Women’s cancers in China: a spatio-temporal epidemiology analysis

Background Women's cancers, represented by breast and gynecologic cancers, are emerging as a significant threat to women's health, while previous studies paid little attention to the spatial distribution of women's cancers. This study aims to conduct a spatio-temporal epidemiology analysis on breast, cervical and ovarian cancers in China, thus visualizing and comparing their epidemiologic trends and spatio-temporal changing patterns. Methods Data on the incidence and mortality of women’s cancers between January 2010 and December 2015 were obtained from the National Cancer Registry Annual Report. Linear tests and bar charts were used to visualize and compare the epidemiologic trends. Two complementary spatial statistics (Moran’s I statistics and Kulldorff’s space–time scan statistics) were adopted to identify the spatial–temporal clusters. Results The results showed that the incidence and mortality of breast cancer displayed slow upward trends, while that of cervical cancer increase dramatically, and the mortality of ovarian cancer also showed a fast increasing trend. Significant differences were detected in incidence and mortality of breast, cervical and ovarian cancer across east, central and west China. The average incidence of breast cancer displayed a high-high cluster feature in part of north and east China, and the opposite traits occurred in southwest China. In the meantime, the average incidence and mortality of cervical cancer in central China revealed a high-high cluster feature, and that of ovarian cancer in northern China displayed a high-high cluster feature. Besides, the anomalous clusters were also detected based on the space–time scan statistics. Conclusion Regional differences were detected in the distribution of women’s cancers in China. An effective response requires a package of coordinated actions that vary across localities regarding the spatio-temporal epidemics and local conditions.

Method of Feature Reduction in Short Text Classification Based on Feature Clustering

One decisive problem of short text classification is the serious dimensional disaster when utilizing a statistics-based approach to construct vector spaces. Here, a feature reduction method is proposed that is based on two-stage feature clustering (TSFC), which is applied to short text classification. Features are semi-loosely clustered by combining spectral clustering with a graph traversal algorithm. Next, intra-cluster feature screening rules are designed to remove outlier feature words, which improves the effect of similar feature clusters. We classify short texts with corresponding similar feature clusters instead of original feature words. Similar feature clusters replace feature words, and the dimension of vector space is significantly reduced. Several classifiers are utilized to evaluate the effectiveness of this method. The results show that the method largely resolves the dimensional disaster and it can significantly improve the accuracy of short text classification.