IEIS 2022

The research topic of this paper mainly focuses on the problem of communication and trust in Chinese enterprises’ telecommuting under the situation of crisis. Through semi-structured interviews with employees of 32 different enterprises during the epidemic isolation period, 290000 Chinese words of text interview data were obtained. With the help of manual analysis and software NVIVO, the grounded theory method was used for multi-stage coding and relationship analysis. We found that unlike virtual teams and temporary organizations, telecommuting required in special environments relies more on past experience, common experience and commitment to tasks in terms of trust.

Despite the continuous expansion of various emerging distribution channels, the agricultural products wholesale market is still the main channel for the distribution of agricultural products in China. But the agricultural products wholesale market is also the prone and hardest hit area for food safety problems in the circulation of agricultural products. The article first reviews the relevant research of other scholars, then analyzes and summarizes the current problems of food safety supervision in the agricultural products wholesale market in China, and constructs a two-level game model of the government department to the wholesale market and the wholesale market to the wholesaler. We find that strengthening the supervision and punishment of the government department and the wholesale market, and reducing supervision costs will improve the level of food safety in agricultural products wholesale markets. Finally, based on the actual situation, the article puts forward policy recommendations on food safety supervision in agricultural products wholesale markets.

Since the web 3.0 era, everyone can be the source of information dissemination, but with the gradual saturation of the self-publishing market, the video style, copy content, and scene layout of individual creators are directly “copied and pasted” by infringers, and the infringement phenomenon is increasing year by year. Based on the analysis of the characteristics of blockchain technology, this paper aims to analyze the difficulties encountered by self-publishing copyright, identify blockchain as the key to copyright storage and protection, and finally reflect on blockchain technology, which will bring challenges to self-publishing dissemination and regulation.

This paper studies a novel tool for describing fuzzy information, called linguistic Fermatean fuzzy sets (LFFSs), in the process of multi-attribute decision-making (MADM). Compared to linguistic intuitionistic fuzzy sets and linguistic Pythagorean fuzzy sets, our LFFSs are more flexible and can depict more complicated decision-making information then the former two. In this study, we first introduce the notion of LFFSs. Afterwards, some other related concepts, such as operational rules, ranking methods as well as distance measure are interpreted. When considering aggregation operators for linguistic Fermatean fuzzy information, we generalize the classical power average (PA) operator into LFFSs and introduce the linguistic Fermatean fuzzy power average operator and its weighted form. Subsequently, a new MADM method based on LFFSs and their aggregation operator is developed. At last, an illustrative example is provided to show how our proposed method can be applied in solving realistic MADM problems.

Under the new development pattern, this article modifies the index weights of the analytic hierarchy method using the entropy weight method and studies the evaluation system to ensure the safety of China’s environmental industry. The results show that from 2014 to 2016, the range of environmental and industrial safety scores in China is 33.4–37.2, and its status is unsafe, but the overall situation continues to improve. The competitiveness of the environmental industry in the international market is improving, greatly increasing the viability of the environmental industry. Independent innovation has made great breakthroughs, showing an obvious improvement trend from 2016 to 2018. In 2017–2021, China's environmental industrial safety score was 41.8–56.7, which improved the industrial safety status and showed a general improvement trend. Foreign-funded enterprises have accounted for a large proportion of China's environmental industry; the environmental industry will continue this trend for a long time, and China should speed up the environmental industry development process. The state should formulate relevant policies in terms of the financial environment, market access, foreign mergers, acquisitions, etc., to ensure China’s control over important environmental industries, improve the competitiveness of environmental industries, strengthen the control of important environmental industries and achieve environmental industry safety.

The development of the Internet has made digital publishing an inevitable trend for the transformation and development of the publishing industry, and with it comes a series of problems: easy access within, easy copying and long time to confirm the rights, high transaction and maintenance costs, which makes the copyright issue always a bottleneck for the development of the digital publishing industry. This paper attempts to explore the application path of copyright deposit, copyright review, copyright maintenance, copyright transaction and copyright supervision from the features of distributed ledger and impenetrability of blockchain technology to promote the high-quality development of China's digital publishing industry.

Rail accidents cause casualty and financial loss to society. In order to extract and identify the key factors from the accident reports more accurately, this study added the word frequency-correlation importance evaluation function(TF-K*) based on complex network on the basis of text mining, and built an importance evaluation model of factors for the railway accidents. When evaluating the importance of factors, the word frequency and the correlation between factors can be considered simultaneously. In this study, 213 railway accident reports from China and Britain were collected to analyze the cause of the accident, and the final results also verified the validity of the model.

Private tutoring industry is a supplement to national basic education. The service provided by this industry is to strengthen the test-taking ability of customers and provide knowledge supply service in line with the examination selection system. By analyzing the characteristics of education industry and the purpose of industry capitalization, and combining with the empirical data of uneven distribution of educational resources in China, this paper illustrates the essential contradiction between private tutoring industry and industry capitalization.

