Improvement of Oracle Bone Inscription Recognition Accuracy: A Deep Learning Perspective

Deep learning techniques have been successfully applied in handwriting recognition. Oracle bone inscriptions (OBI) are the earliest hieroglyphs in China and valuable resources for studying the etymology of Chinese characters. OBI are of important historical and cultural value in China; thus, textual research surrounding the characters of OBI is a huge challenge for archaeologists. In this work, we built a dataset named OBI-100, which contains 100 classes of oracle bone inscriptions collected from two OBI dictionaries. The dataset includes more than 128,000 character samples related to the natural environment, humans, animals, plants, etc. In addition, we propose improved models based on three typical deep convolutional network structures to recognize the OBI-100 dataset. By modifying the parameters, adjusting the network structures, and adopting optimization strategies, we demonstrate experimentally that these models perform fairly well in OBI recognition. For the 100-category OBI classification task, the optimal model achieves an accuracy of 99.5%, which shows competitive performance compared with other state-of-the-art approaches. We hope that this work can provide a valuable tool for character recognition of OBI.

Dictionary Apps in Chinese Language Learning

After evaluating available Chinese-as-a-foreign-language (CFL) dictionaries on the market, this chapter identifies CFL learners' difficulties with learning Chinese and their needs for CFL dictionaries through an exploratory mixed-methods study involving interviews with 30 German CFL learners and a survey with 379 participants. This chapter also proposes a dictionary app for learning Chinese for German CFL learners with a focus on microstructural classes, access and network structures, and additional supporting functions.

A Novel Enhanced Russell Graph Efficiency Measure in Network DEA

Hitherto, the presented models for measuring the efficiency score of multi-stage decision-making units (DMUs) either are nonlinear or require to specify the weights for combining their divisional efficiencies. The nonlinearity leads to high computational complexity for these models, especially when used for problems with enormous dimensions, and also assigning various weights to the divisional efficiencies causes to obtain different efficiency scores for the multi-stage network system. To tackle these problems, this study contributes to network DEA by introducing a novel enhanced Russell graph (ERG) efficiency measure for evaluating the general two-stage series network structures. Then, the proposed model is extended into the general multi-stage series network structures. This study also describes the managerial and economic implications of measuring the efficiency score of the multi-stage DMUs and provides two numerical and empirical examples for illustrating the use of our proposed model.

Fast Algorithm for Calculating Transitive Closures of Binary Relations in the Structure of a Network Object

The article proposes a fast algorithm for constructing the transitive closures between all pairs of nodes in the structure of a network object, which can have both directional and non-directional links. The algorithm is based on the disjunctive addition of the elements of certain rows of the adjacency matrix, which models (describe) the structure of the original network object. The article formulates and proves a theorem that using such a procedure, the matrix of transitive closures of a network object can be obtained from the adjacency matrix in two iterations (traversal) on such an array. An estimate of the asymptotic computational complexity of the proposed algorithm is substantiated. The article presents the results of an experimental study of the execution time of such an algorithm on network structures of different dimensions and with different connection densities. For this indicator, the developed algorithm is compared with the well-known approaches of Bellman, Warshall-Floyd, Shimbel, which can also be used to determine the transitive closures of binary relations of network objects. The corresponding graphs of the obtained dependences are given. The proposed algorithm (the logic embedded in it) can become the basis for solving problems of monitoring the connectivity of various subscribers in data transmission networks in real time when managing the load in such networks, where the time spent on routing information flows directly depends on the execution time of control algorithms, as well as when solving other problems on the network structures.

