Event extraction, which aims to identify event triggers of pre-defined event types and their arguments of specific roles, is a challenging task in NLP. Most traditional approaches formulate this task as classification problems, with event types or argument roles taken as golden labels. Such approaches fail to model rich interactions among event types and arguments of different roles, and cannot generalize to new types or roles. This work proposes a new paradigm that formulates event extraction as multi-turn question answering. Our approach, MQAEE, casts the extraction task into a series of reading comprehension problems, by which it extracts triggers and arguments successively from a given sentence. A history answer embedding strategy is further adopted to model question answering history in the multi-turn process. By this new formulation, MQAEE makes full use of dependency among arguments and event types, and generalizes well to new types with new argument roles. Empirical results on ACE 2005 shows that MQAEE outperforms current state-of-the-art, pushing the final F1 of argument extraction to 53.4% (+2.0%). And it also has a good generalization ability, achieving competitive performance on 13 new event types even if trained only with a few samples of them.
A new concept dealing with digital analysis of loess terrain, slope spectrum, is presented and discussed in this paper, by introducing its characteristic, representation and extracting method from DEMs. Using 48 geomorphological units in different parts of the loess as test areas and 5 m-resolution DEMs as original test data, the quantitative depiction and spatial distribution of slope spectrum in China's Loess Plateau have been studied on the basis of a series of carefully-designed experiments. In addition, initial experiment indicates a strong relationship between the slope spectrum and the loess landform types, displaying a potential importance of the slope spectrum in geomorphological studies. Based on the slope spectrums derived from the 25 m-resolution DEM data in whole loess terrain in northern part of Shaanxi, 13 slope spectrum indices were extracted and integrated into a comprehensive layer with image integration method. Based on that, a series of unsupervised classifications was applied in order to make a landform classification in northern Shaanxi Loess Plateau. Experimental results show that the slope spectrum analysis is an effective method in revealing the macro landform features. A continuous change of slope spectrum from south to north in northern Shaanxi Loess Plateau shows an obvious spatial distribution of different loess landforms. This also proves the great significance of the slope spectrum method in describing the terrain roughness and landform evolution as well as a further understanding on landform genesis and spatial distribution rule of different landforms in the Loess Plateau.Loess Plateau, slope spectrum, slope, DEM, loess landform
Abstract:The Chinese Loess Plateau suffers from serious gully erosion induced by natural and human causes. Gully-affected areas detection is the basic work in this region for gully erosion assessment and monitoring. For the first time, an unmanned aerial vehicle (UAV) was applied to extract gully features in this region. Two typical catchments in Changwu and Ansai were selected to represent loess tableland and loess hilly regions, respectively. A high-powered quadrocopter (md4-1000) equipped with a non-metric camera was used for image acquisition. InPho and MapMatrix were applied for semi-automatic workflow including aerial triangulation and model generation. Based on the stereo-imaging and the ground control points, the highly detailed digital elevation models (DEMs) and ortho-mosaics were generated. Subsequently, an object-based approach combined with the random forest classifier was designed to detect gully-affected areas. Two experiments were conducted to investigate the influences of segmentation strategy and feature selection. Results showed that vertical and horizontal root-mean-square errors were below 0.5 and 0.2 m, respectively, which were ideal for the Loess Plateau region. The overall extraction accuracy in Changwu and Ansai achieved was 84.62% and 86.46%, respectively, which indicated the potential of the proposed workflow for extracting gully features. This study demonstrated that UAV can bridge the gap between field measurement and satellite-based remote sensing, obtaining a balance in resolution and efficiency for catchment-scale gully erosion research.
Digital terrain analysis (DTA) is one of the most important contents in the research of geographical information science (GIS). However, on the basis of the digital elevation model (DEM), many problems exist in the current research of DTA in geomorphological studies. For instance, the current DTA research appears to be focused more on morphology, phenomenon, and modern surface rather than mechanism, process, and underlying terrain. The current DTA research needs to be urgently transformed from the study of landform morphology to one focusing on landform process and mechanism. On this basis, this study summarizes the current research status of geomorphology-oriented DTA and systematically reviews and analyzes the research about the knowledge of geomorphological ontology, terrain modeling, terrain derivative calculation, and terrain analytical methods. With the help of DEM data, DTA research has the advantage of carrying out geomorphological studies from the perspective of surface morphology. However, the study of DTA has inherent defects in terms of data expression and analytic patterns. Thus, breakthroughs in basic theories and key technologies are necessary. Moreover, scholars need to realize that DTA research must be transformed from phenomenon to mechanism, from morphology to process, and from terrain to landform. At present, the research development of earth science has reached the critical stage in which the DTA research should focus more on geomorphological ontology. Consequently, this study proposes several prospects of geomorphology-oriented DTA from the aspects of value-added DEM data model, terrain derivatives and their spatial relations, and macro-terrain analysis. The study of DTA based on DEM is at a critical period along with the issue on whether the current GIS technology can truly support the development of geography. The research idea of geomorphology-oriented DTA is expected to be an important exploration and practice in the field of GIS.
Slope spectrum has been proved to be a significant methodology in revealing geomorphological features in the study of Chinese loess terrain. The determination of critical areas in deriving slope spectra is an indispensable task. Along with the increase in the size of the study area, the derived spectra are becoming more and more alike, such that their differences can be ignored in favor of a standard. Subsequently, the test size is defined as the Slope Spectrum Critical Area (SSCA). SSCA is not only the foundation of the slope spectrum calculation but also, to some extent, a reflection of geomorphological development of loess relief. High resolution DEMs are important in extracting the slope spectrum. A set of 48 DEMs with different landform areas of the Loess Plateau in northern Shaanxi province was selected for the experiment. The spatial distribution of SSCA is investigated with a geo-statistical analysis method, resulting in values ranging from 6.18 km 2 to 35.1 km 2 . Primary experimental results show that the spatial distribution of SSCA is correlated with the spatial distribution of the soil erosion intensity, to a certain extent reflecting the terrain complexity. The critical area of the slope spectrum presents a spatial variation trend of weak-strong-weak from north to south. Four terrain parameters, gully density, slope skewness, terrain driving force (T d ) and slope of slope (SOS), were chosen as indicators. There exists a good exponential function relationship between SSCA and gully density, terrain driving force (T d ) and SOS and a logarithmic function relationship between SSCA and slope skewness. Slope skewness increases, and gully density, terrain driving force and SOS decrease with increasing SSCA. SSCA can be utilized as a discriminating factor to identify loess landforms, in that spatial distributions of SSCA and the evolution of loess landforms are correlative. Following the evolution of a loess landform from tableland to gully-hilly region, this also proves that SSCA can represent the development degree of local landforms. The critical stable regions of the Loess Plateau represent the degree of development of loess landforms. Its chief significance is that the perception of stable areas can be used to determine the minimal geographical unit.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.