Soft Clustering Techniques Research Articles

Investigating the effect of inflation on the consumption pattern of Iranian households

The big data sources of National Statistical Offices (NSOs) are provided to make a superior platform for decision-making. The household income and expenditure survey is one of the economically important surveys especially when the inflation rate varies to assess the changes in households’ consumption patterns. In this case, big data can be beneficial and help to accurately measure consumption patterns of urban and rural households at every geographical level. This analysis is an exploratory study for the extraction of the size of injustice and imparity of household income and facilities implemented by classifying and clustering all Iranian households. Through this study, classification and soft clustering (Fuzzy clustering) techniques are employed to characterize the Iranian household types from 2011 to 2021, which are supervised and unsupervised approaches, respectively. Moreover, association rule mining techniques are employed to discover and extract consumption patterns for each cluster. Obtained results showed that there was a significant gap between purchasing power/receiving energy between lowest and highest income households from 2011 to 2021, and this gap is increasing day by day.

Assessment, Categorisation and Prediction of the Landslide-Affected Regions Using Soft Computing and Clustering Techniques

Assessment, Categorisation and Prediction of the Landslide-Affected Regions Using Soft Computing and Clustering Techniques

Aerosol classification using fuzzy clustering over a tropical rural site

Exact knowledge on the types of aerosols present over a region is very important in the radiative forcing estimates. Several techniques are available to classify aerosols, which have their own pros and cons. In the present study, a soft clustering technique, Fuzzy C-Means (FCM) is employed to classify aerosols using multiple aerosol parameters over a rural site Gadanki (13.48oN, 79.18°E), India. The optical and microphysical properties of the aerosols (26 parameters) derived from long-term measurements of the Sky Radiometer (from April 2008 to October 2020) are utilized for cluster analysis. The three resultant clusters were found to be distinct and compact based on high intra-cluster similarity and low inter-cluster similarity. By comparing the clustering results of different subsets of data and testing the robustness of the FCM algorithm, we found that the correct classification percentage for each scenario was about 98.4%. Even by introducing a 10% random error for each of the aerosol microphysical properties used in the cluster analysis, around 94.5% of the records were found to retain the original clusters. Based on the aerosol properties of the clusters, the three clusters were identified as mixed mode moderately absorbing, mixed mode slightly absorbing, and coarse mode slightly absorbing type of aerosols. While comparing with the global aerosol models, these three aerosol types showed similarities with biomass burning, polluted marine, and polluted continental aerosol types. Significant seasonal variability of the identified aerosol types was observed, with biomass burning type dominating in the pre-monsoon season, polluted marine during the monsoon season, and polluted continental in a relatively higher percentage compared to the other two types in winter and post-monsoon seasons. An increasing trend in polluted marine aerosol type is noticed, whilst biomass burning and polluted continental showed decreasing trends. Furthermore, the existing knowledge on various aspects of aerosols over Gadanki ascertains that the identified aerosol types are reasonable, implying that the FCM clustering technique for aerosol classification is effective and it might be extend to other locations.

Multimorbidity patterns of chronic conditions and geriatric syndromes in older patients from the MoPIM multicentre cohort study

ObjectivesTo estimate the frequency of chronic conditions and geriatric syndromes in older patients admitted to hospital because of an exacerbation of their chronic conditions, and to identify multimorbidity clusters in...

Combined Multimorbidity and Polypharmacy Patterns in the Elderly: A Cross-Sectional Study in Primary Health Care.

(1) Background: The acquisition of multiple chronic diseases, known as multimorbidity, is common in the elderly population, and it is often treated with the simultaneous consumption of several prescription drugs, known as polypharmacy. These two concepts are inherently related and cause an undue burden on the individual. The aim of this study was to identify combined multimorbidity and polypharmacy patterns for the elderly population in Catalonia. (2) Methods: A cross-sectional study using electronic health records from 2012 was conducted. A mapping process was performed linking chronic disease categories to the drug categories indicated for their treatment. A soft clustering technique was then carried out on the final mapped categories. (3) Results: 916,619 individuals were included, with 93.1% meeting the authors’ criteria for multimorbidity and 49.9% for polypharmacy. A seven-cluster solution was identified: one non-specific (Cluster 1) and six specific, corresponding to diabetes (Cluster 2), neurological and musculoskeletal, female dominant (Clusters 3 and 4) and cardiovascular, cerebrovascular and renal diseases (Clusters 5 and 6), and multi-system diseases (Cluster 7). (4) Conclusions: This study utilized a mapping process combined with a soft clustering technique to determine combined patterns of multimorbidity and polypharmacy in the elderly population, identifying overrepresentation in six of the seven clusters with chronic disease and chronic disease-drug categories. These results could be applied to clinical practice guidelines in order to better attend to patient needs. This study can serve as the foundation for future longitudinal regarding relationships between multimorbidity and polypharmacy.