Audio-driven 2D avatar video generation algorithms have a wide range of applications in the media field. The technology of generating 2D avatar videos with only the input of compliant audio and images has been a positive boost to the development of online media and other fields. In such generation algorithms, the accurate coupling of speech audio and appearance changes such as faces and gestures in subtle movements has been a point of continuous improvement, with appearance changes moving from an early focus on matching speech content only to starting to incorporate human emotions expressed by speech. There has been a significant improvement in fidelity and synchronization compared to the early experimental results of the study, and the behavioral performance of the 2D avatars in the generated videos is getting closer to that of humans. This paper provides an overview of existing audio-driven 2D avatar generation algorithms and classifies their tasks into two categories: talking face generation and co-speech gesture generation. Firstly, the article describes the task specifically and describes its application areas. Secondly, we analyze the core algorithms in order of technological advancement and briefly describe the performance effects of the methods or models. Thirdly, we present common datasets for both types of tasks as well as evaluation metrics and compare the performance metrics of some recently proposed algorithms. Finally, the paper discusses the opportunities and challenges faced by the field and gives future research directions.

Starting from the questionnaire survey and statistical analysis of the current situation of postgraduate entrance examination and professional learning in colleges and universities, combined with the policy of training talents in higher education in my country and the background of the integration and development of intelligent media, this paper comprehensively analyzes the current situation between postgraduate entrance examination and professional learning in colleges and universities. It is difficult to take into account the problems of professional learning, ignoring the opportunities for undergraduate teaching practice, the dilemma of choice after failing the postgraduate entrance examination, and the serious mental health problems of students. It also proposes a smart teaching model for postgraduate entrance examination and undergraduate study [1], in order to better serve postgraduate entrance examination students and colleges and universities, improve the quality of undergraduate teaching classrooms, and provide decision-making basis for the improvement of school-running quality.

In recent years, in the field of natural language processing, significant progress has been made in text generation. Text generation has gained widespread popularity in many fields such as abstract extraction, poetry creation, and response to social network comments. Given the excellent generative capabilities of Generative Adversarial Networks (GAN), it is often used as main model for text generation with remarkable results. This review aims to provide the core tasks of generative adversarial network text generation and the architecture used to deal with these tasks, and draw attention to the challenges in text generation with generative adversarial network. Firstly, we outline the mainstream text generation models, and then introduce datasets, advanced models and challenges of text generation tasks in detail. Finally, we discuss the prospects and challenges of the fusion of generative adversarial networks and text generation tasks in the future.

We media has gained rapid development on the fast train of digital era, and everyone has a microphone. However, due to the low threshold, large flow and fast realization, we media is also very prone to problems such as infringement, etc. Blockchain, as a kind of distributed ledger, can provide technical support and application in the problem of copyright protection of we media with its characteristics of decentralization, non-tamperability, trustworthiness and privacy protection, etc. This paper discusses and researches the problems existing in the process of we media copyright protection, and provides reasonable ideas and measures for we media copyright protection based on blockchain technology from three perspectives of confirming, using and defending rights, in order to better use technical means to combat infringement in we media field, safeguard the legitimate rights and interests of we media, and build a positive and clear we media communication environment.

In the context of the rapid development of electric vehicles, the uneven space-time distribution of charging station load has caused the loss of efficiency and user experience. Therefore, the space-time prediction of charging station load has become an important research problem. In this paper, based on the St-ResNet model, which has achieved excellent results in space-time flow prediction in the field of traffic flow, we establish a space-time prediction model for a load of electric vehicle charging stations. In the model, we convert the spatial features of multiple charging stations with different geographical locations into 16*16 charging areas. And then, we fuse the three temporal features of the regional spatial distribution of the charging station load, and then use ResPlus to capture the long-distance spatial dependence of the charging load. Finally, we improved 3% to 20% compared with the baseline model.

With the maturity and improvement of blockchain theory and technology, many media have created platforms based on blockchains such as Civil, Decentralized News Network, Steemit and so on. All these successful cases showed us the potential in protecting media industry especially in digital copyright area. In 2021, the Xinhua News Agency’s search engine ChinaSo created Media Integration Blockchain to protect copyright of news media. This paper introduces the design of the Media Integration Blockchain and try to give some advice to upgrade it to increase the range of it. The paper focuses on not only news media but also on We Media. The purpose of this study is to avoid the copyright infringements and contribute to the protection of digital copyright based on the application of blockchains in China’s digital copyright protection.

The global spotlight has been increasingly drawn to the topic of the greenhouse gas effect. This research endeavors to explore the factors that influence Macau's carbon emissions by analyzing yearly figures from 1991 to 2020 and then proffering strategic proposals for the Macau government. The empirical results of the study demonstrate a long-term cointegration relationship, which is based on the natural logarithm of carbon emissions as a dependent variable, and population, real GDP per worker, financial development, industrial structure upgrading, and fixed capital formation as independent variables. The LMDI model shows that, in Macau, the environmental Kuznets curve is present, with energy consumption and carbon emissions being the primary contributors to carbon emissions. Moreover, real GDP per worker, industrial structure improvement, and fixed capital formation have a beneficial effect on carbon emissions, while financial development has a detrimental effect.

Grapheme-to-phoneme (G2P) conversion techniques have been used in many fields, most notably speech synthesis (text-to-speech, TTS). Nowadays, the development of speech synthesis is facilitated with the continuous improvement of G2P conversion techniques. The purpose of the paper is to provide an review of grapheme-to-phoneme conversion methods. First, the grapheme-to-phoneme conversion methods in recent years are sorted out; then the relevant data sets and evaluation metrics are listed, and finally the problems and development trends faced by grapheme-to-phoneme conversion are described.