Link-based Similarity Measures Research Articles

Landslide Susceptibility Mapping Using DIvisive ANAlysis (DIANA) and RObust Clustering Using linKs (ROCK) Algorithms, and Comparison of Their Performance

Landslide susceptibility mapping (LSM) studies provide essential information that helps various authorities in managing landslide-susceptible areas. This study aimed at applying and comparing the performance of DIvisive ANAlysis (DIANA) and RObust Clustering using linKs (ROCK) algorithms for LSM in the Baota District, China. These methods can be applied when the data has no labels and when there is insufficient inventory data. First, based on historical records, survey reports, and previous studies, 293 landslides were mapped in the study area and 7 landslide-influencing attributes were selected for modeling. Second, the methods were clustered in the study area mapping units into 469 and 476 subsets, respectively; for mapping landslide susceptibility, the subsets were classified into 5 susceptibility levels through the K-means method using landslide densities and attribute values. Then, their performances were assessed and compared using statistical metrics and the receiver operating curve (ROC). The outcomes indicated that similarity measures influenced the accuracy and the predictive power of these clustering models. In particular, when using a link-based similarity measure, the ROCK performed better with overall performance accuracy of 0.8933 and an area under the curve (AUC) of 0.875. The maps constructed from the models can be useful in landslide assessment, prevention, and mitigation strategies in the study area, especially for areas classified with higher susceptibility levels. Moreover, this comparison provides a new perspective in the selection of a considerable model for LSM in the Baota District.

Reconstruction of international energy trade networks with given marginal data: A comparative analysis

The energy commodity plays a significant role in economic development and national security. It is important for various countries to capture the linkages of the international energy trade network and avoid the occurrence of systemic risk. However, the granular data on trade network is often lacking, and instead, the network has to be reconstructed from partial data. In this paper, we compare seven network reconstruction methods by applying them to 16 types of energy trade data extracted from UN-Comtrade. To this end, first we present the topological structures and compute the dependency characteristics for these energy trade networks. Then, we conduct a horse race among different network reconstruction methods. The methods are ranked by comparing their abilities to reproduce the network structures and bilateral weights. Our findings show that Cimi is the winner across all 16 energy trade networks in terms of link-based similarity measures. In addition, the fitness models (Cimi and Musm) tend to perform best at reconstructing the weight matrix. These findings will aid in the accurate disclosure of latent international energy trade relations when only aggregated export and import figures are available, allowing systemic risks to be appropriately assessed.

Scaling High-Quality Pairwise Link-Based Similarity Retrieval on Billion-Edge Graphs

SimRank is an attractive link-based similarity measure used in fertile fields of Web search and sociometry. However, the existing deterministic method by Kusumoto et al. [ 24 ] for retrieving SimRank does not always produce high-quality similarity results, as it fails to accurately obtain diagonal correction matrix D . Moreover, SimRank has a “connectivity trait” problem: increasing the number of paths between a pair of nodes would decrease its similarity score. The best-known remedy, SimRank++ [ 1 ], cannot completely fix this problem, since its score would still be zero if there are no common in-neighbors between two nodes. In this article, we study fast high-quality link-based similarity search on billion-scale graphs. (1) We first devise a “varied- D ” method to accurately compute SimRank in linear memory. We also aggregate duplicate computations, which reduces the time of [ 24 ] from quadratic to linear in the number of iterations. (2) We propose a novel “cosine-based” SimRank model to circumvent the “connectivity trait” problem. (3) To substantially speed up the partial-pairs “cosine-based” SimRank search on large graphs, we devise an efficient dimensionality reduction algorithm, PSR # , with guaranteed accuracy. (4) We give mathematical insights to the semantic difference between SimRank and its variant, and correct an argument in [ 24 ] that “if D is replaced by a scaled identity matrix (1-Ɣ)I, their top-K rankings will not be affected much”. (5) We propose a novel method that can accurately convert from Li et al. SimRank ~{S} to Jeh and Widom’s SimRank S . (6) We propose GSR # , a generalisation of our “cosine-based” SimRank model, to quantify pairwise similarities across two distinct graphs, unlike SimRank that would assess nodes across two graphs as completely dissimilar. Extensive experiments on various datasets demonstrate the superiority of our proposed approaches in terms of high search quality, computational efficiency, accuracy, and scalability on billion-edge graphs.

