Sum-Product Networks模型的研究及其在文本分类的应用

图模型在机器学习有着广泛的应用。相比图模型,Sum-Product Networks模型具有更强表达能力和更快的推理速度,所以其在对文本和图像数据建模有着广泛的应用。本文总结Sum-Product Networks这一新的深度概率模型的研究进展,先介绍了固定结构的Sum-Product Networks的参数学习方法,再介绍了根据不同的输入数据而进行的结构和参数学习方法。并且介绍了判别式和生成模型的Sum-Product Networks,最后介绍了Sum-Product Networks在文本分类上的应用。

‘Partly'globalized networks and driving mechanism in resource-based state-owned enterprises:A case study of J Group

In the context of economic globalization,while multinational enterprises from developed countries occupy a high-end position in the global value chain,enterprises from developing countries are often marginalized in the world market.In China,resource-based state-owned enterprises(SOEs)are tasked with the mission of safeguarding resource security,and their internationalization development ideas and strategic deployment are significantly and fundamentally different from those of other non-state-owned enterprises and large multinational corporations.This study provides ideas for the globalization policies of enterprises in developing countries.We consider J Group in western China as a case and discuss its productive investment and global production network development from 2010 to 2019.We found that J Group was‘Partly'globalized,and there are multiple core nodes with the characteristics of centralized and decentralized coexistence in the production network;in addition,the overall layout centre shifted to Southeast Asia and China;however,its global production was restricted by the enterprise's investment security considerations,support and restrictions of the home country,political security risk of the host country,and sanctions from the West.These findings provide insights for future research:under the wave of anti-globalization and'internal circulation as the main body',resource SOEs should consider the potential risk of investment,especially keeping the middle and downstream industrial chain in China as much as possible.

Coal mine safety production forewarning based on improved BP neural network

Firstly, the early warning index system of coal mine safety production was given from four aspects as per- sonnel, environment, equipment and management. Then, improvement measures which are additional momentum method, adaptive learning rate, particle swarm optimization algorithm, variable weight method and asynchronous learning factor, are used to optimize BP neural network models. Further, the models are applied to a comparative study on coal mine safety warning instance. Results show that the identification precision of MPSO-BP network model is higher than GBP and PSO-BP model, and MPSO- BP model can not only effectively reduce the possibility of the network falling into a local minimum point, but also has fast convergence and high precision, which will provide the scientific basis for the forewarnin~ management of coal mine safetv production.

Networked Evolutionary Model of Snow-Drift Game Based on Semi-Tensor Product

This paper investigates the networked evolutionary model based on snow-drift game with the strategy of rewards and penalty. Firstly, by using the semi-tensor product of matrices approach, the mathematical model of the networked evolutionary game is built. Secondly, combined with the matrix expression of logic, the mathematical model is expressed as a dynamic logical system and next converted into its evolutionary dynamic algebraic form. Thirdly, the dynamic evolution process is analyzed and the final level of cooperation is discussed. Finally, the effects of the changes in the rewarding and penalty factors on the level of cooperation in the model are studied separately, and the conclusions are verified by examples.

Decent Work and Industrial Upgrading in Global Production Network: a Case of China Apparel Industry

China apparel industry, which is deeply embedded in the global production network (GPN), faces two urgent issues, social upgrading and economic upgrading. The study of GPN places great emphasis on the two issues. Based on the survey of Ningbo apparel industry, four key components of decent work in China apparel industry are discussed. The role of buyers in promoting decent work in suppliers can't be neglected. There are significant correlations between business type and some indicators of decent work. Though the majority of the apparel firms are engaging in processing, more and more firms are involved in marketing and branding. The upgrading trajectory of China apparel industry leads to the economic and social performances.

Construction of Network Management Information System of Agricultural Products Supply Chain Based on 3PLs

The necessity to construct the network management information system of 3PLs agricultural supply chain is analyzed,showing that 3PLs can improve the overall competitive advantage of agricultural supply chain.3PLs changes the homogeneity management into specialized management of logistics service and achieves the alliance of the subjects at different nodes of agricultural products supply chain.Network management information system structure of agricultural products supply chain based on 3PLs is constructed,including the four layers (the network communication layer,the hardware and software environment layer,the database layer,and the application layer) and 7 function modules (centralized control,transportation process management,material and vehicle scheduling,customer relationship,storage management,customer inquiry,and financial management).Framework for the network management information system of agricultural products supply chain based on 3PLs is put forward.The management of 3PLs mainly includes purchasing management,supplier relationship management,planning management,customer relationship management,storage management and distribution management.Thus,a management system of internal and external integrated agricultural enterprises is obtained.The network management information system of agricultural products supply chain based on 3PLs has realized the effective sharing of enterprise information of agricultural products supply chain at different nodes,establishing a long-term partnership revolving around the 3PLs core enterprise,as well as a supply chain with stable relationship based on the supply chain network system,so as to improve the circulation efficiency of agricultural products,and to explore the sales market for agricultural products.

