Data clustering is an important field of machine learning having applicability in wide areas, like, business analysis, manufacturing, energy, healthcare, and travel and logistics. Since no clustering method can address every clustering problem, a number of clustering algorithms with a variety of applications have been developed. Data clustering approaches based on self-organizing map (SOM) generally use the map dimensions (of grid) ranging from 2 × 2 (4 neurons) to 8 × 8 (64 neurons) without any explicit reason for using the particular dimension. Also these dimensions of SOM are not capable of giving optimized results because they lack concrete idea or logic behind using the particular dimension. Further, the number of neurons (microclusters) in the above approach is extending from 4 to 64, or any arbitrary value. Therefore these algorithms have to again use some secondary approaches to map these microclusters into the lower dimension (actual number of clusters) like 2, 3, or 4, as the case may be, based on the optimum number of clusters in the specific data set. Additionally, it is discovered that this secondary approach, observed in most of the works, is not SOM and, is an algorithm, like, cut tree or the other. In the current article, the proposed approach is elucidating a logic that will give an idea to select the most optimized dimension of SOM for the given data set. Then this optimum dimension is again clustered into the lower actual dimension. Also, primary and secondary, both approaches utilize the SOM itself to clusterize the data. It is also discovered that the weight matrix of SOM is useful in finding the best suitable dimension for the specific data set. It is to be noted that the optimized two-dimensional configuration of SOM is not the same for every data set, and this work also tries to discover this configuration. The current work proves that SOM is superior to any other clustering approaches, like, k-means or the other, and could be used successfully to clusterize all types of data sets. Seven data sets from diverse domains like medical, biological, and chemical are tested in this work, including the synthetic data sets. The adjusted randomized index obtained on the Iris, Wine, Wisconsin diagnostic breast cancer, New Thyroid, Seeds, A1, and Imbalance is, respectively, 0.7173, 0.9134, 0.7543, 0.8041, 0.7781, 0.8907, and 0.8755, which outperforms all other results on each of the data sets available on the web. The main aim of the current work is to explore the data of the medical domain but to clarify our approach, the algorithm is tested on the data of nonmedical domain and observed excellent results on them also.

The role of community pharmacists and their perception toward antimicrobial stewardship in Baghdad, Iraq.

Based on the largest cohort of the longevous Chinese, the present study comprehensively evaluated the impacts of main contributors of socioeconomic support on the quality of life and prognosis of frailty. The prefrail and frail population were vulnerable to socioeconomic support and more sensitive to it than the robust population. Continuing to work at the advanced age was the risk factor of low quality of life but was the protective factor of worsening of prefrailty and survival. In sum, the present study provides evidence about socioeconomic support and prognosis of frailty for the clinical guidelines and policy makers. Frailty-preferential policies on financial and social support and medical insurance should be developed.