Text Simplification Using Transformer and BERT

Reading and writing are the main interaction methods with web content.Text simplification tools are helpful for people with cognitive impairments,new language learners,and children as they might find difficulties in understanding the complex web content.Text simplification is the process of changing complex text intomore readable and understandable text.The recent approaches to text simplification adopted the machine translation concept to learn simplification rules from a parallel corpus of complex and simple sentences.In this paper,we propose two models based on the transformer which is an encoder-decoder structure that achieves state-of-the-art(SOTA)results in machine translation.The training process for our model includes three steps:preprocessing the data using a subword tokenizer,training the model and optimizing it using the Adam optimizer,then using the model to decode the output.The first model uses the transformer only and the second model uses and integrates the Bidirectional Encoder Representations from Transformer(BERT)as encoder to enhance the training time and results.The performance of the proposed model using the transformerwas evaluated using the Bilingual Evaluation Understudy score(BLEU)and recorded(53.78)on the WikiSmall dataset.On the other hand,the experiment on the second model which is integrated with BERT shows that the validation loss decreased very fast compared with the model without the BERT.However,the BLEU score was small(44.54),which could be due to the size of the dataset so the model was overfitting and unable to generalize well.Therefore,in the future,the second model could involve experimenting with a larger dataset such as the WikiLarge.In addition,more analysis has been done on the model’s results and the used dataset using different evaluation metrics to understand their performance.

Analysis of a simplification strategy in a nonhydrostatic model for surface and internal wave problems

This paper examines the simplification strategy of retaining only the nonhydrostatic effect of local acceleration in a three-dimensional fully nonhydrostatic model regarding the submesoscale wave phenomenon in the ocean.Elaborate scale analysis of the vertical component of the Reynold-averaged Navier-Stokes(RANS)equation was performed,confirming the rationalization of this simplification.Then,the simplification was implemented in a RANS equation-based nonhydrostatic model NHWAVE(nonhydrostatic WAVE)to make a simplified nonhydrostatic model.Numerical examples were taken to test its performance,including surface sinusoidal waves propagating on an idealized East China Sea topography,tidally induced internal lee waves and small-scale solitary waves.The results show that in a considerably wide range of nonlinear strengths,the simplified nonhydrostatic model can obtain similar results as those in the fully nonhydrostatic model,even for smaller-scale solitary waves.Nonlinearity influences the applicability of the simplification.The stronger the nonlinearity is,the worse the simplified model describes the nonhydrostatic phenomenon.In general,the simplified nonhydrostatic model can simulate surface waves better than internal waves.Improvement of computational efficiency in the simplified nonhydrostatic model is reasonable,reducing the central processing unit time duration in the fully nonhydrostatic model by 16.4%–20.6%.The specially designed algorithm based on the simplified nonhydrostatic equation can remarkably reduce the computational time.

Numerical Simulation of the Settling Flux of Biodeposition in a Bay with Cage Culture Through Similarity Theory and a Simplified Pollution Source

The settling flux of biodeposition affects the environmental quality of cage culture areas and determines their environmental carrying capacity.Simple and effective simulation of the settling flux of biodeposition is extremely important for determining the spatial distribution of biodeposition.Theoretically,biodeposition in cage culture areas without specific emission rules can be simplified as point source pollution.Fluent is a fluid simulation software that can simulate the dispersion of particulate matter simply and efficiently.Based on the simplification of pollution sources and bays,the settling flux of biodeposition can be easily and effectively simulated by Fluent fluid software.In the present work,the feasibility of this method was evaluated by simulation of the settling flux of biodeposition in Maniao Bay,Hainan Province,China,and 20 sampling sites were selected for determining the settling fluxes.At sampling sites P1,P2,P3,P4,P5,Z1,Z2,Z3,Z4,A1,A2,A3,A4,B1,B2,C1,C2,C3 and C4,the measured settling fluxes of biodeposition were 26.02,15.78,10.77,58.16,6.57,72.17,12.37,12.11,106.64,150.96,22.59,11.41,18.03,7.90,19.23,7.06,11.84,5.19 and 2.57 g d^(−1)m^(−2),respectively.The simulated settling fluxes of biodeposition at the corresponding sites were 16.03,23.98,8.87,46.90,4.52,104.77,16.03,8.35,180.83,213.06,39.10,17.47,20.98,9.78,23.25,7.84,15.90,6.06 and 1.65 g d^(−1)m^(−2),respectively.There was a positive correlation between the simulated settling fluxes and measured ones(R=0.94,P=2.22×10^(−9)<0.05),which implies that the spatial differentiation of biodeposition flux was well simulated.Moreover,the posterior difference ratio of the simulation was 0.38,and the small error probability was 0.94,which means that the simulated results reached an acceptable level from the perspective of relative error.Thus,if nonpoint source pollution is simplified to point source pollution and open waters are simplified based on similarity theory,the setting flux of biodeposition in the open waters can be simply and effectively simulated by the fluid simulation software Fluent.

