Three Common Machine Learning Algorithms Neither Enhance Prediction Accuracy Nor Reduce Spatial Autocorrelation in Residuals: An Analysis of Twenty-five Socioeconomic Data Sets

Machine learning (ML) is being applied in an increasing volume of geographical research. However, the aspects of spatial autocorrelation (SAC) in the residuals produced by ML models have been understudied compared to the benefit of ML, namely, reduction of prediction errors. In this study, we examined the relationship between predictive accuracy and the reduction in the residual SAC for 597 variables from 25 geographical socio-economic data sets using spatial and nonspatial cross-validation of three ML algorithms such as random forests, support vector machine, and artificial neural network (ANN) to provide an extensive empirical diagnosis—but not a definitive theory—of the relationship between SAC and ML. Our results highlighted that the ML algorithms with tuned hyperparameters yielded marginal predictive accuracy gains and the minimal decreases in residual SAC. ANN revealed lower accuracy and higher reduction in the residual SAC than others. This implies ML algorithms in geographical research in socio-economic domains would not always result in higher prediction accuracy. We suggest that ML in geographical research should be cautiously employed when the main objective is related to the residual SAC. We also showed that spatial cross-validation neither improves predictive accuracy substantially nor reduce the residual SAC effectively.     

从红山文化女神头像谈中国古代唇妆艺术的产生

唇妆,在古代又称点唇,今天的口红在古代则称唇脂,这在东汉刘熙《释名．释首饰》中便有记载,“唇脂,以丹作之,象唇赤也”。隋唐以后又有称“口脂”者,如韦庄《江城子》“朱唇动,先觉口脂香”。用唇脂涂唇在古代又称“点唇术”,可以增加口唇的鲜艳光泽,并给人以健康、年轻、充满活力的印象,所以自古以来点唇术便备受众多     

Plant exploitation of the last foragers at Shizitan in the Middle Yellow River Valley China: evidence from grinding stones

China is one of the few centers in the world where plant domestication evolved independently, but its developmental trajectory is poorly understood. This is because there is considerably less data from the terminal Pleistocene and early Holocene documented in China than in other regions, such as the Near East and Mesoamerica, and previous studies on Paleolithic subsistence in China have largely focused on animal hunting rather than plant gathering. To resolve these problems the current research investigated the range of plants used by late Paleolithic hunter–gatherers in the middle Yellow River region where some of the earliest farmers emerged. We employed usewear and starch analyses on grinding stones to recover evidence for plant use in a hunting–gathering population at a late Paleolithic site, Shizitan Locality 9 in Jixian, Shanxi (ca. 13,800–8500 cal. BP). The usewear analysis shows that all artifacts preserved a range of usewear patterns best matching multiple tasks and indicating multi-functional use. Starch remains recovered from these tools indicate that the Shizitan people collected and processed many types of grass seeds (Panicoideae and Pooideae subfamilies), acorns (Quercus sp.), beans (Phaseoleae tribe) and yams (Dioscorea sp.). The Shizitan people represented some of the last hunter–gatherers in the middle Yellow River region. Their broad spectrum subsistence strategy was apparently carried on by the first Neolithic farmers in the same region, who collected similar wild plants and eventually domesticated millets. The trajectory from intensified collection of a wide range of wild plants to domestication of a small number of species was a very long process in north China. This parallels the transition from the “broad spectrum revolution” to agriculture in the Near East.