Search

The COVID-19 pandemic has a strong worldwide impact on not only the health and economic sectors but also the (socio-)psychological functioning of individuals. Since psychological health is an important protective factor to prevent diseases, it is crucial to identify individuals with increased vulnerability during the crisis. 275 adults participated in a German online survey from April until August 2020 which investigated health-related, social, behavioral, and psychological effects of the COVID-19 pandemic. We here introduce an unsupervised clustering approach suitable for mixed data types combining the Gower distance with the Partitioning Around Medoids (PAM) algorithm k-Medoids. We were able to identify three clusters differing significantly in subjects’ well-being, psychological distress, and current financial and occupational concerns. The clusters also differed in age with younger persons reporting greater financial and occupational concerns, increased anxiety, higher psychological distress, and reduced subjective well-being. Features with the strongest impact on the clustering were examined using a wrapping method and the feature importance implemented in the random forest. Particularly, answers regarding financial and occupational concern, psychological distress, and current well-being were decisive for the assignment to a cluster. In summation, the clustering approach can identify persons with weakened psychological protective factors allowing them to provide tailored recommendations for preventive actions based on the cluster affiliation, e.g., via a web application.

Induction motor is widely used in industrial applications. The research paper presents the diagnosis of the induction motor bearing faults using Discrete Wavelet Transforms and Partitioning Around Medoids algorithm methods. The experimental test rig was developed to obtain data of the bearings on healthy or damaged conditions. Several mother-level wavelets are tried in order to get the best performance to find bearing faults. The wavelet transform results are used as an input of the Partitioning Around Medoids algorithm to cluster the bearing condition. The results showed that the methods proposed could provide an accurate diagnosis of the bearing condition.

If you want to classify object and you do not have any specific labels, then how can you classify that objects? Clustering is the best way to classify this kind of object. Whenever we are talking about large-scale data relating to variety of fields, beginning of new gen techniques for concentrate collection of data was resulted. Typical database query processing is not able to find exact information from large number of data, and hence, clustering is important analysis method for large-scale data. Among these many clustering algorithms, k-means and k-medoids are superiorly utilized for large-scale database. Initially, select centroids in k-mean algorithm and medoids for k-medoids algorithm for batter quality of the resulting clusters. Key highlight of this algorithm is that, whenever number of iterations increase, it will also increase the computational time. Proposed k-means algorithm initially finds the starting level centroids “k” as per requirements and returns with better compare to previous, effective and very stabilized cluster. Here proposed algorithm consumes very less time for execution as it segregates unnecessary computational distance because it uses the previous iteration of cycle. On the basis on initial centroids, here proposed algorithm systematically selects initial k-medoids. And thus, it will produce stable clusters for efficiency improvement.

In this rapidly rising field of Web, volume of online social networks has increased exponentially. This inspires the researchers to work in the area of information diffusion, i.e., spread of information through “word of mouth” effect. Information maximization is an important research problem of information diffusion, i.e., selection of k most influential nodes in the network such that they can maximize the information spread. In this paper, we proposed an influence maximization model that identifies optimal seeds to maximize the influence spread in the network. Our proposed algorithm is a hybrid approach, i.e., GA with k-medoid approach using dynamic edge strength. To analyze the efficiency of the proposed algorithm, experiments are performed on two large-scale datasets using fitness score measure. Experimental outcome illustrated 8–16% increment in influence propagation by proposed algorithm as compared to existing seed selection methods, i.e., general greedy, random, discounted degree, and high degree.