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.

Measurement of health service performance through machine learning using clustering techniques

The health center is the first level public health service center that is run by the Indonesian government. Organizing quality health center services of health center is the hope and satisfaction of every patient. The dimensions of patient satisfaction vary greatly and the scope is very wide. Patient satisfaction can be used as an indicator of quality and service performance of health center. The purpose of this study was to measure the performance of health center services in order to determine the level of patient satisfaction and grouping the Patient Satisfaction Index. With the knowledge of the dimensions of performance and quality of health services provided by the health center, it will facilitate the government in carrying out the function of guidance and control of the health center. This study uses a Machine Learning technology approach with clustering techniques, by grouping the Patient Satisfaction Index with K-Means (Hard Clustering) and Fuzzy C-Mean (Soft Clustering) methods. Based on the subsets produced, clustering techniques can be divided into 2 methods, namely hard clustering techniques and soft clustering techniques. The K-Mean method is widely used in clustering techniques. K-Mean has advantages in computational speed and relatively easy process stages. While the Fuzzy C-Means method has advantages in terms of flexibility in determining clusters so that there is little possibility of converging failure. The experimental results of the Patient Satisfaction Index show that the K-Means method provides better performance with a value of 96% compared to the Fuzzy C-Means method with a value of 76%.

A soft clustering technique with layered feature extraction for social image mining

Social image mining is beneficial to accomplish tasks like event detection, suspicious activity detection, prediction of future trends, identification of mentally depressed people, etc. To carry out social image mining, data mining techniques need to be used. Clustering is one of the most important tasks of data mining which is able to deal with the unlabelled data. But, less number of clustering approaches are having ability to handle the uncertain image data. Thus, in this paper we proposed a soft clustering algorithm named as ROugh Mean Shift clustering (ROMS) with layered feature extraction model for social images. Effectiveness of the rough set theory and mean shift concepts are incorporated in this algorithm. It makes the ROMS to deal with the vagueness and the automatic determination of cluster numbers in given data. Proposed method is experimented on three datasets- synthetic, standard and real-world datasets and compared with existing techniques. Experimental results show that ROMS performs better as compared to other techniques.

Soft-clustering Technique for Fingerprint-based Localization

Soft-clustering Technique for Fingerprint-based Localization

Analyzing cloud based reviews for product ranking using feature based clustering algorithm

The rapid expansion of e-commerce improves customer experience and online businesses largely. Customers buy products based on reviews provided by the users of the product through cloud. Opinion of customers plays a huge role in decision making for business. Opinion mining helps in comparison of products by identifying the main features commented upon by customers and also ranking the products. Most of the opinion mining systems identify only the explicitly mentioned features from the reviews. Identification of implicit features and considering the context of the reviewer about a product plays a huge role in better decision making. The proposed system uses a novel approach for identifying explicit features, implicit features and the context of the reviewer through soft clustering techniques for forming the clusters. Since opinion comparison and ranking of products are based on multi-alternative criteria, a multi criteria decision making approach is used for ranking the products. An experiment is designed to test the proposed method using the reviews collected from Amazon cloud set for mobile phone and camera reviews. The result shows that the proposed method identifies the key features of the products and provides the better ranking of products as it considers the explicit features, implicit features and the context of the reviewer.

Secure weighted possibilistic c-means algorithm on cloud for clustering big data

The weighted possibilistic c-means algorithm is an important soft clustering technique for big data analytics with cloud computing. However, the private data will be disclosed when the raw data is directly uploaded to cloud for efficient clustering. In this paper, a secure weighted possibilistic c-means algorithm based on the BGV encryption scheme is proposed for big data clustering on cloud. Specially, BGV is used to encrypt the raw data for the privacy preservation on cloud. Furthermore, the Taylor theorem is used to approximate the functions for calculating the weight value of each object and updating the membership matrix and the cluster centers as the polynomial functions which only include addition and multiplication operations such that the weighed possibilistic c-means algorithm can be securely and correctly performed on the encrypted data in cloud. Finally, the presented scheme is estimated on two big datasets, i.e., eGSAD and sWSN, by comparing with the traditional weighted possibilistic c-means method in terms of effectiveness, efficiency and scalability. The results show that the presented scheme performs more efficiently than the traditional weighted possiblistic c-means algorithm and it achieves a good scalability on cloud for big data clustering.

