Conventional video summarization approaches based on reinforcement learning have the problem that reward can only be received after whole summary is generated. Such kind of sparse and it makes hard to converge. Another labelling each shot tedious costly, which usually prohibits construction large-scale datasets. To solve these problems, we propose a weakly supervised hierarchical framework, decomposes task into several subtasks enhance quality. This framework consists manager network worker...

Introduction The treatment cost expenditure for family members with illness is a significant factor exacerbating poverty. Using nationally representative data concerning the migrant population, this study analyzes impact of economic burden from members’ on poverty among young workers. Methods Firstly, we construct multidimensional index system based five dimensions: income, education, employment, health, and livelihood,and used A-F method calculation. Moreover, diverging existing studies...

Abstract Based on China’s new development philosophy, and following the idea of “assessing overall levels firstly, then anchoring subsystems contributions finally identifying obstacle factors”, an evaluation system high-quality (HQD) is constructed from five innovative (ID), coordinated (CD), green (GD), open (OD) shared (SD). In this paper, entropy weight TOPSIS model used to measure HQD 19 provincial units located in Yellow River Basin (YRB) Yangtze Economic Belt (YREB) 2010–2019, with...

Poverty is a critical social problem in numerous countries. It result of many aspects and has been addressed worldwide for long time. In this study, we construct the multidimensional poverty index (MPI) migrant workers China based on Amartya Sen’s capabilities approach. Using Chinese Family Panel Studies (CFPS) data during 2014–2020 Alkire–Foster methodology, our study examines using static to dynamic analyses. The results show following: (1) shows general decreasing trend over (2) Over one...

A growing body of work has been devoted to studying teachers’ digital competence, but not much is known about how competence affects online teaching behavior. Guided by the technological pedagogical content knowledge (TPACK) framework, this study investigated behavior based on Chinese primary and secondary teachers. total 1833 teachers completed self-report scales measuring behavior, use intention teaching, students’ learning difficulties using platform Questionnaire Star. The results...

Introduction Reducing carbon emissions from agriculture is essential to ensuring food security and human prosperity. As a country with approximately 20% of the global population, China has begun actively practicing low-carbon agricultural development conception. Against backdrop disruptive technologies that continue be integrated into various industries, massive application robots opened way intelligent agriculture. This paper tries answer whether there some non-linear nexus between in...

The purpose of water resources carrying capacity (WRCC) research is to meet the sustainable utilization and realize its development strategy. Scientifically evaluating regional extremely important. In this study, a variable precondition S-type cloud algorithm (VPSCA) developed for assessment WRCC based on social economic subsystem, ecological environmental subsystem including thirteen indicators, which can overcome shortcoming that existing evaluation methods rarely consider fuzziness,...

Abstract Image inpainting is the task to fill missing regions of an image. Recently, researchers have achieved a great performance by using convolutional neural networks (CNNs) with conventional patch-matching method. Existing methods compute attention scores, which are based on similarity patches between known and regions. Considering that at different spatial positions can convey levels detail, we propose spatially adaptive multi-scale score uses scales scores for each pixel positions....

Abstract This article expounds on the challenges of sustainable global energy development to conventional econometrics in research. It introduces big data's “4V” characteristics, that is, volume, variety, velocity, and veracity. Further, presents a series feature brings research issues. Then, by reviewing existing literature, proposes improvement promotion three directions data exploration variable generation, prediction, causal inference machine learning data. Finally, concludes discipline...

This paper made improvements on clustering by fast search and find of density peaks (CFSFDP) algorithm extended this to fuzzy numbers (FN-CFSFDP algorithm). Using FN-CFSFDP algorithm, classical information included in the samples are sets, can be cl ustered searching peak. Firstly, means error analysis, improved Euclidean distance between was defined, some key parameters or operating quantities mainly including cut-off Gaussian Kernel function were introduced detail. Next, 76 random...

If we endow an intelligent system with fuzzy logic, hope that it can deal data, including the clustering of data. This paper proposes a mixed data algorithm by fast search and find density peaks (FMTD-CFSFDP), which is development CFSFDP lgorithm. The proposed kind density-based method established using sets for Mathematical definitions are presented. Combined definition traditional Euclidean distance, defined improved distance both continuous discrete smaller error. On this basis, weight...

With development of the infrastructures supporting smart home which has entered in a new stage featured by intelligent services.The services based on natural language instructions are aimed at simplifying and improving our lives.To customize develop these more efficiently, this paper proposes an approach to model execute runtime introduces knowledge graph into process.Firstly, concept is introduced for services.Secondly, we put forward mechanism construct instance services.Finally, rules...

For a continuous random variable in real number field, there must be distribution and also probability density function of this variable. If is known with as independent variable, its image smooth or piecewise line, at least one that takes these functions are bounded on the first function. Any conduct line integral operation to segment arc length certain cumulative interval corresponding section for operation. A general designation linear variables. Conduct have same results By way, Line...

<p><span style="font-family: Times New Roman; font-size: medium;">The random complex variable is a kind of which in number as the variable. In this paper, study on mathematical conjecture variable, two-dimensional joint distribution and expected, variance other common statistics real imaginary part variables satisfy independent standard normal condition.</span></p>

The extended Cobb-Douglas production function is used based on the transition probabilities matrix with reflecting wall and absorbable determined by non-homogeneous Markov chain. According to current number ratio of various types enterprises under different carbon emission conditions, after multi-stage unlocking evolution transfer process determined. And finally economic output conditions corresponding valuation model are

The extended Cobb-Douglas production function is used based on the transition probabilities matrix with reflecting wall and absorbable determined by non-homogeneous Markov chain. According to current number ratio of various types enterprises under different carbon emission conditions, after multi-stage unlocking evolution transfer process determined. And finally economic output conditions corresponding valuation model are

The economic ethics is an abstract variable in system, and there a correlation between the sustainable growth. First of all, existing researches are discussed from macro perspective. On this basis, set system constructed. Afterwards, micro perspective, necessary conditions for growth economics obtained through positive negative benefit functions three dimensions government ethics, enterprise personal under condition whether to abide by law ethics. Finally, whole analyzed conclusion that...