A Progressive Line Simplification Algorithm

An application of line simplification considering spatial knowledge is described.A method for identifying potential conflict regions,in order to avoid the self_intersection of generalized lines,is also discussed.Furthermore,a new progressive line simplification algorithm is presented.From the view point of spatial cognition,a spatial hierarchical structure is proposed,and its application to construction of spatial knowledge related to a line is explained.

A Collaborative Simplification Method for Multiple Coastlines Based on the Hierarchical Triangulation Network Partition

For the current automatic coastline generalization method,only one-line element is considered separately,but the relationship between the nearby elements is not effectively considered.A synergistic simplification method for multiple coastlines based on the hierarchical triangulation network partition(HTNP)is proposed in this paper.Firstly,the constrained Delaunay triangulation is constructed to partition the regions that can be simplified.Then,a hierarchical binary tree model to structure the morphological characteristics of the above several coastlines and the spatial proximity between different coastlines is constructed.Finally,the small curved and curved invisible parts of the coastline are deleted according to the visual constraints,and the narrow part between the coastline itself or the different coastlines is exaggerated appropriately,and the automatic simplification of the coastline is realized.The experimental results show that:①Relationships between the different coastlines are considered,and the shortcomings of considering the coastline separately are overcome;②Under the condition of the multiple coastlines in complex sea areas,the problem of collaborative simplification is solved,the quality of the coastlines are improved obviously,and the proposed method can be applied into more types of coastlines.

Simplification of English Sentence Pattern Structure in the Teaching of Translation and Grammar

In the teaching of translation and grammar,it is essential that the English sentence structure is analyzed and transformation of sentence structures both in English and Chinese in class and after class. However,many kinds of English sentences structures prevent the students who study English from getting the basic idea and concept of a sentence in translation or grammatical analysis. In this paper,the author submitted the basic mould of simplification of English sentence structure,and the students can learn and master the translating technique through the simplified English sentence structure. Thus it helps to improve students' competence of translation.

Complex Model Simplification Algorithm with Improved Error Evaluation

Given the framework of incremental mesh simplification based on edgecollapse, the paper proposes a mesh simplification algorithm using an improved approach formeasuring simplification error. The algorithm uses edge collapse to simplify the trianglemesh andmaintains surface error approximations using c-error for the faces which have changed after edgecollapse and d-crror for the faces which become degenerated after edge collapse. Also, we reportsome results using a variety of computer graphics models, which can show that the algorithm canachieve the desired simplification effect.

The Solid-liquid Transformation of Low-grade Solid Potash Deposit in Dalangtan Basin and the Simplification of the Liquid Phase System

We studied the solid-liquid transformation of low-grade solid potash deposit in Dalangtan Basin and simplified the liquid phase system.We did experiments to optimize conditions of the solid-liquid transformation.The Suitable

Terrain Simplification Research in Augmented Scene Modeling

Augmented reality is the merging of synthetic sensory information into a user's perception of a real environment. As one of the most important tasks in augmented scene modeling, terrain simplification research has gained more and more attention. In this paper, we mainly focus on point selection problem in terrain simplification using triangulated irregular network. Based on the analysis and comparison of traditional importance measures for each input point, we put forward a new importance measure based on local entropy. The results demonstrate that the local entropy criterion has a better performance than any traditional methods. In addition, it can effectively conquer the 'short-sight' problem associated with the traditional methods.

