operations during the maintenance process. In this paper, we model, design and implement a novel framework called IClustMaint for efficiently clustering and maintaining health data clusters incrementally. A two-phase algorithm is embedded in the framework. We first employ the Principal Component Analysis (PCA) method to efficiently reduce the high costs of the initial clustering phase. Next, in the maintenance phase, we propose the incremental Cluster maintenance (ICM) approach for managing the generated cluster during a period of time. Technically, when the data clusters are evolving over time and need to be maintained frequently, the ICM approach improves the performance of cluster maintenance by only tracking the edge points. The experimental results on a real medical dataset verify the efficiency of the proposed approaches.

Multi-input address incremental clustering for the Bitcoin blockchain based on Petri net model analysis

Bitcoin is a cryptocurrency based on blockchain. All historical Bitcoin transactions are stored in the Bitcoin blockchain, but Bitcoin owners are generally unknown. This is the reason for Bitcoin's pseudo-anonymity, therefore it is often used for illegal transactions. Bitcoin addresses are related to Bitcoin users' identities. Some Bitcoin addresses have the potential to be analyzed due to the behavior patterns of Bitcoin transactions. However, existing Bitcoin analysis methods do not consider the fusion of new blocks' data, resulting in low efficiency of Bitcoin address analysis. In order to address this problem, this paper proposes an incremental Bitcoin address cluster method to avoid re-clustering when new block data is added. Besides, a heuristic Bitcoin address clustering algorithm is developed to improve clustering accuracy for the Bitcoin Blockchain. Experimental results show that the proposed method increases Bitcoin address cluster efficiency and accuracy.

Cluster Expansion Method for Critical Node Problem Based on Contraction Mechanism in Sparse Graphs

The objective of critical nodes problem is to minimize pair-wise connectivity as a result of removing a specific number of nodes in the residual graph. From a mathematical modeling perspective, it comes the truth that the more the number of fragmented components and the evenly distributed of disconnected sub-graphs, the better the quality of the solution. Basing on this conclusion, we proposed a new Cluster Expansion Method for Critical Node Problem (CEMCNP), which on the one hand exploits a contraction mechanism to greedy simplify the complexity of sparse graph model, and on the other hand adopts an incremental cluster expansion approach in order to maintain the size of formed component within reasonable limitation. The proposed algorithm also relies heavily on the idea of multi-start iterative local search algorithm, whereas brings in a diversified late acceptance local search strategy to keep the balance between interleaving diversification and intensification in the process of neighborhood search. Extensive evaluations show that CEMCNP running on 35 of total 42 benchmark instances are superior to the outcome of KBV, while holding 3 previous best results out of the challenging instances. In addition, CEMCNP also demonstrates equivalent performance in comparison with the existing MANCNP and VPMS algorithms over 22 of total 42 graph models with fewer number of node exchange operations.

A scalable network intrusion detection system towards detecting, discovering, and learning unknown attacks

Network intrusion detection systems (IDSs) based on deep learning have reached fairly accurate attack detection rates. But these deep learning approaches usually have been performed in a closed-set protocol that only known classes appear in training are considered during classification, the existing IDSs will fail to detect the unknown attacks and misclassify them as the training known classes, hence are not scalable. Furthermore, these IDSs are not efficient for updating the deep detection model once new attacks are discovered. To address those problems, we propose a scalable IDS towards detecting, discovering, and learning unknown attacks, it has three components. Firstly, we propose the open-set classification network (OCN) to detect unknown attacks, OCN based on the convolutional neural network adopts the nearest class mean (NCM) classifier, two new loss are designed to jointly optimize it, including Fisher loss and maximum mean discrepancy (MMD) loss. Subsequently, the semantic embedding clustering method is proposed to discover the hidden unknown attacks from all unknown instances detected by OCN. Then we propose the incremental nearest cluster centroid (INCC) method for learning the discovered unknown attacks through updating the NCM classifier. Extensive experiments on KDDCUP’99 dataset and CICIDS2017 dataset indicate that our OCN outperforms the state-of-the-art comparison methods in detecting multiple types of unknown attacks. Our experiments also verify the feasibility of the semantic embedding clustering method and INCC in discovering and learning unknown attacks.

