Clustering-based Model Research Articles

Whale Optimization Based Recommendation System

Recommendation System is an information filtering system which seeks to predict the “liking” of a user for an item, with the aim to suggest the user those items which he/she is most likely to select/buy. The focus of this paper is on rating prediction whose main objective is to predict the ratings the current user is going to give to the items which are yet to be rated/viewed by him/her. This paper uses a collaborative filtering based approach for generating recommendation, and the model used is a clustering-based model. In this approach all the existing users are clustered using whale optimization technique, instead of traditional clustering approaches like k-means, EM algorithm, etc. The appropriate cluster is then identified for the active user, and the ratings of the active user are predicted based on ratings given by other users belonging to the same cluster. Different measures like MAE, SD, RMSE and t-value are used for performance analysis of the proposed method and the results obtained are found to be highly accurate.

Entity resolution for media metadata based on structural clustering

An increasing amount of media metadata are published by different organizations on the Web which leads to a fragmented dataset landscape. Identifying media metadata from disparate datasets and integrating heterogeneous datasets have many applications but also pose significant challenges. To tackle this problem, entity resolution methods are commonly used as an essential prerequisite for integrating media information from different sources and effectively foster the re-use of existing data sources. While the amount of media metadata published on the Web grows steadily, how to scale it well to large media knowledge bases while maintaining a high matching quality is a critical challenge. This article investigates the relationships between media entities. To that end, the media database is formulated as a knowledge graph with entities as nodes and the associations between related entities as edges. Thus, media entities are grouped into communities by how they share neighbors. Then, a structural clustering-based model is proposed to detect communities and discover anchor vertices as well as isolated vertices. Specifically, an initial seed set of matched anchor vertex pairs is obtained. Furthermore, an iterative propagation approach for identifying the matched entities in the whole graph is developed, where community similarity is introduced into the measure function to control the total measurement of candidate pairs. Therefore, starting with the elements of the initial seed set, the entity resolution algorithm updates the matching information over the whole network along with the neighbor relationships iteratively. Extensive experiments are conducted on real datasets to evaluate how the seed set impacts the matching process and performance. The experiment results show this model can achieve an excellent balance between accuracy and efficiency and is a clear improvement compared to state-of-the-art methods.

A Clustering-Based Equipment Condition Model of Chemical Vapor Deposition Process

In semiconductor manufacturing, equipment condition monitoring is important to improve the efficiency of the manufacturing process by performing equipment maintenance in a timely manner. In this paper, we propose the clustering-based equipment condition model to select key sensors relating to a maintenance. During the manufacturing process, huge amounts of data are collected in real time from sensors on the equipment. The sensor data has various patterns, such as increased pattern, decreased pattern, unchanged pattern, and other patterns. We apply five clustering algorithms to group the sensors with similar characteristics and extract key sensors that are highly correlated with equipment health condition. The health condition monitoring model consists of the combination of key sensors. To validate proposed method, the empirical study is conducted using collected sensor data from a chemical vapor deposition (CVD) process in a semiconductor company in the Republic of Korea. The model with clustered sensors outperforms the model with full sensors. The health condition monitoring model assists engineers in making decisions regarding the equipment maintenance.

Development of a clustering-based model for enhancing acoustic leak detection

According to the Canadian Infrastructure Report of 2016, Canada’s water and wastewater infrastructures are in a declining state. One of the problems plaguing water systems is leakage. Leaks are costly as they create losses in precious water resources as well as treatment chemicals and energy required to produce drinking water. Therefore, the city of Montréal has implemented a pilot project to detect the leaks in a portion of its water supply network using noise loggers. The main shortcoming tackled is the inaccuracy of the current system as it can regularly present false rulings on new events. This article presents a novel approach for the analysis of the signals using k-means clustering and provides a set of models for leak detection. The developed model was tested against real-life conditions and detected two possible leaks that were undetected by the current system in addition to its ability to detect all confirmed leak conditions.