Application of equivalent resistance to simplification of Sutong Bridge piers in tidal river section modeling

This paper describes some details and procedural steps in the equivalent resistance （E-R） method for simplifying the pier group of the Sutong Bridge, which is located on the tidal reach of the lower Yangtze River, in Jiangsu Province. Using a two-dimensional tidal current numerical model, three different models were established： the non-bridge pier model, original bridge pier model, and simplified bridge pier model. The difference in hydrodynamic parameters, including water level, velocity, and diversion ratio, as well as time efficiency between these three models is discussed in detail. The results show that simplifying the pier group using the E-R method influences the water level and velocity near the piers, but has no influence on the diversion ratio of each cross-section of the Xuliujing reach located in the lower Yangtze River. Furthermore, the simplified bridge pier model takes half the calculation time that the original bridge pier model needs. Thus, it is concluded that the E-R method can be use to simplify bridge piers in tidal river section modeling reasonably and efficiently.

An Adaptive SVP Simplification Based on Area Difference

Sound velocity profile(SVP)data is indispensable in the multi-beam data processing.The sampling density is of great importance for SVP to represent the vertical variation of sound velocity accurately and guarantee the accuracy of sound ray tracing(SRT).However,the SVP also affects the SRT efficiency significantly,especially in deep-sea multi-beam sounding data processing.To improve SRT efficiency and ensure SRT accuracy,an adaptive SVP simplification method based on area difference is proposed in this article.Firstly,the relationship between the area difference of the raw SVP and the simplified one and SRT bias is studied,and the relationship model of them is built.Then,by considering the constraint of SRT accuracy,the SVP simplification method and the simplifying SVP procedure SVP are given.Finally,a deep water experiment is conducted to verify the proposed method.Compared to the existing method,the proposed method improves the robustness,feasibility of SVP simplification as well as the accuracy and efficiency of SRT.

Simplification and Experimental Verification for Temperature and Humidity Field Coupling Model of Conservatory Soil

Based on the surface energy balance model which is widely used abroad, a temperature and humidity field coupling model of conservatory soil without crop vegetation in full illumination was established. Considering the relatively closed environment in conservatory, weak solar radiation and little surface evaporation of soil, the daily variation of water content in different soil layers may be neglected, then the temperature and humidity field coupling model was simplified to a one-dimensional thermal diffusion model. The simplified model and the temperature and humidity field coupling model adopted the same computational method of soil physical parameters and discrete format of heat diffusion differential equations, and were applied to the continuous simulation of temperature field in conservatory soil without crop vegetation in full illumination. Through the comparison between simulation results and experimental data, the precision of the simplified model was verified. The typical rule of soil heat flux variation in a 24 h cycle was also obtained.

The Spatial Characteristics of a Line and Their Application to Line Simplification

This paper discusses the spatial knowledge related to a line ,and the characteristic points of lines is detected.According to the requirements of line generalization,new algorithms for identifying characteristic line points are presented.These characteristic points are used to improve the algorithms of line generalization.An algorithm for identifying bends is shown.In this paper,improved algorithms based on those by Douglas_Peucker,Visvalingam and Whyatt are shown.In this test,the progressive process of line generalization is emphasized.

城市实景模型结构化线面特征重构方法

为了改善城市实景模型边缘模糊,提出了一种曲率引导的结构化线面特征重构方法。根据曲率特征将网格分割为平面、可展凹、可展凸以及不可展曲面4类,在平面分割结果内提取平面,在可展凹和可展凸分割结果内提取直线,对过度弯曲的不可展区域进行保留,最终形成包含几何特征的复合网格模型。结果表明,结合曲率信息预先设置几何特征的潜在范围,使得结构化线面特征更可靠,同时保证城市实景中复杂的树结构不被错误地提取为平面。

桥墩受侧向撞击作用下静力简化分析方法

为了深入探究钢筋混凝土桥墩受侧向撞击时的静力简化方法,基于桥墩受撞击作用与静力作用下产生的侧向变形效应相等,建立了桥墩弯曲应变能和剪切应变能的计算公式.针对不同截面形式和不同约束条件下的桥墩模型,分别得出了桥墩在正弦、三角形和矩形撞击荷载波形作用下的等效撞击力简化公式,并分析了撞击力峰值和等效值之间的关系.结果表明:当矩形截面桥墩撞击点距墩底距离与截面长度的比值小于3.7,或圆形截面桥墩撞击点距墩底距离与截面半径的比值小于6.2时,需充分考虑桥墩剪切应变能的影响;荷载波形对等效撞击力取值影响较大,当撞击荷载波形接近正弦、三角形和矩形波时,等效撞击力系数的建议值分别为0.64、0.52和0.91.研究结果可为桥墩等效撞击力取值提供参考依据.