Incremental Cluster Validity Indices for Online Learning of Hard Partitions: Extensions and Comparative Study

Validation is one of the most important aspects of clustering, particularly when the user is designing a trustworthy or explainable system. However, most clustering validation approaches require batch calculation. This is an important gap because of the value of clustering in real-time data streaming and other online learning applications. Therefore, interest has grown in providing online alternatives for validation. This paper extends the incremental cluster validity index (iCVI) family by presenting incremental versions of Calinski-Harabasz (iCH), Pakhira-Bandyopadhyay-Maulik (iPBM), WB index (iWB), Silhouette (iSIL), Negentropy Increment (iNI), Representative Cross Information Potential (irCIP), Representative Cross Entropy (irH), and Conn_Index (iConn_Index). This paper also provides a thorough comparative study of correct, under- and over-partitioning on the behavior of these iCVIs, the Partition Separation (PS) index as well as four recently introduced iCVIs: incremental Xie-Beni (iXB), incremental Davies-Bouldin (iDB), and incremental generalized Dunn’s indices 43 and 53 (iGD43 and iGD53). Experiments were carried out using a framework that was designed to be as agnostic as possible to the clustering algorithms. The results on synthetic benchmark data sets showed that while evidence of most under-partitioning cases could be inferred from the behaviors of the majority of these iCVIs, over-partitioning was found to be a more challenging problem, detected by fewer of them. Interestingly, over-partitioning, rather then under-partitioning, was more prominently detected on the real-world data experiments within this study. The expansion of iCVIs provides significant novel opportunities for assessing and interpreting the results of unsupervised lifelong learning in real-time, wherein samples cannot be reprocessed due to memory and/or application constraints.

Quantum-chemical study of 7Li NMR shifts in the context of delithiation of paramagnetic lithium vanadium phosphate, Li3V2(PO4)3 (LVP)

The lithium NMR shifts of three paramagnetic materials important in the charging/discharging processes of lithium vanadium phosphate cathode materials have been studied by large-scale quantum-chemical methodology. Namely, the 7Li NMR shifts of the fully lithiated Li3V2(PO4)3 (LVP3.0), and of the partly delithiated Li2.5V2(PO4)3 (LVP2.5) and Li2V2(PO4)3 (LVP2.0), have been computed and analyzed using a recently proposed approach (A. Mondal, M. Kaupp J. Phys. Chem. C 123 (2019) 8387-8405) that accounts for the Fermi-contact, pseudo-contact, as well as orbital shifts, combining periodic computations with an incremental cluster model. LVP3.0 and LVP2.0 exhibit three and two unique Li sites, respectively, which could be assigned to their experimental 7Li NMR signals. In case of LVP2.0, the computations clearly assigned the signals at 143 ppm and 77 ppm to the Li(1) and Li(2) sites, respectively, even though the latter is connected to Vb(+III d2 sites and the former to Va(+IV) d1 sites. LVP2.5 is the most complex of these three materials, exhibiting a 50% occupation of Li(3) sites, which generates much more complicated Li NMR spectra with seven peaks that partly are closely spaced. Exploring three different occupation patterns, the computations can clearly assign five of the seven signals to one type of Li site and give most probable assignments for the two remaining signals. Notably, the calculations support seven signals to be assigned to LVP2.5, while previous interpretations took two of the signals as being entirely due to contamination by LVP2.0. The accuracy of the computations could probably be improved further by full DFT optimization of large super-cell structures. This work suggests that first-principles computations of NMR shifts of extended paramagnetic solids provide an important tool for the analysis of even rather complex NMR spectra.

An effective density-based clustering and dynamic maintenance framework for evolving medical data streams

BackgroundMedical data stream clustering has become an integral part of medical decision systems since it extracts highly-sensitive information from a tremendous flow of medical data. However, clustering and maintaining of medical data streams is still a challenging task. That is because the evolving of medical data streams imposes various challenges for clustering such as the ability to discover the arbitrary shape of a cluster, the ability to group data streams without a predefined number of clusters, and the ability to maintain the data clusters dynamically. ObjectiveTo support the online medical decisions, there is a need to address the clustering challenges. Therefore, in this paper, we propose an effective density-based clustering and dynamic maintenance framework for grouping the patients with similar symptoms into meaningful clusters and monitoring the patients’ status frequently. MethodsFor clustering, we generate a set of initial medical data clusters based on the combination of Piece-wise Aggregate Approximation and the density-based spatial clustering of applications with noise called (PAA+DBSCAN) algorithm. For maintenance, when new medical data streams arrive, we maintain the initially generated medical data clusters dynamically. Since the incremental cluster maintenance is time-consuming, we further propose an Advanced Cluster Maintenance (ACM) approach to improve the performance of the dynamic cluster maintenance. ResultsThe experimental results on real-world medical datasets demonstrate the effectiveness and efficiency of our proposed approaches. The PAA+DBSCAN algorithm is more efficient and effective than the exact DBSCAN algorithm. Moreover, the ACM approach requires less running time in comparison with the Baseline Cluster Maintenance (BCM) approach using different tuning parameter values in all datasets. That is because the BCM approach tracks all the data points in the cluster. ConclusionThe proposed framework is capable of clustering and maintaining the medical data streams effectively by means of grouping the patients who share similar symptoms and tracking the patients status that naturally tends to be changing over time.