Territory-based and Ownershipbased Gains from Trade under the Global Production Network——An Empirical Study Based on China

Under global production network,export cannot represent a country's gains from trade,and territory-based gains from trade refer to the remainder of export after deducting the input of intermediate goods and re-export after value-added return.Ownership-based gains from trade refer to the remainder of territory-based gains from trade after further deducting the trade in value added(TVA) realized through the inflow of foreign factors.By creating a multicountry input and output model,this paper calculates the territory-based gains from trade,ownership-based gains from trade,and territory-based gains from trade for foreign countries realized through China's export,as well as valueadded return and territory-based gains from trade for foreign countries realized through China's import.This paper has arrived at the following conclusions:behind China's status as the largest goods exporting country in the world,most of Chinese exports contribute to the gains of foreign countries;value addition for foreign countries realized through China's export and value-added return realized through China's import mostly come from Taiwan region,Japan and South Korea in East Asia;a considerable part of gains from trade for the United States realized through China-US trade is achieved through indirect trade.

The Network Marketing of Fresh Agricultural Products in China

The paper introduced the connotation of network marketing of fresh agricultural products and analyzed the fundaments of launching the C2C marketing of fresh agricultural products in China.The fundaments mainly cover the development of e-commerce,the transformation of consumption concept,the change of life style and the features of Chinese agricultural production.The developmental status of the C2C marketing of Chinese fresh vegetables and fresh fruits is introduced and the plights,including imperfect rural network infrastructure,inadequate talents specialized in network marketing of fresh agricultural products,uneven quality of agricultural products,immature logistics system,uncertainty existed in faced by the C2C marketing of fresh agricultural products and imperfect governmental protection system,are pointed out.In the end,the developmental trend of Chinese network marketing of fresh agricultural products on the basis of C2C mode is described.

China reshapes the East Asian production network

From the perspective of intra-product specialization and with in-depth analysis of trade statistics,this paper investigates the influence of China's rise on the East Asian production network.Our conclusions suggest that in integrating into the East Asian production network,China has gradually emerged as the manufacturing center of East Asia,weakening the regional influence of the Four Asian Tigers.Meanwhile,the competitive effect of China's rise has helped promote the specialization levels of the network's members and even the network as a whole.With cooperation in various processes of intra-product specialization,internal connections of the East Asian production network were further strengthened.In addition,China became an export platform of East Asia,transforming the export pattern of the East Asian production network to world markets from "bilateral trade" into "triangular trade," trade via China.

Intermodal Freight Network Design for Transport of Perishable Products

The amount of perishable products transported via the existing intermodal freight networks has significantly increased over the last years. Perishable products tend to decay due to a wide range of external factors. Supply chain operations mismanagement causes waste of substantial volumes of perishable products every year. The heretofore proposed mathematical models optimize certain supply chain processes and reduce decay of perishable products, but primarily deal with local production, inventory, distribution, and retailing of perishable products. However, significant quantities of perishable products are delivered from different continents, which shall increase the total transportation time and decay potential of perishable products as compared to local deliveries. This paper proposes a novel optimization model to design the intermodal freight network for both local and long-haul deliveries of perishable products. The objective of the model aims to minimize the total cost associated with transportation and decay of perishable products. A set of piecewise approximations are applied to linearize the non-linear decay function for each perishable product type. CPLEX is used to solve the problem. Comprehensive numerical experiments are conducted using the intermodal freight network for import of the seafood perishable products to the United States to draw important managerial insights. Results demonstrate that increasing product decay cost may significantly change the design of intermodal freight network for transport of perishable products, cause modal shifts and affect the total transportation time and associated costs.

ZTE Achieves Strong Growth of Bearer Network Products

ZTE Corporation continues to register remarkable growthrate for its bearer network products according to theannouncement the

AQSIQ Popularizing Nation-wide Launching of Electronic Network for Product Quality Supervision

According to AQSIQ, it has made an effort to popularize the launching ofelectronic ''network for product quality supervision in the whole nation. China Product QualitySupervision Electronic Network is a nation-wide network, which employs modern informationtechnology, network technology and encoding technology to supervise enterprises'' products in anelectronic way. The new system labels digital information on all items of products that are includedin the network.