Applying evolutionary optimization algorithms for improving fuzzy C-mean clustering performance to predict the deformation modulus of rock mass

This paper focuses on the capability of the evolutionary computation methods namely, genetic algorithm (GA) and particle swarm optimization (PSO) in design and optimizing the fuzzy c-means clustering (FCM) structure and their applications to predict the deformation modulus of rock masses. Accordingly, evolutionary algorithms are used to tune the pre-determined FCM clustering-based model to make better the accuracy of modulus estimation. A new empirical equation with the aid of multiple regression analysis is also suggested and on the basis of it a prediction chart is presented for determination of the rock mass deformation modulus. Finally, a comprehensive credibility assessment of the prediction performances of some existing empirical equations is done and the results are compared with that obtained by the evolutionary algorithms based FCM clustering models. It is concluded that the new proposed approaches provide more accurate results compared with the existing empirical equations.

K-means++ clustering-based active contour model for fast image segmentation

Intensity inhomogeneity often occurs in real-life images because of nonuniform illumination, device operating, and technical limitation. We propose a K-means++ clustering-based active contour model for fast image segmentation. The key point is two fitting functions whose value computed by K-means++ clustering algorithm before level set function evolution. At first, we set up a rectangular local window. Two fitting functions represent the center points of brighter and darker subregions in the moving rectangular local windows. The method avoids repeating calculation of the fitting functions during curve evolution compared with the traditional region-based active contour models. Therefore, the proposed model has lower computational costs, and we can obtain correct segmentation results in less time and fewer iterations. The proposed model can efficiently segment images with intensity inhomogeneity. In addition, the experiments have proved that the proposed model has strong robustness to initialization.

Random generation of industrial pipelines’ data using Markov chain model

Random generation of data sets is a vital step in simulation modeling. It involves in generating the variation associated with the real system behavior. In the industrial fabrication of construction components, unique products such as pipelines are produced. The fabrication processes are dependent on pipelines features, and complexity; randomly generating pipelines structure is imperative in the simulation of such processes. This paper investigates the nature of industrial pipelines and proposes a Markov chain model to randomly generate pipelines data structure. The performance of Markov chain model was tested against real pipelines through a three-stage validation process. The validation process includes (1) a validation based on the number of components and the pipelines components correlation analysis, (2) clustering-based model validation, and (3) model validation using similarity distances between pipelines feature vectors. The Markov chain model was found to generate a reasonable pipelines data structure when compared with real pipelines. It was found that 89% of the generated pipelines share similar properties equivalent to 0.88 (a scale from 0 (not identical) to 1 (identical)) to 85.5% of the original pipelines.

Improved fragment-based protein structure prediction by redesign of search heuristics

Difficulty in sampling large and complex conformational spaces remains a key limitation in fragment-based de novo prediction of protein structure. Our previous work has shown that even for small-to-medium-sized proteins, some current methods inadequately sample alternative structures. We have developed two new conformational sampling techniques, one employing a bilevel optimisation framework and the other employing iterated local search. We combine strategies of forced structural perturbation (where some fragment insertions are accepted regardless of their impact on scores) and greedy local optimisation, allowing greater exploration of the available conformational space. Comparisons against the Rosetta Abinitio method indicate that our protocols more frequently generate native-like predictions for many targets, even following the low-resolution phase, using a given set of fragment libraries. By contrasting results across two different fragment sets, we show that our methods are able to better take advantage of high-quality fragments. These improvements can also translate into more reliable identification of near-native structures in a simple clustering-based model selection procedure. We show that when fragment libraries are sufficiently well-constructed, improved breadth of exploration within runs improves prediction accuracy. Our results also suggest that in benchmarking scenarios, a total exclusion of fragments drawn from homologous templates can make performance differences between methods appear less pronounced.

Novel Clustering-based Hidden Markov Model (CHMM) optimization approach for optimal PMU placement

A Phasor Measurement Unit (PMU) is an important device for monitoring the wide-area power distribution network. Placement of the PMUs across the network enables reliable monitoring of the network to identify the faults in the bus system. Due to the increase in the installation cost of the PMUs, optimal placement of the PMU is a significant task. But the existing techniques do not provide the optimal solution for the PMU placement. To overcome this issue, this paper proposes a novel Clustering-based Hidden Markov Model (CHMM) optimization approach to achieve optimal placement of PMUs in the power distribution network. Optimal PMU placement is achieved by applying cluster formation in the bus system to extract the data with neighboring buses. Best optimal position for placing the PMU is estimated by using Fuzzy logic-based rale formation to update the binary table of the bus system. The HMM approach is used for updating weight in the cluster formation. Our system is implemented in various bus systems like IEEE 28-bus system, 69-bus system and also in Karnataka 155-bus system. The proposed approach is implemented in the IEEE bus system and Karnataka Power Transmission Corporation Limited (KPTCL) bus system and compared with the existing approaches, based on the total number of PMU placement. The proposed approach achieves better performance in the optimal placement of PMU than the existing optimization algorithms.

