Search

Ports are essential and strategic nodes of international trade and economic activities. The integration of port resources helps enhance the optimal division of port functions to further promote economic development. This paper explores the evolutionary trends of port integration policy from 2011 to 2021 in China through a text-mining approach. We first used the Latent Dirichlet Allocation (LDA) topic analysis method to analyze the port integration policies and summarize the evolutionary trends of port policies. In addition, an association analysis was conducted to explore port integration policies' impacts by examining their relationships among the environmental protection, digitalization level, and port development scale. Our findings suggested that China's port integration policy has evolved from specific and simple to abstract and complicated at the managerial level. Port development has enriched from infrastructure construction to the ideology of economic development. Our findings show that China's port integration is gradually improving the port operation. The findings of the study contribute to the extant literature by analyzing the port integration policy evolution in China and can be referred to by other countries. • Text mining method was used to explore the port integration policy evolution. • Port integration policy evolved from specific and simple to abstract and complicated. • Port development enriched from basic infrastructure to the governance model. • The evolution of port integration policy was related to reducing carbon emissions in the port sector. • The evolution of port integration policy made the ports intelligent in association with digital technology such as big data and AI. [ABSTRACT FROM AUTHOR]

Colour is an important factor in the expression of recognizability and cultural identity of regional cultural and creative design. At present, the colour recognition of regional characteristic and the colour association of regional culture mainly rely on the designer's subjective perception. To obtain the target colour resources with reference for regional cultural and design need, this study proposes a scientific method of colour extraction and strong colour association matching of Shanghai style Chinese paintings by machine learning, and applies the related results to the colour design of cultural and creative products. Firstly, using the SLIC superpixel algorithm and Mean shift algorithm to realize the overall dimensionality reduction of the image features and the colour aggregation gradually, so as to extract the characteristic colours of Shanghai style Chinese paintings; Secondly, we introduce a data mining algorithm (Apriori) to mine out the association rules from multiple characteristics colours and filter out strongly associated colour combinations; Finally, we apply the colour combinations and colour tones to the colour design of the creative products. In order to verify the scientificity of the colour extraction and colour matching method proposed in this paper, we selected another painter's paintings in the same school as the algorithm experimental validation sample, similar results were obtained. In addition, we measured user satisfaction using the degree of awakening to Shanghai style culture and the propensity to make consumer decisions as the evaluation dimensions, which proves that the method of this study is effective. [ABSTRACT FROM AUTHOR]

The Cyber-Physical System covers a wide range of applications, many of which are involved in critical infrastructure, and the cybersecurity attacks on them become more and more threatening. Currently, most of the comprehensive analysis of compound attacks depend on the experience of security analysts. To improve the efficiency and accuracy of compound attack research, this paper introduces a knowledge graph into compound attack detection and constructs a cybersecurity knowledge graph based on the knowledge of known attacks. The cybersecurity knowledge graph can carry out correlation analysis on real-time data to restore the attack process. The main work of this paper is to construct the cybersecurity knowledge graph and to apply mining found compound attacks automatically. Besides, a multi-dimensional data association analysis algorithm based on dynamic clustering mechanism, and an attack chain complementation-pruning method based on optimal reaching path queries are proposed to solve the problem of low efficiency in correlation analysis caused by redundant data and the problem of missing and misunderstandings in the collection data. Experiments show that the cyber security knowledge graph construction method and attack chain optimization-pruning method proposed in this paper improve the accuracy and efficiency of attack chain mining. [ABSTRACT FROM AUTHOR]

With the development of artificial intelligence, the most representative deep learning has been applied to various fields, which is greatly influencing human society. However, deep neural networks (DNNs) are still a black-box model, and the process how they make decisions internally is still difficult to understand and control. At the same time, DNNs take up more hardware resources, resulting in high energy consumption. Therefore, it is significant to study the characteristics of deep AI models and deeply understand the interactions between parameters within AI models so as to improve the interpretability of DNNs, optimize their structure and increase their computational efficiency. In this paper, we propose a graph-based interpretability method for deep neural networks (GIMDNN). The running parameters of DNNs are modeled as a graph by using a kernel function or the Graph Transformer Networks (GTN), where the nodes of the graph are obtained by dimensional mapping of the parameters of the DNNs, and the edges are calculated by the Gaussian kernel function. The generated graphs are classified by a graph convolutional network (GCN). The association relationship between the adjacent layers and the running mechanism of DNNs are analyzed, and the importance of the parameters of each layer in the DNNs for the final classification result can be obtained. Convolutional neural networks (CNNs) are one of the most representative models in DNNs. The proposed method is experimentally evaluated on the CNNs. The experimental results show that the proposed method can interpret the associations among the weight parameters as well as the correlation between two adjacent layers. Therefore, the DNNs for special tasks, such as portable applications, edge computing, and so on, can be customized, the number of parameters can be reduced. It is valuable to interpret the operation and principle of CNNs. [ABSTRACT FROM AUTHOR]

