Narrow Results

Link prediction aims to identify missing links within static networks or estimate the probability of emerging links in dynamic networks, representing a critical and challenging research direction in complex network. Many similarity-based methods have been established from various viewpoints, however, there are relatively few methods that take into account both the path information between node pairs and the nodes along the path. To fill this gap, we propose two novel link prediction algorithms, namely LPRA and LPH. The core idea is that the similarity between node pairs is closely related to the local paths connecting the two nodes and the resource transition capability of the nodes along those paths. The algorithms utilize the local paths with adjustable lengths between node pairs and the topological information of the nodes on the path to calculate the similarity index, which incorporate the resource transition capabilities of all the nodes on the possible paths between node pairs. We conducted multiple groups of comparative experiments on 10 real-world networks to validate the effectiveness of the proposed algorithm. Experimental results demonstrate that LPRA and LPH outperform nine classical methods and five recently popular methods. Moreover, LPRA demonstrates a significant difference level (P-value $<0.05$) compared to most of the methods in the ANOVA of AUC accuracy, indicating a statistically significant performance advantage of LPRA.

Two main difficulties in the problem of classification in partially labeled networks are the sparsity of the known labeled nodes and inconsistency of label information. To address these two difficulties, we propose a similarity-based method, where the basic assumption is that two nodes are more likely to be categorized into the same class if they are more similar. In this paper, we introduce ten similarity indices defined based on the network structure. Empirical results on the co-purchase network of political books show that the similarity-based method can, to some extent, overcome these two difficulties and give higher accurate classification than the relational neighbors method, especially when the labeled nodes are sparse. Furthermore, we find that when the information of known labeled nodes is sufficient, the indices considering only local information can perform as good as those global indices while having much lower computational complexity.

Recent years have witnessed the increasing of available network data; however, much of those data is incomplete. Link prediction, which can find the missing links of a network, plays an important role in the research and analysis of complex networks. Based on the assumption that two unconnected nodes which are highly similar are very likely to have an interaction, most of the existing algorithms solve the link prediction problem by computing nodes' similarities. The fundamental requirement of those algorithms is accurate and effective similarity indices. In this paper, we propose a new similarity index, namely similarity based on activity and connectivity (SAC), which performs link prediction more accurately. To compute the similarity between two nodes, this index employs the average activity of these two nodes in their common neighborhood and the connectivities between them and their common neighbors. The higher the average activity is and the stronger the connectivities are, the more similar the two nodes are. The proposed index not only commendably distinguishes the contributions of paths but also incorporates the influence of endpoints. Therefore, it can achieve a better predicting result. To verify the performance of SAC, we conduct experiments on 10 real-world networks. Experimental results demonstrate that SAC outperforms the compared baselines.

The link prediction algorithm is one of the key technologies to reveal the inherent rule of network evolution. This paper proposes a novel link prediction algorithm based on the properties of the local community, which is composed of the common neighbor nodes of any two nodes in the network and the links between these nodes. By referring to the node degree and the condition of assortativity or disassortativity in a network, we comprehensively consider the effect of the shortest path and edge clustering coefficient within the local community on node similarity. We numerically show the proposed method provide good link prediction results.

With the proliferation of available network data, link prediction has become increasingly important and captured growing attention from various disciplines. To enhance the prediction accuracy by making full use of community structure information, this paper proposes a new link prediction model, namely CMS, in which different community memberships of nodes are investigated. In the opinion of CMS, different memberships can have different influence to link’s formation. To estimate the connection likelihood between two nodes, the CMS model weights the contribution of each shared neighbor according to the corresponding community membership. Three CMS-based methods are derived by introducing three forms of contribution that neighbors make. Extensive experiments on 12 networks are conducted to evaluate the performance of CMS-based methods. The results manifest that CMS-based methods are more effective and robust than baselines.

Link prediction in complex networks has attracted much attention recently. Many local similarity measures based on the measurements of node similarity have been proposed. Among these local similarity indices, the neighborhood-based indices Common Neighbors (CN), Adamic-Adar (AA) and Resource Allocation (RA) index perform best. It is found that the node similarity indices required only information on the nearest neighbors are assigned high scores and have very low computational complexity. In this paper, a new index based on the contribution of common neighbor nodes to edges is proposed and shown to have competitively good or even better prediction than other neighborhood-based indices especially for the network with low clustering coefficient with its high efficiency and simplicity.

Link prediction can estimate the probablity of the existence of an unknown or future edges between two arbitrary disconnected nodes (two seed nodes) in complex networks on the basis of information regarding network nodes, edges and topology. With the important practical value in many fields such as social networks, electronic commerce, data mining and biological networks, link prediction is attracting considerable attention from scientists in various fields. In this paper, we find that degree distribution and strength of two- and three-step local paths between two seed nodes can reveal effective similarity information between the two nodes. An index called local major path degree (LMPD) is proposed to estimate the probability of generating a link between two seed nodes. To indicate the efficiency of this algorithm, we compare it with nine well-known similarity indices based on local information in 12 real networks. Results show that the LMPD algorithm can achieve high prediction performance.

Epileptic seizures are generated by abnormal activity of neurons. The prediction of epileptic seizures is an important issue in the field of neurology, since it may improve the quality of life of patients suffering from drug resistant epilepsy. In this study a new similarity index based on symbolic dynamic techniques which can be used for extracting behavior of chaotic time series is presented. Using Freiburg EEG dataset, it is found that the method is able to detect the behavioral changes of the neural activity prior to epileptic seizures, so it can be used for prediction of epileptic seizure. A sensitivity of 63.75% with 0.33 false positive rate (FPR) in all 21 patients and sensitivity of 96.66% with 0.33 FPR in eight patients were achieved using the proposed method. Moreover, the method was evaluated by applying on Logistic and Tent map with different parameters to demonstrate its robustness and ability in determining similarity between two time series with the same chaotic characterization.