A Rough–Fuzzy approach for Support Vector Clustering

Support Vector Clustering (SVC) is an important density-based clustering algorithm which can be applied in many real world applications given its ability to handle arbitrary cluster silhouettes and detect the number of classes without any prior knowledge. However, if outliers are present in the data, the algorithm leaves them unclassified, assigning a zero membership degree which leads to all these objects being treated in the same way, thus losing important information about the data set. In order to overcome these limitations, we present a novel extension of this clustering algorithm, called Rough–Fuzzy Support Vector Clustering (RFSVC), that obtains rough–fuzzy clusters using the support vectors as cluster representatives. The cluster structure is characterized by two main components: a lower approximation, and a fuzzy boundary. The membership degrees of the elements in the fuzzy boundary are calculated based on their closeness to the support vectors that represent a specific cluster, while the lower approximation is built by the data points which lie inside the hyper-sphere obtained in the training phase of the SVC algorithm. Our computational experiments verify the strength of the proposed approach compared to alternative soft clustering techniques, showing its potential for detecting outliers and computing membership degrees for clusters with any silhouette.

Decision Theoretic Evaluation of Rough Fuzzy Clustering

Clustering is the process of organizing dissimilar objects into natural groups in such a way objects in the same group is more similar than objects in the different groups. Since we know clustering is an unsupervised learning problem, typical clustering algorithms not achieving its end to handle uncertainty that exists in the real life experience. Though fuzzy clustering handles incompleteness and vagueness in the data set efficiently, it is highly descriptive than hard clustering algorithm. Rough clustering algorithm is the popular soft clustering technique which uses rough set to handle uncertainty. In Rough Fuzzy clustering, each cluster is represented by centroid, crisp lower approximation and fuzzy boundary. Clustering undergoes sequence of partitions where cluster evaluation is the final step in clustering process. Efficient clustering structure can be obtained through validity measures. Various validity measures have been proposed to evaluate rough fuzzy clustering. Since those measures are Geometric measures, this paper proposes decision theoretic measure for validating rough fuzzy clustering structure.

On Clustering Histograms with k-Means by Using Mixed α-Divergences

Clustering sets of histograms has become popular thanks to the success of the generic method of bag-of-X used in text categorization and in visual categorization applications. In this paper, we investigate the use of a parametric family of distortion measures, called the α-divergences, for clustering histograms. Since it usually makes sense to deal with symmetric divergences in information retrieval systems, we symmetrize the α -divergences using the concept of mixed divergences. First, we present a novel extension of k-means clustering to mixed divergences. Second, we extend the k-means++ seeding to mixed α-divergences and report a guaranteed probabilistic bound. Finally, we describe a soft clustering technique for mixed α-divergences.

Feature weighted unsupervised classification algorithm and adaptation for software cost estimation

Multi-attribute data, needed to be clustered may have different consequences of attributes on clustering criteria. In this paper, a new soft clustering technique is proposed in which similarity measures between data points and impact of each attribute is calculated using grey relational analysis. Algorithm provides the flexibility to choose significant number of attributes for classification purpose using feature subset selection. An iterative approach is adopted to find desired number of clusters having more appropriate and unique centroid. In addition, the use of proposed technique is instanced on software cost estimation because inherent uncertainty in software attributes due to the measurement by expert judgment has a significant impact on estimation accuracy. Combination of clustering and regression technique reduces the potential problem in efficacy of predictive assays due to heterogeneity of the data. Clustered approach creates the subsets of data having a degree of homogeneity that elaborate more accurate prediction.