ZTE's Optical Transmission Products Used in National Backbone Network of Indonesia

1 Technical Features of ZTE’sDWDM SystemThe toll backbone wave division equipmentof ZXWM M900 backbone transmissionplatform is well designed forlarge-capacity optical transmission. It canfully satisfy the networking and managementrequirements of diversified users and be fit forvarious toll backbone networks. The systemsupports up to 40 working wavelengths and up to400 Gb/s transmission capacity, and thewavelength selection and interval are in strictcompliance with ITU-T Recommendations. It

Modelling the impact of climate change on rangeland forage production using a generalized regression neural network:a case study in Isfahan Province,Central Iran

Monitoring of rangeland forage production at specified spatial and temporal scales is necessary for grazing management and also for implementation of rehabilitation projects in rangelands. This study focused on the capability of a generalized regression neural network（GRNN） model combined with GIS techniques to explore the impact of climate change on rangeland forage production. Specifically, a dataset of 115 monitored records of forage production were collected from 16 rangeland sites during the period 1998–2007 in Isfahan Province, Central Iran. Neural network models were designed using the monitored forage production values and available environmental data（including climate and topography data）, and the performance of each network model was assessed using the mean estimation error（MEE）, model efficiency factor（MEF）, and correlation coefficient（r）. The best neural network model was then selected and further applied to predict the forage production of rangelands in the future（in 2030 and 2080） under A1 B climate change scenario using Hadley Centre coupled model. The present and future forage production maps were also produced. Rangeland forage production exhibited strong correlations with environmental factors, such as slope, elevation, aspect and annual temperature. The present forage production in the study area varied from 25.6 to 574.1 kg/hm^2. Under climate change scenario, the annual temperature was predicted to increase and the annual precipitation was predicted to decrease. The prediction maps of forage production in the future indicated that the area with low level of forage production（0–100 kg/hm^2） will increase while the areas with moderate, moderately high and high levels of forage production（≥100 kg/hm^2） will decrease both in 2030 and in 2080, which may be attributable to the increasing annual temperature and decreasing annual precipitation. It was predicted that forage production of rangelands will decrease in the next couple of decades, especially in the western and southern parts of Isfahan Province. These changes are more pronounced in elevations between 2200 and 2900 m. Therefore, rangeland managers have to cope with these changes by holistic management approaches through mitigation and human adaptations.

Study on the Product Description Method for Networked Customization

Networked customization had gradually become one of t he import manufacturing and marketing patterns in the information age. Product d escription had an important effect on realizing networked customization. From th is point of view, this paper firstly analyzed four requirements of adaptability, stability, conciseness and definitude that product description methods for netw orked customization must satisfy. According to these requirements, the paper com mented on current methods of product description, pointing out their advantages and disadvantages. As none of current methods could satisfy the four requirement s simultaneously, they were not fully applicable to networked customization. The n a plug-in based product description method was brought forward and discussed in detail. In this method, product description for networked customization w as abstracted as three component parts of basic description, basic feature and c ustomization feature. These parts had different characteristics. Basic descripti on was applicable to all products types. Basic feature was stable as concerning specific product type. Variants of certain product type all possessed identi cal basic features. Customization features were variable. Different combinations of customization features would give birth to different product variants. Apply ing object-oriented method, three standard product description plug-ins could be got by encapsulating these component parts. As "basic feature" and "customiza tion feature" plug-in had similar structures, they could be further abstracted as "feature" plug-in. When describing a type of product, a "basic description" plug-in would be instantiated firstly. Then according to product’s specific ch aracteristics, necessary number of "basic feature" and "customization feature" p lug-ins would be instantiated and appended, such as "color", "material", etc. E ach "basic feature" plug-in related to an instantiated "feature value" plug-in , while each "customization feature" plug-in related to one or multiple instant iated "feature value" plug-ins. Finally, "associate" and "exclude" relation ship should be specified among associated "customization feature" plug-ins. Thr ough appending and combining various plug-ins, this method was capable of descr ibing any type of product while meeting the four requirements simultaneously. Th is paper also gave an example to clarify the application of plug-in based produ ct description method. This example firstly described a kind of customizable TV provided by certain company. Then it showed how to describe an individual co nfiguration of this customizable TV in XML. Furthermore, suppose a consumer want ed to search for relevant information of a TV with game function, this example explained how to utilize XML query language to benefit consumer’s searching for product information. From analysis, we could draw the conclusion that plug -in based product description method was superior to current description method s and fully adapted to networked customization environment.