辽宁省日光温室蓝莓轻简化高效栽培技术

蓝莓是一种新兴的小浆果,因其较高的营养价值和保健功能,越来越受到人们重视和喜爱。辽宁省大部分地区位于北温带,日光温室蓝莓产业起步较早,设施蓝莓发展面积和果品质量在全国位居前列。近年,随着人口老龄化及城市化加剧,青壮劳动力不足、生产成本上涨等问题已成为制约辽宁温室蓝莓产业发展的瓶颈之一,传统的常规生产方式用工量多、劳动强度大、效益低等问题日益凸显。该文从机械化建园、智能温湿度调控、水肥一体化、设施抑草、蜜蜂授粉、简化修剪、病虫害防治、采收等方面进行详细介绍,总结辽宁温室蓝莓部分环节轻简化栽培措施,旨在为辽宁温室蓝莓产业高质量发展提供指导与借鉴。

采用混合策略联合优化的模糊C-均值聚类信息熵点云简化算法

针对传统聚类算法处理点云简化问题时精度低、耗时长且易丢失特征信息等问题,提出了一种基于动态精英自适应混合策略的鹈鹕算法(DEAMPOA)与加权熵法联合优化的模糊C-均值聚类(FCM)信息熵点云简化算法。采用动态自适应种群混合策略,同时融合了精英反向化思路,显著提升了鹈鹕优化算法(POA)的收敛趋势和全局寻优能力,提高了寻找FCM最优聚类中心的成功率;利用DEAMPOA结合加权熵法对FCM进行优化,提高鲁棒性的同时增强了搜索精度,得到较好的聚类结果;在8种UCI标准数据集上与4种算法对比进行聚类性能评估实验,验证了所提方法综合性能优越;将所提方法与信息熵融合,并应用在三维点云KITTI数据集简化中。实验结果表明:与包围框简化法、随机采样简化法和特征选择简化法对比,所提方法全局误差简化前后点集之间平均欧式距离(MED)指标分别降低了2.25%、6.93%、5.74%,点云简化效果最优且运行速度满足要求。

兼具求解速度与隐私性的复杂供热管网等值简化模型

电热能源系统模型维度高将会导致求解难度较大。为此,提出一种兼具求解速度与隐私性的复杂供热管网等值简化模型。首先,基于热网拓扑结构特点,提出可对供热管网进行灵活化简的供热管网拓扑简化方法,并根据简化后的管道结构和参数建立复杂供热管网等值简化模型。其次,充分考虑热电联产(combined heat and power,CHP)机组以热定电模式和热网热惯性对旋转备用容量的影响,建立了系统旋转备用模型。基于此,建立了考虑旋转备用约束的电热能源系统调度模型。最后,采用信息间隙决策理论(information gap decision theory, IGDT)处理系统成本与风电不确定度之间的关系。算例结果分析了供热管网拓扑简化方法在不同简化程度下的简化效果,验证了所提模型适用于电热能源系统优化调度,有效地减少优化模型的求解时间。

气井拟压力弹性二相法的推导、简化及应用

弹性二相法是评价气井控制原始地质储量的动态法,又称为拟稳态法(pseudo-steady-state method)或气藏范围测试法(reservoir limit testing method)。该法主要用于气井初期测试,评价岩性、断块和裂缝类型气藏井控的原始地质储量。由压力平方表示的弹性二相法,在中国从1994年起,连续4次被列入国家油气行业标准(SY/T 6098—1994,SY/T 6098—2000,SY/T 6098—2010,SY/T 6098—2022)。与拟压力表示的弹性二相法相比,由压力平方表示的弹性二相法是一种较好的近似评价方法。利用AL-HUSSAINY(1966)指出的拟压力关系式,经理论推导,得到了拟压力的弹性二相法,并应用WATTENBARGER(1968)对µgZ随p的变化关系研究,对拟压力弹性二相法进行了简化,分别得到了由压力一次方和压力平方表示的弹性二相法。目前,在国内外的文献中,尚看不到有关拟压力弹性二相法推导的报道。通过实例应用表明,压力一次方表示的弹性二相法评价的井控原始地质储量明显偏低,而由压力平方表示的弹性二相法评价的结果明显偏高。