Household monthly electricity consumption pattern mining: A fuzzy clustering-based model and a case study

Household monthly electricity consumption pattern mining is to discover different energy use patterns of households in a month from their daily electricity consumption data. In this study, we develop an improved fuzzy clustering model for the monthly electricity consumption pattern mining of households. First, the background of clustering and fuzzy c-means clustering is introduced. Then a process model of household electricity consumption pattern mining and an improved fuzzy c-means clustering model are provided. Three key aspects of the improved fuzzy c-means clustering model, namely fuzzifier selection, cluster validation and searching capability optimization, are discussed. Finally, the daily electricity consumption data of 1200 households in Jiangsu Province, China, during a month from December 1, 2014 to December 31, 2014 are used in the experiment. With the proposed model, 938 valid households are successfully divided into four and six groups respectively, and the characteristics of each group are extracted. The results revealed the different electricity consumption patterns of different households and demonstrated the effectiveness of the clustering-based model. The customer segmentation based on consumption pattern mining in electric power industry is of great significance to support the development of personalized and targeted marketing strategies and the improvement of energy efficiency.

Clustering-Based Model Reduction of Networked Passive Systems

The model reduction problem for networks of interconnected dynamical systems is studied in this paper. In particular, networks of identical passive subsystems, which are coupled according to a tree topology, are considered. For such networked systems, reduction is performed by clustering subsystems that show similar behavior and subsequently aggregating their states, leading to a reduced-order networked system that allows for an insightful physical interpretation. The clusters are chosen on the basis of the analysis of controllability and observability properties of associated edge systems, representing the importance of the couplings and providing a measure of the similarity of the behavior of neighboring subsystems. This reduction procedure is shown to preserve synchronization properties (i.e., the convergence of the subsystem trajectories to each other) and allows for the a priori computation of a bound on the reduction error with respect to external inputs and outputs. The method is illustrated by means of an example of a thermal model of a building.

Exploring More Representative States of Hidden Markov Model in Optical Character Recognition: A Clustering-Based Model Pre-Training Approach

Hidden Markov Model (HMM) is an effective method to describe sequential signals in many applications. As to model estimation issue, common training algorithm only focuses on the optimization of model parameters. However, model structure influences system performance as well. Although some structure optimization methods are proposed, they are usually implemented as an independent module before parameter optimization. In this paper, the clustering feature of states in HMM is discussed through comparing the mechanism of Quadratic Discriminant Function (QDF) classifier and HMM. Then, through the clustering effect of Viterbi training and Baum–Welch training, a novel clustering-based model pre-training approach is proposed. It can optimize model parameters and model structure by turns, until the representative states of all models are explored. Finally, the proposed approach is evaluated on two typical OCR applications, printed and handwritten Arabic text line recognition. And it is compared with some other optimization methods. The improvement of character recognition performance proves the proposed approach can make more precise state allocation. And the representative states are benefit to HMM decoding.

Clustering-based model of cysteine co-evolution improves disulfide bond connectivity prediction and reduces homologous sequence requirements.

Cysteine residues have particular structural and functional relevance in proteins because of their ability to form covalent disulfide bonds. Bioinformatics tools that can accurately predict cysteine bonding states are already available, whereas it remains challenging to infer the disulfide connectivity pattern of unknown protein sequences. Improving accuracy in this area is highly relevant for the structural and functional annotation of proteins. We predict the intra-chain disulfide bond connectivity patterns starting from known cysteine bonding states with an evolutionary-based unsupervised approach called Sephiroth that relies on high-quality alignments obtained with HHblits and is based on a coarse-grained cluster-based modelization of tandem cysteine mutations within a protein family. We compared our method with state-of-the-art unsupervised predictors and achieve a performance improvement of 25-27% while requiring an order of magnitude less of aligned homologous sequences (∼10(3) instead of ∼10(4)). The software described in this article and the datasets used are available at http://ibsquare.be/sephiroth. wvranken@vub.ac.be Supplementary material is available at Bioinformatics online.