JacSim*: An Effective and Efficient Solution to the Pairwise Normalization Problem in SimRank

Despite the fact that SimRank has been successfully applied to various applications as a link-based similarity measure, it suffers from a counter-intuitive property called a pairwise normalization problem ; JacSim is a powerful variant of SimRank that alleviates this problem. In this paper, we first point out three existing drawbacks of JacSim and then propose JacSim* to effectively solve them; JacSim* exploits those paths neglected by JacSim in similarity computation, its matrix form provides the exact similarity scores while not being sensitive to the number of node-pairs with common neighbors, and it has simpler, easier to understand, and easier to implement formulas in both iterative and matrix forms than those of JacSim. We conduct extensive experiments with eight real-world datasets to evaluate both the accuracy and performance of JacSim* in comparison with those of JacSim. Our experimental results demonstrate that JacSim* shows better accuracy than JacSim and the JacSim* matrix form is dramatically faster than its own iterative form and also than the two forms of JacSim with all datasets.

On Investigating Both Effectiveness and Efficiency of Embedding Methods in Task of Similarity Computation of Nodes in Graphs

One of the important tasks in a graph is to compute the similarity between two nodes; link-based similarity measures (in short, similarity measures) are well-known and conventional techniques for this task that exploit the relations between nodes (i.e., links) in the graph. Graph embedding methods (in short, embedding methods) convert nodes in a graph into vectors in a low-dimensional space by preserving social relations among nodes in the original graph. Instead of applying a similarity measure to the graph to compute the similarity between nodes a and b, we can consider the proximity between corresponding vectors of a and b obtained by an embedding method as the similarity between a and b. Although embedding methods have been analyzed in a wide range of machine learning tasks such as link prediction and node classification, they are not investigated in terms of similarity computation of nodes. In this paper, we investigate both effectiveness and efficiency of embedding methods in the task of similarity computation of nodes by comparing them with those of similarity measures. To the best of our knowledge, this is the first work that examines the application of embedding methods in this special task. Based on the results of our extensive experiments with five well-known and publicly available datasets, we found the following observations for embedding methods: (1) with all datasets, they show less effectiveness than similarity measures except for one dataset, (2) they underperform similarity measures with all datasets in terms of efficiency except for one dataset, (3) they have more parameters than similarity measures, thereby leading to a time-consuming parameter tuning process, (4) increasing the number of dimensions does not necessarily improve their effectiveness in computing the similarity of nodes.

An effective similarity measure based on kernel spectral method for complex networks

Node similarity measure is a special important task in complex network analysis and plays a critical role in a multitude of applications, such as link prediction, community detection, and recommender systems. In this study, we are interested in link-based similarity measures, which only concern the structural information of networks when estimating node similarity. A new algorithm is proposed by adopting the idea of kernel spectral method to quantify the similarity of nodes. When computing the kernel matrix, the proposed algorithm makes use of local structural information, but it takes advantage of global information when constructing the feature matrix. Thence, the proposed algorithm could better capture potential relationships between nodes. To show the superiority of our algorithm over others, we conduct experiments on 10 real-world networks. Experimental results demonstrate that our algorithm yields more reasonable results and better performance of accuracy than baselines.

JacSim: An accurate and efficient link-based similarity measure in graphs

SimRank is a well-known link-based similarity measure successfully applied in many graph-related applications. Despite of the current success of SimRank, it suffers from the problem caused by its pairwise normalization paradigm in similarity computation. In this paper, we propose JacSim (Jaccard-based SimRank) that solves the pairwise normalization problem in an effective way. JacSim computes the similarity score of a node-pair by combining two different computation manners: Jaccard coefficient and pairwise normalization. We point out two problems of existing measures targeted at solving the pairwise normalization problem and provide effective solutions to them: (1) JacSim eliminates the redundancy hidden in their similarity computation; (2) JacSim enables to control the degree of importance of the two scores obtained by employing Jaccard coefficient and pairwise normalization. In order to take advantage of links weights in similarity computation, we propose a weighted version of JacSim applicable to weighted graphs. Furthermore, to accelerate JacSim, we provide a linear recursive matrix form of JacSim, which is composed of only linear operations. We demonstrate the effectiveness and efficiency of our JacSim by conducting extensive experiments with real-world datasets. The results show that JacSim outperforms existing measures significantly in term of accuracy and also provides better performance than the similarity measures targeted to solve the pairwise normalization problem.

