The ten algorithm released the latest data scientists most often use the Sohu Technology Review: investigation ten from the KDnuggets algorithm in this paper, the algorithm of data engineers commonly used to rank, and the period of 2011-2016 were introduced. Based on the investigation of KDnuggets, summed up the ten most commonly used algorithm of data scientists, they are: 1 Regression regression algorithm 2 clustering algorithm Clustering 3 Decision Trees Rules 4 Visualization 5 k-Nearest decision tree visualization Neighbor neighbor algorithm (Principal Component Analysis) 6 PCA principal component analysis algorithm Statistics 8 Random Forests 7 statistical algorithm of random forest algorithm 9 Time series Sequence 10 Text time series of Mining text mining which uses an average of 8.1 of respondents said a survey in 2011 than the similar algorithm, can greatly improve the. Compared with a similar survey in 2011, we find that the most popular algorithms are regression algorithm, clustering algorithm, decision tree and visualization. Relatively speaking, the biggest growth is from (pct2016 pct2011 1) the following algorithm determination: Boosting, from 23.5% in 2011 to 32.8% in 2016, an increase of 40% text mining, from 2011 27.7% to 2016 35.9%, an increase of 30% visualization, from 38.3% in 2011 to 48.7% in 2016, an increase of 27% time series. From 2011 29.6% to 2016 37%, an increase of 25% abnormal deviation detection, from 2011 16.4% to 2016 19.5%, an increase of 19% integrated methods, from 28.3% in 2011 to 33.6% in 2016, an increase of 19% support vector machine, from 2011 28.6% to 2016 33.6%, 18% year-on-year growth in 2011 from the regression algorithm. 57.9% to 201 6 years, an increase of 16%, in addition, in 2016 the most popular new algorithms are: K- nearest neighbor, the 46% principal component analysis, the 43% random forest algorithm, the 38% optimization, K相关的主题文章: