The paper proposes a new multiple-factor clustering method(NMFCM)with consideration of both box fractal dimension(BFD)and orientation of joints.This method assumes that the BFDs of different clusters were uneven,and c...The paper proposes a new multiple-factor clustering method(NMFCM)with consideration of both box fractal dimension(BFD)and orientation of joints.This method assumes that the BFDs of different clusters were uneven,and clustering was performed by redistributing the joints near the boundaries of clusters on a polar map to maximize an index for estimating the difference of the BFD(DBFD).Three main aspects were studied to develop the NMFCM.First,procedures of the NMFCM and reasonableness of assumptions were illustrated.Second,main factors affecting the NMFCM were investigated by numerical simulations with disk joint models.Finally,two different sections of a rock slope were studied to verify the practicability of the NMFCM.The results demonstrated that:(1)The NMFCM was practical and could effectively alleviate the problem of hard boundary during clustering;(2)The DBFD tended to increase after the improvement of clustering accuracy;(3)The improvement degree of the NMFCM clustering accuracy was mainly influenced by three parameters,namely,the number of clusters,number of redistributed joints,and total number of joints;and(4)The accuracy rate of clustering could be effectively improved by the NMFCM.展开更多
Let Ld=(Zd, Ed) be the d-dimensional lattice, suppose that each edge of Ld be oriented in a random direction, i.e., each edge being independently oriented positive direction along the coordinate axises with probabilit...Let Ld=(Zd, Ed) be the d-dimensional lattice, suppose that each edge of Ld be oriented in a random direction, i.e., each edge being independently oriented positive direction along the coordinate axises with probability p and negative direction otherwise. Let Pp be the percolation measure, η(p) be the probability that there exists an infinite oriented path from the origin. This paper first proves η(p) θ(p) for d 2 and 1/2 p 1, where θ(p) is the percolation probability of bond model; then, as corollaries, the author gets η(1/2) = 0 for d = 2 and dc(1/2) = 2, where dc(1/2) = sup{d: η(1/2) = 0}. Next, based on BK Inequality for arbitrary events in percolation (see[2]), two inequalities are proved, which can be used as FKG Inequality in many cases (note that FKG Inequality is absent for Random-Oriented model). Finally, the author proves the uniqueness of infinite cluster and a theorem on geometry of the infinite cluster (similar to theorem (6.127) in [1] for bond percolation).展开更多
This paper proposed a new approach of sample part classification and design, a so called Or-dered-object-oriented method (O-O-O method). Based on the theory of neural networks, fuzzy clustering algorithm and adaptive ...This paper proposed a new approach of sample part classification and design, a so called Or-dered-object-oriented method (O-O-O method). Based on the theory of neural networks, fuzzy clustering algorithm and adaptive pattern recognition, O-O-O method can be used to classify and design the sample parts automatically. The basic theory, the main step as well as the characteristics of the method are analysed. The construction of the ordered object in application is also presented in this paper.展开更多
本文使用SqueezeNet网络作为基础,构建深度学习点云结构面智能识别模型,案例采用RockBench公开数据库中的数据来验证深度学习模型的识别效果。扫描岩体位于西班牙拜克斯营地地区的TP-7101公路沿线,使用Optech LLRIS 3D激光扫描仪获取点...本文使用SqueezeNet网络作为基础,构建深度学习点云结构面智能识别模型,案例采用RockBench公开数据库中的数据来验证深度学习模型的识别效果。扫描岩体位于西班牙拜克斯营地地区的TP-7101公路沿线,使用Optech LLRIS 3D激光扫描仪获取点云数据。以点云XYZ坐标和法向量作为输入数据,之后挑选550个点构建训练数据集,设置模型的最优参数并以此训练深度学习模型。将训练好的深度学习模型应用到整个点云数据中,对点云数据进行结构面分组。同时,引入Ordering Points to Identify the Clustering Structure(OPTICS)和Density-Based Spatial Clustering of Applications with Noise(DBSCAN)对得到的结构面组进行划分,得到单个结构面。最后,计算出各个结构面的产状。为了验证深度学习模型的识别准确率,采用传统的BP神经网络方法进行验证,计算5组结构面的平均值产状,并与前人计算结果进行对比。经过对比,深度学习模型的计算准确率明显高于浅层BP神经网络模型,其倾向平均误差为9.6456°、倾角平均误差为7.6890°。整体来看,深度学习模型结构更加复杂、提取信息的能力更强,产状计算误差更小。展开更多