Y. Miyazawa introduced a two-variable polynomial invariant of virtual knots in 2006 [Magnetic graphs and an invariant for virtual links, J. Knot Theory Ramifications 15 (2006) 1319–1334] and then generalized it to give a multi-variable one via decorated virtual magnetic graph diagrams in 2008. A. Ishii gave a simple state model for the two-variable Miyazawa polynomial by using pole diagrams in 2008 [A multi-variable polynomial invariant for virtual knots and links, J. Knot Theory Ramifications 17 (2008) 1311–1326]. H. A. Dye and L. H. Kauffman constructed an arrow polynomial of a virtual link in 2009 which is equivalent to the multi-variable Miyazawa polynomial [Virtual crossing number and the arrow polynomial, preprint (2008), arXiv:0810.3858v3, http://front.math.ucdavis.edu.]. We give a bracket model for the multi-variable Miyazawa polynomial via pole diagrams and polar tangles similarly to the Ishii's state model for the two-variable polynomial. By normalizing the bracket polynomial we get the multi-variable Miyazawa polynomial f_K ∈ ℤ[A, A^-1, K₁, K₂, …] of a virtual link K. n-similar knots take the same value for any Vassiliev invariant of degree < n. We show that f_K1 ≡ f_K2 mod (A⁴ - 1)ⁿ if two virtual links K₁ and K₂ are n-similar. Also we give a necessary condition for a virtual link to be periodic by using n-similarity of virtual tangles and the Miyazawa polynomial.

We re-take the possibilistic (as opposed to probabilistic) approach to information coding put forward in ^1,2. To enhance the possibilistic approach also outside the realm of "subjective" uncertainties, in this paper we adopt an "objective" interpretation of possibilistic source coding based on utility functions and an "objective" interpretation of possibilistic channel coding based on distortion measures and similarity indices. We stress the relationship between possibilistic coding as based on distortions between sequences and algebraic coding as based on minimum distances between codewords. We compute the operational (coding-theoretic) entropy for a new class of possibilistic sources.

This paper deals with the problem of multicriteria clusters construction. The aim is to propose a multicriteria clustering procedure aiming at discovering data structures from a multicriteria perspective by defining a dissimilarity measure which takes into account the multicriteria nature of the problem. Comparing two objects in the multicriteria context is based on the preference information that expresses whether these objects are indifferent, incomparable or one is preferred to the other. The proposed approach uses this preference information with an agreement–disagreement similarity index to compute a dissimilarity measure. The approach generates, according to the preference relations, a set of clusterings. Each clustering expresses a way of grouping objects according to the preference relation used. A good quality final clustering is obtained by combining the clusterings generated previously using a clustering ensemble technique.

In this paper some of the characteristic of Hückel method, were exploited in order to obtain some important results, through a new technique with which it is possible to obtain non-degenerate characteristic values as in the case of pyrrole and allowing the expression of conjugated ring systems (Pentalene) as function of a system of diene monomer (Cyclopentadienyl). The local similarity index based on the Hirshfeld partitioning in the framework of conceptual Density Functional Theory (DFT), was introduced in the secular determinant of the Hückel method and was applied to Pyrrole molecule in order to express their orbital energies as a function of the orbital energy of Cyclopentadienyl, to express the energies of molecular orbitals of the Cyclopentadienyl as a function of Pentalene, resolved by the Hückel method and applied to Cyclopentadienyl, by means of six local similarity index: Overlap, Overlap-Interaction, Coulomb, Coulomb-Interaction, with their respective Euclidean distances, using the Topo-Geometrical Superposition Approach (TGSA) as a method of alignment, which allowed us to obtain good results in local similarity indices.

This technique will permit the study of some molecular systems that differ in one atom in its molecular structure, resolving the Hückel method for the Pyrrole and Thiophene system without taking into account the considerations with its neighboring atoms. This proposed technique reduces the symmetry of fused ring systems which are Cyclopentadienyl derivatives, allowing to express the orbital energy of a diene dimmer (Pentalene) as a function of diene monomer systems, creating a tool of calculation to solve the problem of obtaining non-degenerate values in systems where the approximations in the Hückel method approximation provide degenerate values and providing a symmetry reduction technique.

Recommendation systems are becoming more and more present in our daily lives, whether it is for suggesting items to buy, movies to watch or music to listen. They can be used in a large number of contexts. In this paper, we propose the use of a recommendation system in the context of a recruitment platform. The use of the recommendation system allows to obtain precise profile recommendations based on the competences of a candidate to meet the stated requirements and to avoid companies to have to perform a very time-consuming manual sorting of the candidates. Thus, this paper presents the context in which we propose this recommendation system, the data preprocessing, the general approach based on a hybrid content-based filtering (CBF) and similarity index (SI) system, as well as the means implemented to reduce the computational cost of such a system with the increasing evolution of the platform.

Seismic performance assessment of highway is important to mitigate future earthquake disaster. In general, the seismic motion at a site is different from that at another, and seismic zoning is often applied to consider the difference in anticipated input motions. H/V spectrum of microtremor measurement is a technique to clarify the difference in anticipated strong motions at the sites. However, there is no quantitative index for the difference in H/V spectrum, which was caused by the difference in site amplification characteristics. In this study, a quantitative index for the similarity of H/V spectrum was proposed. The proposed method is applied to several sites along a highway in Japan, showing its applicability with more reasonable judgment in seismic zoning.