a b s t r a c tThis article aims to evaluate the impact of urbanization and different urbanization modes on cultivated land changes using an econometric model that incorporates socio-economic and policy factors in the eastern China, which experience the great urbanization in recent years. Based on land-use remote sensing data interpreted from Landsat Thematic Mapper/Enhanced Thematic Mapper digital images of Chinese Academy of Sciences and a unique set of socio-economic data, an econometric model is developed to empirically estimate the impacts on cultivated land changes. Although urbanization has an effect on the changes of cultivated land, its effect is marginal. Moreover, the expansion of built-up areas in different urbanization modes causes varying impacts on changes in cultivated land use in different regions. Assuming that other factors remain constant, compared with the expansion of villages or the development of small towns, in the periods of 1995-2000, the urbanization in the more developed eastern region alleviates the loss of cultivated land by 7%, while during 2000-2008 the rapid urbanization lead to the cultivated land loss increase by 29.2%. The policies designed to protect cultivated land by encouraging people move to small towns may actually accelerate the occupation of cultivated land.

While the recent tree-based neural models have demonstrated promising results in generating solution expression for the math word problem (MWP), most of these models do not capture the relationships and order information among the quantities well. This results in poor quantity representations and incorrect solution expressions. In this paper, we propose Graph2Tree, a novel deep learning architecture that combines the merits of the graph-based encoder and tree-based decoder to generate better solution expressions. Included in our Graph2Tree framework are two graphs, namely the Quantity Cell Graph and Quantity Comparison Graph, which are designed to address limitations of existing methods by effectively representing the relationships and order information among the quantities in MWPs. We conduct extensive experiments on two available datasets. Our experiment results show that Graph2Tree outperforms the state-of-the-art baselines on two benchmark datasets significantly. We also discuss case studies and empirically examine Graph2Tree's effectiveness in translating the MWP text into solution expressions 1 .

This paper documents the changes in China's Hukou reform before and after 2014 based on a unique data set of Hukou policy documents from Chinese cities between 2000 and 2016. We construct a Hukou registration index to measure the stringency of local Hukou qualification in Chinese cities. There are four main channels for migrants to get local urban Hukou: investment, home purchase, talent program, and employment. The requirements of the four channels have large variations across different tiers of cities between the two periods of 2000–2013 and 2014–2016. First‐tier and some second‐tier cities set high criteria for local Hukou registration that have become more stringent over time, while other cities have much lower requirements. The point account system for Hukou registration shows that cities have different preferences over workers eligible for local urban Hukou. The quantitative measures developed in this paper can be used to study a variety of topics on the social and economic consequences of labor mobility barriers.

The design of automatic solvers to arithmetic math word problems has attracted considerable attention in recent years and a large number of datasets and methods have been published. Among them, Math23K is the largest data corpus that is very helpful to evaluate the generality and robustness of a proposed solution. The best performer in Math23K is a seq2seq model based on LSTM to generate the math expression. However, the model suffers from performance degradation in large space of target expressions. In this paper, we propose a template-based solution based on recursive neural network for math expression construction. More specifically, we first apply a seq2seq model to predict a tree-structure template, with inferred numbers as leaf nodes and unknown operators as inner nodes. Then, we design a recursive neural network to encode the quantity with Bi-LSTM and self attention, and infer the unknown operator nodes in a bottom-up manner. The experimental results clearly establish the superiority of our new framework as we improve the accuracy by a wide margin in two of the largest datasets, i.e., from 58.1% to 66.9% in Math23K and from 62.8% to 66.8% in MAWPS.