Oriented aggregation of nanoparticles has been accomplished by means of solid state reac- tion. Non-crystallized and crystallized ZnO nanoparticles/clusters could be accommodated in the lamellar spacing of inorganic-o...Oriented aggregation of nanoparticles has been accomplished by means of solid state reac- tion. Non-crystallized and crystallized ZnO nanoparticles/clusters could be accommodated in the lamellar spacing of inorganic-organic composite, which were prepared by thermolysis of layered solid zinc-oleate complex at 260 and 300 ℃ in air, respectively. High-resolution transmission electron microscopy and selected area electron diffraction patterns indicate that aggregates are single crystals with various defects. The photoluminescence excitation spectra of both samples show two bands at 272 and 366 nm. The former may originate from electron transfer from valence band to conduction band in ZnO clusters composed of less than 200 ZnO molecules （2R〈2 nm）.展开更多
This paper proposes a clustering technique that minimizes the need for subjective human intervention and is based on elements of rough set theory （RST）. The proposed algorithm is unified in its approach to clusterin...This paper proposes a clustering technique that minimizes the need for subjective human intervention and is based on elements of rough set theory （RST）. The proposed algorithm is unified in its approach to clustering and makes use of both local and global data properties to obtain clustering solutions. It handles single-type and mixed attribute data sets with ease. The results from three data sets of single and mixed attribute types are used to illustrate the technique and establish its efficiency.展开更多
Class cohesion is considered as one of the most important object-oriented software attributes. High cohesion is, in fact, a desirable property of software. Many different metrics have been suggested in the last severa...Class cohesion is considered as one of the most important object-oriented software attributes. High cohesion is, in fact, a desirable property of software. Many different metrics have been suggested in the last several years to measure the cohesion of classes in object-oriented systems. The class of structural object-oriented cohesion metrics is the most in-vestigated category of cohesion metrics. These metrics measure cohesion on structural information extracted from the source code. Empirical studies noted that these metrics fail in many situations to properly reflect cohesion of classes. This paper aims at exploring the use of hierarchical clustering techniques to improve the measurement of cohesion of classes in object-oriented systems. The proposed approach has been evaluated using three particular case studies. We also used in our study three well-known structural cohesion metrics. The achieved results show that the new approach appears to better reflect the cohesion (and structure) of classes than traditional structural cohesion metrics.展开更多
基金funded by the National Natural Science Foundation of China(Grant Nos.41972264 and 52078093)Liaoning Revitalization Talents Program,China(Grant No.XLYC1905015)。
文摘The paper proposes a new multiple-factor clustering method(NMFCM)with consideration of both box fractal dimension(BFD)and orientation of joints.This method assumes that the BFDs of different clusters were uneven,and clustering was performed by redistributing the joints near the boundaries of clusters on a polar map to maximize an index for estimating the difference of the BFD(DBFD).Three main aspects were studied to develop the NMFCM.First,procedures of the NMFCM and reasonableness of assumptions were illustrated.Second,main factors affecting the NMFCM were investigated by numerical simulations with disk joint models.Finally,two different sections of a rock slope were studied to verify the practicability of the NMFCM.The results demonstrated that:(1)The NMFCM was practical and could effectively alleviate the problem of hard boundary during clustering;(2)The DBFD tended to increase after the improvement of clustering accuracy;(3)The improvement degree of the NMFCM clustering accuracy was mainly influenced by three parameters,namely,the number of clusters,number of redistributed joints,and total number of joints;and(4)The accuracy rate of clustering could be effectively improved by the NMFCM.