Highlights • A data mining method for topology optimization is proposed. • The method sequentially applies clustering and association rule analysis. • A distance measure for pairs of configurations is defined. • Confidence and lift indexes in multiobjective topology optimizations are defined. • The method is applied in 3- and 4-objective design problems in 2 and 3 dimensions. Abstract Optimum design problems, including structural optimization problems, often include multiple objectives. A multiobjective optimization problem usually provides a number of optimal solutions, called non-dominated solutions or Pareto-optimal solutions. In multiobjective topology optimization scenarios, decision makers face the challenging task of choosing the most effective solution that meets their needs; serial comparisons among a set of Pareto-optimal solution are cumbersome, as are trial-and-error attempts to find an appropriate solution among a host of alternatives. On the other hand, the recent integration of data mining techniques in multiobjective optimization methods can provide decision makers with important, highly pertinent, and useful knowledge. In this paper, we propose a data mining technique for knowledge discovery in multiobjective topology optimization. The proposed method sequentially applies clustering and association rule analysis to a Pareto-optimal solution set. First, clustering is applied in the design space and the result is then visualized in the objective space. After clustering, detailed features in each cluster are analyzed based on the concept of association rule analysis, so that characteristic substructures can be extracted from each cluster of solutions. In four numerical examples, we demonstrate that the proposed method provides pertinent knowledge that aids comprehension of the key substructures responsible for one or more desired performances, thereby giving decision makers a useful tool for discovery of particularly effective design solutions. [ABSTRACT FROM AUTHOR]

Abstract Data mining is one of the most prominent methodologies demonstrating efficiency in various tasks related to pattern detection. Applications of data mining for fraud detection are numerous, for the detection of fraud, ranging from credit card fraud to taxation fraud, are numerous. However, the applications of data mining for fraud related to internal control are scarce. The goal of this paper is to present the application of data mining techniques (the CHAID decision tree) for discovering patterns in internal control-related fraud in one project-based organization. [ABSTRACT FROM AUTHOR]

The on-demand parameters of heat source are the precondition for ensuring the safe, stable and energy-saving operation of heating system. For large-scale heating system, the existing predictive methods are not applicable due to the complexity of modeling, while the design heating load index method has many influencing factors, resulting in a low accuracy of obtaining accurate values. In this paper, firstly, the simplified mathematical model of the heating substation is built, and the calculation methods of under-demand rate (η) and energy-saving rate (ς) are proposed for diagnosis thermal balance and evaluation energy-saving potential. Secondly, the variation relationship among heat source parameters is analyzed and the input parameters of the analysis process are determined, then cluster analysis is adopted to identify the operation strategy, and non-on-demand clusters are eliminated from the perspective of professional knowledge. Thirdly, the data in the remaining clusters are discretized, association analysis is used to obtain the frequent item-sets of each cluster, and on-demand heating parameters of each cluster are obtained. Finally, η and ς are used to evaluate the object heating system. This application reveals that the descriptive data mining techniques combined with professional knowledge can successfully identify the on-demand parameters from the historical data of heat source. • Operation model of the heating substation is built. • Evaluation indexes are proposed for diagnosis thermal balance and evaluation energy-saving potential. • Cluster analysis is used to identify operation strategy. • Association analysis is used to obtain on-demand parameters. [ABSTRACT FROM AUTHOR]

Hepatocyte nuclear factor-4α ( HNF-4α ), a member of the hepatocyte nuclear factor family, plays an important role in regulating the expression of genes involved in the development, differentiation and normal function of liver and pancreatic β cells, as well as the maintenance of glucose homeostasis. Tetra-primer amplification refractory mutation system PCR (T-ARMS-PCR) is a new method offering fast detection and extreme simplicity at a negligible cost for SNP genotyping. In this paper, we characterize the polymorphisms of the bovine HNF-4α gene in three Chinese indigenous cattle breeds (n = 660). Six novel SNPs were identified including 1 mutation in the coding region and others in introns. The statistical analyses indicated that 4 SNPs (g.T53729C, g.A53861G, g.A65188C and g.T65444C) affected growth traits markedly ( P < 0.05) in Qinchuan cattle (2 years after birth). Besides, haplotypes involving these 4 SNP sites in the bovine HNF-4α gene were identified and their effects on growth traits were also analyzed. The results showed that haplotypes 2, 7, 9 and 11 were predominant and accounted for 73.2%, 59.6%, and 67.1% in Qinchuan, Nanyang and Jiaxian cattle breeds, respectively. Hap9 (TAAT) was extremely predominant in all test populations, which suggested that individuals with Hap9 were more adapted to the environment. Furthermore, 4 combined haplotypes were constructed to guarantee the reliability of analysis results in Qinchuan cattle. There were also significant differences in body length ( P < 0.05). These findings will benefit for the application of DNA marker related to the growth traits on marker-assisted selection (MAS), and improve the performance of beef cattle. [ABSTRACT FROM AUTHOR]