A Study of Optimum Switching Problem for Production Systems Considering Efficiency, Delivery Time and Green Evaluation

This paper aims to derive the optimal switching strategy for production system considering efficiency, delivery time and green evaluation. Nowadays more and more manufacturing and logistics systems not only pursue better work efficiency, but also focus on green energy evaluation issues. Cost reduction and shortening of delivery time are always important management issues in pursuit of efficiency and optimization of the entire production system because of global production competition. In a market situation where customer needs change in various ways, in particular, due to inadequate quality, changes in the local environment, natural disasters and so on. Therefore, prompt planning of management measures such as switching work processes and changing production methods has become an important issue. On the other hand, since the Paris Agreement came into effect, the construction of an environment-friendly production system has been required as an approach to environmental problems such as global warming. In this paper, we propose an optimum switching model of production systems considering efficiency, delivery time and green evaluation using a green evaluation index (GEC: Green Energy Coefficient). We also discuss the optimal switching strategy by numerical observation.

Research on Networked Rapid Product Development Process

Today the cycle time of the product develop is requ ir ed to be shortened. At the same time the requirement of the customers becomes mo re and more diverse and complex. The capability of the develop unit is limited b ecause of the existence of heterogeneous systems and distributed environments. I n this paper, we bring forward a new approach to solve the problem in product de velopment process. We also settle part key technologies in it. A great deal of information from all kinds of sources in the distributed develop ment process is interweaved. The solution to organize the workflow and manage th e information in the process is called for anxiously. We use a new approach that is asynchronous and synchronous coupling product development approach based on the network. The approach extends the develop process from the time axis. Then t he activities in the process are organized from the asynchronous and synchronous aspects. The state of every activity projects at the ASN (active semantic netwo rk). The ASN includes decision system, intelligent agent, user interface and net work. The ASN decides the types and states of the activities and deals with the couple relationship among them. The knowledge stored in ASN is open to all users through the relative interfaces. Every specialist keeps contact with their user s relying on collaborative platform implements CSCW (computer support collaborat ive work) that integrated product/process design and development. The lack of gl obal communication in product development process can be prevented in the most d egree. The key technologies that exist in the asynchronous and synchronous coupling pro duct develop approach include: integrated development structure, orderly organiz ation of information, transparent management of process, agile transfer of infor mation and rapid prototype. The development process can be completed quickly by these technologies. The technologies involve wide content. In this paper, we dis cuss some key technologies. We validate the approach by the projectrapid response manufacturing a pplication in the distributed environment. The expensive device, high technology and low using lead to RE (Rapid engineering) and RP (Rapid prototype) service a pplication by the network. RE and RP develop rapidly due to the accelerated prod uct development process. RE and RP application service platform is built in the project.

Fine-Grained Classification of Product Images Based on Convolutional Neural Networks

With the rapid development of the Internet of things and e-commerce, feature-based image retrieval and classification have become a serious challenge for shoppers searching websites for relevant product information. The last decade has witnessed great interest in research on content-based feature extraction techniques. Moreover, semantic attributes cannot fully express the rich image information. This paper designs and trains a deep convolutional neural network that the convolution kernel size and the order of network connection are based on the high efficiency of the filter capacity and coverage. To solve the problem of long training time and high resource share of deep convolutional neural network, this paper designed a shallow convolutional neural network to achieve the similar classification accuracy. The deep and shallow convolutional neural networks have data pre-processing, feature extraction and softmax classification. To evaluate the classification performance of the network, experiments were conducted using a public database Caltech256 and a homemade product image database containing 15 species of garment and 5 species of shoes on a total of 20,000 color images from shopping websites. Compared with the classification accuracy of combining content-based feature extraction techniques with traditional support vector machine techniques from 76.3% to 86.2%, the deep convolutional neural network obtains an impressive state-of-the-art classification accuracy of 92.1%, and the shallow convolutional neural network reached a classification accuracy of 90.6%. Moreover, the proposed convolutional neural networks can be integrated and implemented in other colour image database.

交通改善、产业链贸易与区域生产网络

理论上,交通改善能够促进企业间贸易与产业间的分工合作,进而形成区域乃至全国性生产网络。本文利用福建省企业增值税发票数据构建了产业链关联变量,考察了高速公路对省内企业间产业分工与产业链形成的影响,有以下发现:首先,高速公路促进了企业的产业间销售,加深了与产业链企业的贸易联系,并提升了销售对象的产业多样性;其次,上游企业贸易对下游企业贸易有显著推动作用,而且上述结果在考虑内生性以及可能产生估计偏误的情形下保持稳健;最后,对不同产业以及生产网络中心度的异质性分析发现,交通改善对制造业下游的产业影响强于服务业,而且高速公路能够带动生产网络中的弱势产业,从而降低网络的结构失衡。