Computation of NMR Shifts for Paramagnetic Solids Including Zero-Field-Splitting and Beyond-DFT Approaches. Application to LiMPO4 (M = Mn, Fe, Co, Ni) and MPO4 (M = Fe, Co)

A protocol for the computation of NMR shifts of paramagnetic spin-coupled solids is described and applied to the olivine-type lithium-ion battery cathode materials LiMPO4 (M = Mn, Fe, Co, Ni) and to the related binary phosphates MPO4 (M = Fe, Co). The computational approach includes Fermi-contact (FC), pseudocontact (PC), as well as orbital contributions to the shifts and combines periodic density functional computations with multireference post-Hartree–Fock methods. A shift formalism based on EPR spin-Hamiltonian parameters corrected for residual spin couplings within the Curie–Weiss regime is used. Orbital shieldings as well as hyperfine couplings (HFCs) are obtained at DFT levels for large simulation cells using the CP2K code, taking advantage of hybrid functionals for the HFCs. g-Tensors and zero-field-splitting tensors are computed within an incremental cluster model, using CASSCF and NEVPT2 approaches, providing thus the first post-Hartree–Fock treatments of these properties for extended solids. Com...

Online cluster validity indices for performance monitoring of streaming data clustering

Cluster analysis is used to explore structure in unlabeled batch data sets in a wide range of applications. An important part of cluster analysis is validating the quality of computationally obtained clusters. A large number of different internal indices have been developed for validation in the offline setting. However, this concept cannot be directly extended to the online setting because streaming algorithms do not retain the data, nor maintain a partition of it, both needed by batch cluster validity indices. In this paper, we develop two incremental versions (with and without forgetting factors) of the Xie-Beni and Davies-Bouldin validity indices, and use them to monitor and control two streaming clustering algorithms (sk-means and online ellipsoidal clustering), In this context, our new incremental validity indices are more accurately viewed as performance monitoring functions. We also show that incremental cluster validity indices can send a distress signal to online monitors when evolving structure leads an algorithm astray. Our numerical examples indicate that the incremental Xie-Beni index with a forgetting factor is superior to the other three indices tested.

Quantum-Chemical Approach to NMR Chemical Shifts in Paramagnetic Solids Applied to LiFePO4 and LiCoPO4.

A novel protocol to compute and analyze NMR chemical shifts for extended paramagnetic solids, accounting comprehensively for Fermi-contact (FC), pseudocontact (PC), and orbital shifts, is reported and applied to the important lithium ion battery cathode materials LiFePO4 and LiCoPO4. Using an EPR-parameter-based ansatz, the approach combines periodic (hybrid) DFT computation of hyperfine and orbital-shielding tensors with an incremental cluster model for g- and zero-field-splitting (ZFS) D-tensors. The cluster model allows the use of advanced multireference wave function methods (such as CASSCF or NEVPT2). Application of this protocol shows that the 7Li shifts in the high-voltage cathode material LiCoPO4 are dominated by spin-orbit-induced PC contributions, in contrast with previous assumptions, fundamentally changing interpretations of the shifts in terms of covalency. PC contributions are smaller for the 7Li shifts of the related LiFePO4, where FC and orbital shifts dominate. The 31P shifts of both materials finally are almost pure FC shifts. Nevertheless, large ZFS contributions can give rise to non-Curie temperature dependences for both 7Li and 31P shifts.

Large-Scale Computation of Nuclear Magnetic Resonance Shifts for Paramagnetic Solids Using CP2K.

Large-scale computations of nuclear magnetic resonance (NMR) shifts for extended paramagnetic solids (pNMR) are reported using the highly efficient Gaussian-augmented plane-wave implementation of the CP2K code. Combining hyperfine couplings obtained with hybrid functionals with g-tensors and orbital shieldings computed using gradient-corrected functionals, contact, pseudocontact, and orbital-shift contributions to pNMR shifts are accessible. Due to the efficient and highly parallel performance of CP2K, a wide variety of materials with large unit cells can be studied with extended Gaussian basis sets. Validation of various approaches for the different contributions to pNMR shifts is done first for molecules in a large supercell in comparison with typical quantum-chemical codes. This is then extended to a detailed study of g-tensors for extended solid transition-metal fluorides and for a series of complex lithium vanadium phosphates. Finally, lithium pNMR shifts are computed for Li3V2(PO4)3, for which detailed experimental data are available. This has allowed an in-depth study of different approaches (e.g., full periodic versus incremental cluster computations of g-tensors and different functionals and basis sets for hyperfine computations) as well as a thorough analysis of the different contributions to the pNMR shifts. This study paves the way for a more-widespread computational treatment of NMR shifts for paramagnetic materials.