Abstract: Since 1978, China''s economy and science & technology boom and it has shown great power and immense potential to the world. This prosperity is inseparable from China''s international collaboration. This paper is focused on China''s international technology collaboration activities based on the USPTO patent data from 1976 to 2010. By using of patent statistics analysis and association analysis method, we analyze the US patent data on co-inventor level. The analysis showed that the number of joint patent got a leap and increasing at a high speed at the beginning of 1997. The cooperation scope is gradually expanded and the innovation capacity improved significantly. Most of the co-inventors are from the United States, Taiwan province, Hong Kong and Japan. In the areas of cooperation research field, the joint patents mainly distribute in G06, H01 and H05. [Copyright &y& Elsevier]

Abstract: The problem of protein secondary structure prediction is one of the most important problems in Bioinformatics. After the study of this problem for 30 years and more, there have been some breakthroughs. Especially, the introduction of ensemble prediction model and hybrid prediction model makes the accuracy of prediction better, but there is a long distance to induce the tertiary structures from the secondary ones. As one of the extension researches of KDTICM [Bingru, Yang (2004). Knowledge discovery based on theory of inner cognition mechanism and application. Beijing: Electronic Industry Press] theory, this paper proposed a method KAAPRO, which is based on Maradbcm algorithm which is induced by KDD∗ model and combined with CBA, for protein secondary structure prediction. And a gradually enhanced, multi-layer systematic prediction model, compound pyramid model, is proposed. The kernel of this model is KAAPRO. Domain knowledge is used through the whole model, and the physical–chemical attributes are chosen by causal cellular automata. In the experiment, the test proteins used in reference Muggleton et al. (Muggleton, S. H., King, R., Sternberg, M. (1992). Protein secondary structure prediction using logic-based machine learning. Protein Engineering, 5(7), 647–657) are predicted. The structures of amino acids, whose structural traits are obscure, are predicted well by KAAPRO. Hence, the result of this model is satisfying too. [Copyright &y& Elsevier]

Abstract: Though frequent itemsets and association rules express interesting association among items of frequently occurring itemsets in a database, there may exist other types of interesting associations among the items. A critical analysis of frequent itemsets would provide more insight about a database. In this paper, we introduce the notion of conditional pattern in a database. Conditional patterns are interesting and useful for solving many problems. We propose an algorithm for mining conditional patterns in a database. Experiments are conducted on three real datasets. The results of the experiments show that conditional patterns store significant nuggets of knowledge about a database. [Copyright &y& Elsevier]

Abstract: The extent to which elephants hold behavioural traits in common with human beings is relevant to the ethics of how we treat them. Observations show that elephants, like humans, are concerned with distressed or deceased individuals, and render assistance to the ailing and show a special interest in dead bodies of their own kind. This paper reports helping and investigative behaviour of different elephants and their families towards a dying and deceased matriarch. We make use of long-term association records, GPS tracking data and direct observations. Records made around the time of death, shows that the helping behaviour and special interest exhibited was not restricted to closely related kin. The case is made that elephants, like human beings, can show compassionate behaviour to others in distress. They have a general awareness and curiosity about death, as these behaviours are directed both towards kin and non-related individuals. [Copyright &y& Elsevier]

The aim of this study was to investigate 96 single-nucleotide polymorphisms (SNPs) from 54 candidate genes, and test the associations of the polymorphic SNPs with milk yield, composition, milk urea nitrogen (MUN) content and somatic cell score (SCS) in individual milk samples from Italian Brown Swiss cows. Milk and blood samples were collected from 1271 cows sampled once from 85 herds. Milk production, quality traits (i.e. protein, casein, fat and lactose percentages), MUN and SCS were measured for each milk sample. Genotyping was performed using a custom Illumina VeraCode GoldenGate approach. A Bayesian linear animal model that considered the effects of herd, days in milk, parity, SNP genotype and additive polygenic effect was used for the association analysis. Our results showed that 14 of the 51 polymorphic SNPs had relevant additive effects on at least one of the aforementioned traits. Polymorphisms in the glucocorticoid receptor DNA-binding factor 1 (GRLF1), prolactin receptor (PRLR) and chemokine ligand 2 (CCL2) were associated with milk yield; an SNP in the stearoyl-CoA desaturase (SCD-1) was related to fat content; SNPs in the caspase recruitment domain 15 protein (CARD15) and lipin 1 (LPIN1) affected the protein and casein contents; SNPs in growth hormone 1 (GH1), lactotransferrin (LTF) and SCD-1 were relevant for casein number; variants in beta casein (CSN2), GH1, GRLF1 and LTF affected lactose content; SNPs in beta-2 adrenergic receptor (ADRB2), serpin peptidase inhibitor (PI) and SCD-1 were associated with MUN; and SNPs in acetyl-CoA carboxylase alpha (ACACA) and signal transducer and activator of transcription 5A (STAT5A) were relevant in explaining the variation of SCS. Although further research is needed to validate these SNPs in other populations and breeds, the association between these markers and milk yield, composition, MUN and SCS could be exploited in gene-assisted selection programs for genetic improvement purposes. [ABSTRACT FROM PUBLISHER]