A New Method for Clustering Based on Development of Imperialist Competitive Algorithm

Clustering is one of the most widely used data mining techniques that can be used to create homogeneous clusters.K-means is one of the popular clustering algorithms that,despite its inherent simplicity,has also some major problems.One way to resolve these problems and improve the k-means algorithm is the use of evolutionary algorithms in clustering.In this study,the Imperialist Competitive Algorithm(ICA) is developed and then used in the clustering process.Clustering of IRIS,Wine and CMC datasets using developed ICA and comparing them with the results of clustering by the original ICA,GA and PSO algorithms,demonstrate the improvement of Imperialist competitive algorithm.

A Comparison of Atmospheric Temperature over China between Radiosonde Observations and Multiple Reanalysis Datasets

The quality controlled（RAW） and homogenized（ADJ） radiosonde temperatures within 850-30 hPa collected at 118 stations in China are compared,on a monthly mean basis,with the temperatures extracted from 8 reanalysis datasets（REA） including NCEP-1,NCEP-2,ERA-40（ECMWF 45-yr Reanalysis）,ERAInterim,JRA-55（Japanese 55-yr Reanalysis）,20CR（20th Century Reanalysis）,MERRA（Modern Era Retrospective-Analysis）,and CFSR（Climate Forecast System Reanalysis）.Average differences,correlations,standard deviations,and linear trends among RAW,ADJ,and REA for the period 1981-2010 are analyzed.The results reveal significant inhomogeneity in the time series of RAW radiosonde temperature in China;an overall negative adjustment was thus employed to obtain the ADJ temperatures,and the effect of the negative adjustment is the most significant within 200-100 hPa.Such a homogenization process has removed the system errors in RAW,possibly caused by radiosonde instrument changes and observation system upgrades.Hence,the correlation is higher between ADJ and REA than that between RAW and REA.The mean difference between ADJ and REA is about 1℃ during 1981-2010,while REA are mostly cooler in the troposphere and warmer in the stratosphere than ADJ;nonetheless,they have a significant high and positive correlation and their annual variability is notably consistent.Furthermore,the linear trends in REA and ADJ both demonstrate warming in the lower-mid troposphere and cooling in the mid stratosphere,with large uncertainties found in the upper troposphere and lower stratosphere.In general,ERA-Interim,JRA-55,and MERRA are more consistent with ADJ than other reanalysis datasets.