BIG DATA AND ANALYSIS OF WEATHER FORECASTING SYSTEM

The whole world is suffering from the changing clement and their side. To reduce this side effects up to some extent there are many techniques and algorithms through which we can predict the weather on the basis of given data. After doing the analysis of existing techniques we conclude that data mining technique enable us to analyze the given set of data and extract the useful information from the given data. Therefore in order to understand the fluctuating patterns of weather conditions, a predictive model is studied. In this paper, we are using incremental K-means cluster algorithm which groups the same type data sets together and to prefigure the data we are using R-tool that gives structural data. At the end, result will be calculated on the basis of some mathematical conditions.

Incremental fuzzy cluster ensemble learning based on rough set theory

To deal with the uncertainty, vagueness and overlapping distribution within the data sets, a novel incremental fuzzy cluster ensemble method based on rough set theory (IFCERS) is proposed by the idea of combining clustering analysis task with classification techniques. Firstly, on the basis of soft clustering results, the positive region, boundary region and negative region of clustering ensemble are obtained by applying the construction of rough approximation in rough set theory, and then a group structure within data points of positive region is obtained by adopting a fuzzy cluster ensemble method. Secondly, by combining with the supervised ensemble learning method, e.g., random forests, the obtained group structure is used to construct the random forests classifier to classify the data points in boundary region. Finally, all the acquired group structure is used to train the random forests classifier to classify the data points of negative region. Experimental evaluations on UCI machine learning repository datasets verify the effectiveness of the proposed method. It is also shown that the quality of the final solution has a weak correlation with the ensemble size, the parameter setting on the rough approximations construction is appropriate, and the proposed method is robust towards the diversity from hard clustering members.

Sub-10 nm nano-gap device for single-cluster transport measurements

We present a versatile procedure for the fabrication of single electron transistor (SET) devices with nanometer-sized clusters and embedded back gate electrode. The process uses sputtering gas-aggregation for the growth of clusters and e-beam lithography with double angle shadow-edge deposition to obtain electrodes separated by nano-gaps with width below 10 nm. The nano-gap width is easily controlled only by geometrical factors such as deposited thin film thickness and evaporation angles. The usefulness of this technique is demonstrated by measuring the SET behavior of a device with a 4 nm cobalt cluster embedded in alumina, where the Coulomb blockade and incremental cluster charging can be readily identified without resorting to the differential conductivity.

An Incremental Rapid DBSCAN Clustering Algorithm for Detecting Software Vulnerabilities

Many current methods for detecting software vulnerabilities couldn’t use the collected vulnerability features to describe the unknown vulnerability, so the unknown vulnerabilities couldn’t be detected efficiently. In this paper, an incremental rapid DBSCAN clustering algorithm for detecting software vulnerabilities is proposed. Sequence pattern library is built by rapid DBSCAN, and the analyzing software vulnerability sequence is clustered based on incremental rapid DBSCAN algorithm. Firstly, the affected set and seeds set of analyzing sequence are calculated. Then the cluster is extended from seed set. For the analyzing sequence, the insert operation is completed. It doesn’t need to re-cluster vulnerabilities on the entire sequence set, but implement incremental cluster by using the previous cluster results. Lastly, the vulnerability type is similar to the sequence with the maximum similarity in its cluster. Experimental analysis shows that the efficiency of detection unknown vulnerability is greatly improved by using this method.

Adaptive online solder joint inspection algorithm based on incremental clustering

An incremental cluster based adaptive online solder joint inspection algorithm is proposed and demonstrated. First, the framework of an intelligent automated optical inspection (AOI) system is designed. Then, the initial solder joint classifier is established with the clustering algorithm. Finally, with a new incremental clustering algorithm, the AOI system can automatically adjust inspection parameters according to the feedback from the repair station. Experiments have shown that the proposed algorithm has a high inspection accuracy, and it is adaptive to environmental changes.

Cluster Analysis for Drilling-Quality Based on the Modified Algorithm of InDBSCAN

To detect the quality of batch drilling quickly，a new approach based on Acoustic Emission signals is presented. The signals’ statistical characteristics are extracted from acoustic emission signals in Time-domain, and then the signals’ eigenvectors are constructed to reflect each drilling process. A modified incremental clustering algorithm InDBSCAN is used to cluster these eigenvectors，and the batch drilling-quality can be analysed indirectly. Calculation and analysis results show that: the conclusion of incremental cluster analysis is more reasonable by the modified incremental clustering method of InDBSCAN. The detection accuracy of the batch drilling-quality is up to 84.3% according to the manual quality inspection.