IoT is a revolutionary vision pertaining to the fact that everything could be connected to the internet. With the increase in the number of internet users, IoT clients are also increasing. In IoT environment, the data generated is enormous. So, we need an efficient approach to manage such a gigantic multi-dimensional IoT data. It can only be done by applying some appropriate data mining algorithms. These algorithms organize and transform the data into a structured form. To generate such structured information, majorly adaptive clustering techniques are employed in data mining. Hence, in the proposed work, the authors focus on generating algorithm which enhances the performance and compares the proposed fuzzy k-medoid clustering with the existing clustering algorithm pertaining to IoT data collected in intelligent real-time traffic system. First, the data streams are uploaded and then adaptive k-means clustering is applied to classify the data. Then k-medoid clustering algorithm is applied to the same data streams, and equivalent distance matrix is generated based on the centroid. After that, the proposed fuzzy k-medoid clustering algorithm is applied. The proposed work gives better performance than the existing ones. So with this approach it is easier to manage a huge IoT data. One can easily extract the information when it is in groups. Conceivable result of this proposed approach is: generating algorithm which gives better performance and also there is a comparison of clustering techniques that helps in deciding better clustering scheme for the given IoT dataset which results in optimal performance.

Differently from hierarchical clustering procedures, non-hierarchical clustering methods need the user to specify in advance the number of clusters; therefore, in this case, a single partition is obtained. The two most famous non-hierarchical clustering algorithms are the k-Means and the k-Medoids one. They differ in the definition of the cluster prototypes. In particular, the k-Means prototypes, called centroids, are defined to be the average values of units assigned to the clusters, while the k-Medoids prototypes, called medoids, identify the most representative observed units for each cluster. In this chapter, non-hierarchical clustering methods will be briefly introduced from a theoretical point of view and their implementation will be presented in detail by means of some real-life case studies.

70% of Indian population depends on farming; agriculture contributes 18% of GDP. According to government statistics 60% of crop production depends on monsoon rainfall. Hence, it is foremost important to understand the factors affecting the crop yield and there is need for development of prediction model for crop yield prediction. In this paper, we have taken Meteorological Data of Chhattisgarh (C.G.). Gathered crop production data in different districts of C.G. in last years, also collected rainfall in last years in different districts of C.G… We have proposed a machine learning model for crop yield prediction, in which dimension reduction algorithm applied to reduce the dimension of gathered data, it will suppress those data that will affect the prediction algorithm accuracy, K-medoid clustering algorithm has been applied to improve the prediction accuracy. Finally, performance of K-means clustering and K-medoid clustering is compared. For preprocessing of input dataset we have used PCA.

Due to the rapid technological changes, these days collaborative filtering-based recommender systems are being widely used worldwide. Collaborative filtering approach is more vulnerable from being attacked because of its open nature. The attackers may rate the fake ratings to disturb the systems. In this paper, unsupervised Kohonen Self-Organizing Map (KSOM) clustering technique is used to make a better detection between genuine and fake profiles to reduce profile injection attacks and compared with existing techniques Enhanced Clustering Large Applications Based on Randomized Search (ECLARANS) and Partition Around Medoids (PAM) with variants of attack size. It has been noticed that KSOM outperforms over ECLARANS and PAM techniques with good outlier accuracy.

This paper compares the performance of k-means and k-medoids in clustering objects with mixed variables. The k-means initially means for clustering objects with continuous variables as it uses Euclidean distance to compute distance between objects. While, k-medoids has been designed suitable for mixed type variables especially with PAM (partition around medoids). By using a mixed variables data set on a modified cancer data, we compared k-means and k-medoids on internal validity set up in R package. The result indicates that k-medoids is a good clustering option when the measured variables are mixed with different types.

Efficient utilization of energy for each sensor node is the key to prolong the network lifetime. The balance of energy consumption among the nodes located at central and edge areas also plays another important role in the situation of lifetime extension. In earlier work, LEACH algorithm has been investigated in the existing cluster head selection and the edge sub-clustering scheme is further developed to ease the effect of non-uniform node distribution on the edge, but it has not covered entire coverage area of the network due to failure of some nodes. The proposed work uses partition around medoid algorithm to improve optimization of number and size of sub-clusters along with the communication range for better improvement in cluster head selection and overcoming failure of edge nodes. Thus, it not only improves the energy conservation but also drastically improves the balance energy consumption, which can contribute to longer network lifetime and outage probability.