Object tracking using adaptive covariance descriptor and clustering-based model updating for visual surveillance.

We propose a novel approach for tracking an arbitrary object in video sequences for visual surveillance. The first contribution of this work is an automatic feature extraction method that is able to extract compact discriminative features from a feature pool before computing the region covariance descriptor. As the feature extraction method is adaptive to a specific object of interest, we refer to the region covariance descriptor computed using the extracted features as the adaptive covariance descriptor. The second contribution is to propose a weakly supervised method for updating the object appearance model during tracking. The method performs a mean-shift clustering procedure among the tracking result samples accumulated during a period of time and selects a group of reliable samples for updating the object appearance model. As such, the object appearance model is kept up-to-date and is prevented from contamination even in case of tracking mistakes. We conducted comparing experiments on real-world video sequences, which confirmed the effectiveness of the proposed approaches. The tracking system that integrates the adaptive covariance descriptor and the clustering-based model updating method accomplished stable object tracking on challenging video sequences.

A clustering-based model for class responsibility assignment problem in object-oriented analysis

Assigning responsibilities to classes is a vital task in object-oriented analysis and design, and it directly affects the maintainability and reusability of software systems. There are many methodologies to help recognize the responsibilities of a system and assign them to classes, but all of them depend greatly on human judgment and decision-making. In this paper, we propose a clustering-based model to solve the class responsibility assignment (CRA) problem. The proposed model employs a novel interactive graph-based method to find inheritance hierarchies, and two novel criteria to determine the appropriate number of classes. It reduces the dependency of CRA on human judgment and provides a decision-making support for CRA in class diagrams. To evaluate the proposed model, we apply three different hierarchical agglomerative clustering algorithms and two different types of similarity measures. By comparing the obtained results of clustering techniques with the models designed by multi-objective genetic algorithm (MOGA), it is revealed that clustering techniques yield promising results.

Estimation of deformation modulus of rock masses by using fuzzy clustering-based modeling

The paper describes a method which incorporates Takagi–Sugeno (TS) fuzzy modeling with two data clustering approaches including fuzzy- c-means (FCM) clustering and subtractive clustering to estimate the rock mass modulus of deformation. For this aim, a database including 120 cases collected from several galleries of dam sites locations was established. The information returned by fuzzy clustering was initially used to define the number of rules and antecedent membership functions and afterwards linear least squares estimation implemented to obtain fuzzy consequent parameters. An adaptive neuro-fuzzy inference system (ANFIS) was applied to modify the pre-determined TS clustering-based model structures to improve the generalization performance of those. For evaluation of the performance, root mean square error ( RMSE) and variance account for ( VAF) values have been utilized as performance criteria. It can be said, that ANFIS approach enhances the performances of fuzzy clustering-based models in predicting modulus of deformation of rock masses successfully.

A clustering-based color model and integral images for fast object tracking

The paper presents a clustering-based color model and develops a fast algorithm for object tracking. The color model is built upon K-means clustering, by which the color space of the object can be partitioned adaptively and the histogram bins can be determined accordingly. In addition, in each bin the multi-channel gray level is modelled as Gaussian distribution. Defined in this way the color model can describe accurately the color distribution with very little bins. To evaluate similarity between the reference model and the candidate model, a similarity measure based on Bhattacharrya distance is introduced and its simplified form is derived under assumption that in each bin distribution of gray level in different channel is independent of each other. Motivated by the paper of Viola and Jones, the Integral Images for computation of histogram, mean and variance are proposed, with which the similarity measure can be evaluated at negligible computational cost. Thus, exhaustive search is made efficiently for object localization which guarantees the global maximum be achieved. Comparisons with the well-known mean shift algorithm demonstrate that the proposed algorithm has better performance while having the same (or less) computational cost.