DEVELOPMENT OF STATISTICALLY AVERAGED MODELS OF DEFORMATION OF MATERIALS WITH RANDOM NETWORK STRUCTURE OF DIFFERENTLY ORIENTED FIBERS

The paper describes the developed statistically averaged models of deformation of materials with a random network structure of differently oriented fibers. New methods of stress-strain analysis and micromacromechanical models of material deformation in the volume of bodies made of material with a network structure taking into account structural and physical nonlinearities have been created. These models are based on the micromechanics of network structures at the level of statistical sets of their chains. The novelty of approaches, models, methods and results is the creation of theoretical foundations for the analysis of the deformation of non-traditional network materials. Nonlinear mathematical models of material deformation in the form of a chaotic network structure of one-dimensional fragments are proposed, which are constructed involving fundamentally new approaches to the description of physical and mechanical properties at the micro level of statistical sets of fiber chains and spatial homogenization of their macroproperties. Compared to traditional models, they more adequately model the features of material deformation in the form of spatial chaotic and ordered network structures, as they do not involve a number of additional non-physical hypotheses. This creates fundamentally new opportunities not only for analyzing the properties of such materials, but also when creating new ones with specified properties. Using the created methods, models and research tools, the basis for solving a number of model and applied problems has been created. The nature of deformation of non-traditional materials with a network structure of one-dimensional elements is determined. The macro-properties of these materials are established on the basis of the developed micromechanical models, variational formulations and averaging methods. Keywords: stress-strain state, network structures, contact interaction, finite element method, contact pressure, machine parts, variational formulation

Macroscopic dynamics of gene regulatory networks revealed by individual entropy

Complex systems are usually high-dimensional with intricate interactions among internal components, and may display complicated dynamics under different conditions. While it is difficult to measure detailed dynamics of each component, proper macroscopic description of a complex system is crucial for quantitative studies. In biological systems, each cell is a complex system containing a huge number of molecular components that are interconnected with each other through intricate molecular interaction networks. Here, we consider gene regulatory networks in a cell, and introduce individual entropy as a macroscopic variable to quantify the transcriptional dynamics in response to changes in random perturbations and/or network structures. The proposed individual entropy measures the information entropy of a system at each instant with respect to a basal reference state, and may provide temporal dynamics to characterize switches of system states. Individual entropy provides a method to quantify the stationary macroscopic dynamics of a gene set that is dependent on the gene regulation connections, and can be served as an indicator for the evolution of network structure variation. Moreover, the individual entropy with reference to a preceding state enables us to characterize different dynamic patterns generated from varying network structures. Our results show that the proposed individual entropy can be a valuable macroscopic variable of complex systems in characterizing the transition processes from order to disorder dynamics, and to identify the critical events during the transition process.

Fish provision in a changing environment: The buffering effect of regional trade networks

Local and regional trade networks in small-scale fisheries are important for food security and livelihoods across the world. Such networks consist of both economic flows and social relationships, which connect different production regions to different types of fish demand. The structure of such trade networks, and the actions that take place within them (e.g., people fishing, buying, selling), can influence the capacity of small-scale fisheries to provide sufficient fish in a changing social and ecological context. In this study, we aim to understand the importance of networks between different types of traders that access spatially-distinct fish stocks for the availability and variability of fish provision. We deployed a mixed-methods approach, combining agent-based modelling, network analysis and qualitative data from a small-scale fishery in Baja California Sur, Mexico. The empirical data allowed us to investigate the trade processes that occur within trade networks; and the generation of distinct, empirically-informed network structures. Formalized in an agent-based model, these network structures enable analysis of how different trade networks affect the dynamics of fish provision and the exploitation level of fish stocks. Model results reveal how trade strategies based on social relationships and species diversification can lead to spillover effects between fish species and fishing regions. We found that the proportion of different trader types and their spatial connectivity have the potential to increase fish provision. However, they can also increase overexploitation depending on the specific connectivity patterns and trader types. Moreover, increasing connectivity generally leads to positive outcomes for some individual traders, but this does not necessarily imply better outcomes at the system level. Overall, our model provides an empirically-grounded, stylized representation of a fisheries trading system, and reveals important trade-offs that should be considered when evaluating the potential effect of future changes in regional trade networks.