Evolving soft subspace clustering

A key challenge to most conventional clustering algorithms in handling many real world problems is that, data points in different clusters are often correlated with different subsets of features. To address this problem, subspace clustering has attracted increasing attention in recent years. In practical data mining applications, data points may arrive in continuous streams with chunks of samples being collected at different time points. In addition, huge amounts of data often cannot be kept in the main memory due to memory restriction. Accordingly, a range of evolving clustering algorithms has been proposed, however, traditional evolving clustering methods cannot be effectively applied to large-scale high dimensional data and data streams. In this study, we extend the online learning strategy and scalable clustering technique to soft subspace clustering to form evolving soft subspace clustering. We propose two online soft subspace clustering algorithms, OFWSC and OEWSC, and two streaming soft subspace clustering algorithms, SSSC_F and SSSC_E. The proposed evolving soft subspace clustering leverages on the effectiveness of online learning scheme and scalable clustering methods for streaming data by revealing the important local subspace characteristics of high dimensional data. Substantial experimental results on both artificial and real-world datasets demonstrate that our proposed methods are generally effective in evolving clustering and achieve superior performance over existing soft subspace clustering techniques.

Novel soft subspace clustering with multi-objective evolutionary approach for high-dimensional data

Many conventional soft subspace clustering techniques merge several criteria into a single objective to improve performance; however, the weighting parameters become important but difficult to set. In this paper, a novel soft subspace clustering with a multi-objective evolutionary approach (MOEASSC) is proposed to this problem. This clustering method considers two types of criteria as multiple objectives and optimizes them simultaneously by using a modified multi-objective evolutionary algorithm with new encoding and operators. An indicator called projection similarity validity index (PSVIndex) is designed to select the best solution and cluster number. Experiments on many datasets demonstrate the usefulness of MOEASSC and PSVIndex, and show that our algorithm is insensitive to its parameters and is scalable to large datasets.

Profiling of sunk cost industries by soft clustering techniques: Turkey case

The aim of this study is to determine the convenience of Sutton’s game-theoretic sunk cost theory by profiling Turkish manufacturing industries via fuzzy c-means (FCM) and expectation-maximisation (EM) clustering methods. The effects of sunk costs on market structure are analysed separately for exogenous and endogenous sunk cost industries in Turkish manufacturing sectors in the period 1992–2001. The results suggest that as market size increases, so does the level of advertising and R&D outlays leading to a concentrated market structure in endogenous sunk cost industries. On the other hand, in exogenous sunk cost industries, market structure becomes more competitive and number of new entrant firms rise as market size growths. According to the profiles constituted by FCM and EM methods, these findings are consistent with Sutton’s predictions. Furthermore, the most important factors which discriminates the clusters are revealed by FCM method.

Soft and hard clustering methods for delineation of hydrological homogeneous regions in the southern strip of the Caspian Sea Watershed

AbstractRegionalisation is a fundamental task for transferring hydrological information from gauged to un‐gauged catchments. This study employs several soft and hard clustering approaches for regionalisation of the 61 catchments located in the southern strip of the Caspian Sea Watershed. Factor analysis using Principal Component Analysis resulted to four out of 16 catchment's attributes relating to flooding which were then used for regionalisation. Hierarchical and Non‐Hierarchical Clustering, K‐Means, Fuzzy C‐Means and Kohonen methods were studied and compared by L‐Moment tests. While in hard clustering approach, distinct clusters of catchments were formed, soft clustering techniques allocated catchments to more than one cluster according to their membership probability. The performance of different clustering methods resulted to the similar number of homogeneous groups. However, the number of sites allocated to the clusters is different. Hard clustering resulted in three clusters at 38, 10 and 13 sites, while Soft clustering allocated 26, 20 and 15 sites to the clusters, respectively. Results indicate a superiority of the soft clustering in the study area where deriving hydrologic groups by hard clustering is problematic.

Improving the quality of predictions using textual information in online user reviews

Online reviews are often accessed by users deciding to buy a product, see a movie, or go to a restaurant. However, most reviews are written in a free-text format, usually with very scant structured metadata information and are therefore difficult for computers to understand, analyze, and aggregate. Users then face the daunting task of accessing and reading a large quantity of reviews to discover potentially useful information. We identified topical and sentiment information from free-form text reviews, and use this knowledge to improve user experience in accessing reviews. Specifically, we focus on improving recommendation accuracy in a restaurant review scenario. We propose methods to derive a text-based rating from the body of the reviews. We then group similar users together using soft clustering techniques based on the topics and sentiments that appear in the reviews. Our results show that using textual information results in better review score predictions than those derived from the coarse numerical star ratings given by the users. In addition, we use our techniques to make fine-grained predictions of user sentiments towards the individual topics covered in reviews with good accuracy.