基金Research supported by the National Natural Science Foundation of China (1977100819571011)Doctoral Programm Fundation of Ins
文摘Let Ld=(Zd, Ed) be the d-dimensional lattice, suppose that each edge of Ld be oriented in a random direction, i.e., each edge being independently oriented positive direction along the coordinate axises with probability p and negative direction otherwise. Let Pp be the percolation measure, η(p) be the probability that there exists an infinite oriented path from the origin. This paper first proves η(p) θ(p) for d 2 and 1/2 p 1, where θ(p) is the percolation probability of bond model; then, as corollaries, the author gets η(1/2) = 0 for d = 2 and dc(1/2) = 2, where dc(1/2) = sup{d: η(1/2) = 0}. Next, based on BK Inequality for arbitrary events in percolation (see[2]), two inequalities are proved, which can be used as FKG Inequality in many cases (note that FKG Inequality is absent for Random-Oriented model). Finally, the author proves the uniqueness of infinite cluster and a theorem on geometry of the infinite cluster (similar to theorem (6.127) in [1] for bond percolation).
文摘This paper proposed a new approach of sample part classification and design, a so called Or-dered-object-oriented method (O-O-O method). Based on the theory of neural networks, fuzzy clustering algorithm and adaptive pattern recognition, O-O-O method can be used to classify and design the sample parts automatically. The basic theory, the main step as well as the characteristics of the method are analysed. The construction of the ordered object in application is also presented in this paper.
文摘本文使用SqueezeNet网络作为基础,构建深度学习点云结构面智能识别模型,案例采用RockBench公开数据库中的数据来验证深度学习模型的识别效果。扫描岩体位于西班牙拜克斯营地地区的TP-7101公路沿线,使用Optech LLRIS 3D激光扫描仪获取点云数据。以点云XYZ坐标和法向量作为输入数据,之后挑选550个点构建训练数据集,设置模型的最优参数并以此训练深度学习模型。将训练好的深度学习模型应用到整个点云数据中,对点云数据进行结构面分组。同时,引入Ordering Points to Identify the Clustering Structure(OPTICS)和Density-Based Spatial Clustering of Applications with Noise(DBSCAN)对得到的结构面组进行划分,得到单个结构面。最后,计算出各个结构面的产状。为了验证深度学习模型的识别准确率,采用传统的BP神经网络方法进行验证,计算5组结构面的平均值产状,并与前人计算结果进行对比。经过对比,深度学习模型的计算准确率明显高于浅层BP神经网络模型,其倾向平均误差为9.6456°、倾角平均误差为7.6890°。整体来看,深度学习模型结构更加复杂、提取信息的能力更强,产状计算误差更小。
文摘Oriented aggregation of nanoparticles has been accomplished by means of solid state reac- tion. Non-crystallized and crystallized ZnO nanoparticles/clusters could be accommodated in the lamellar spacing of inorganic-organic composite, which were prepared by thermolysis of layered solid zinc-oleate complex at 260 and 300 ℃ in air, respectively. High-resolution transmission electron microscopy and selected area electron diffraction patterns indicate that aggregates are single crystals with various defects. The photoluminescence excitation spectra of both samples show two bands at 272 and 366 nm. The former may originate from electron transfer from valence band to conduction band in ZnO clusters composed of less than 200 ZnO molecules （2R〈2 nm）.
文摘This paper proposes a clustering technique that minimizes the need for subjective human intervention and is based on elements of rough set theory （RST）. The proposed algorithm is unified in its approach to clustering and makes use of both local and global data properties to obtain clustering solutions. It handles single-type and mixed attribute data sets with ease. The results from three data sets of single and mixed attribute types are used to illustrate the technique and establish its efficiency.
文摘Class cohesion is considered as one of the most important object-oriented software attributes. High cohesion is, in fact, a desirable property of software. Many different metrics have been suggested in the last several years to measure the cohesion of classes in object-oriented systems. The class of structural object-oriented cohesion metrics is the most in-vestigated category of cohesion metrics. These metrics measure cohesion on structural information extracted from the source code. Empirical studies noted that these metrics fail in many situations to properly reflect cohesion of classes. This paper aims at exploring the use of hierarchical clustering techniques to improve the measurement of cohesion of classes in object-oriented systems. The proposed approach has been evaluated using three particular case studies. We also used in our study three well-known structural cohesion metrics. The achieved results show that the new approach appears to better reflect the cohesion (and structure) of classes than traditional structural cohesion metrics.