SimCC: A novel method to consider both content and citations for computing similarity of scientific papers

To compute the similarity of scientific papers, text-based similarity measures, link-based similarity measures, and hybrid methods can be applied. The text-based and link-based similarity measures take into account only a single aspect of scientific papers, content or citations, respectively. The hybrid methods consider both content and citations; however, they do not carefully consider the relation between the content of a pair of papers involved in a citation relationship. In this paper, we propose a novel method, SimCC (similarity based on content and citations), that considers both aspects, content and citations, to compute the similarity of scientific papers. Unlike previous methods, SimCC effectively reflects both content and authority of scientific papers simultaneously in similarity computation by applying a new RA (relevance and authority) weighting scheme. Also, we propose an RA+R weighting scheme to consider the recency of papers and an RA+E weighting scheme to take into account the author expertise of papers in similarity computation. The effectiveness of our proposed method is demonstrated by extensive experiments on a real-world dataset of scientific papers. The results show that our method achieves more than 100% improvement in accuracy in comparison with previous methods.

C-Rank: A link-based similarity measure for scientific literature databases

As the number of people who use scientific literature databases has grown, the demand for literature retrieval services has steadily increased. One of the most popular retrieval service methods is to find a set of papers similar to the paper under consideration, which requires a measure that computes the similarities between the papers. Scientific literature databases exhibit two interesting characteristics that are not found in general databases. First, the papers cited by older papers are often not included in the database due to technical and economic reasons. Second, since a paper references previously published papers, few papers cite recently published papers. These two characteristics cause all existing similarity measures to fail in at least one of the following cases: (1) measuring the similarity between old, but similar papers, (2) measuring the similarity between recent, but similar papers, and (3) measuring the similarity between two similar papers: one old, the other recent. In this paper, we propose a new link-based similarity measure called C-Rank, which uses both in-link and out-link references, disregarding the direction of the references. In addition, we discuss the most suitable normalization method for scientific literature databases and we propose an evaluation method for measuring the accuracy of similarity measures. For the experiments, we used real-world papers from DBLP’s database with reference information crawled from Libra. We then compared the performance of C-Rank with that of existing similarity measures. Experimental results showed that C-Rank achieved a higher accuracy than existing similarity measures.

Accuracy estimation of link-based similarity measures and its application

Link-based similarity measures play a significant role in many graph based applications. Consequently, measuring node similarity in a graph is a fundamental problem of graph datamining. Personalized pagerank (PPR) and simrank (SR) have emerged as the most popular and influential link-based similarity measures. Recently, a novel link-based similarity measure, penetrating rank (P-Rank), which enriches SR, was proposed. In practice, PPR, SR and P-Rank scores are calculated by iterative methods. As the number of iterations increases so does the overhead of the calculation. The ideal solution is that computing similarity within the minimum number of iterations is sufficient to guarantee a desired accuracy. However, the existing upper bounds are too coarse to be useful in general. Therefore, we focus on designing an accurate and tight upper bounds for PPR, SR, and P-Rank in the paper. Our upper bounds are designed based on the following intuition: the smaller the difference between the two consecutive iteration steps is, the smaller the difference between the theoretical and iterative similarity scores becomes. Furthermore, we demonstrate the effectiveness of our upper bounds in the scenario of top-k similar nodes queries, where our upper bounds helps accelerate the speed of the query. We also run a comprehensive set of experiments on real world data sets to verify the effectiveness and efficiency of our upper bounds.

Link-Based Similarity Measures Using Reachability Vectors

We present a novel approach for computing link-based similarities among objects accurately by utilizing the link information pertaining to the objects involved. We discuss the problems with previous link-based similarity measures and propose a novel approach for computing link based similarities that does not suffer from these problems. In the proposed approach each target object is represented by a vector. Each element of the vector corresponds to all the objects in the given data, and the value of each element denotes the weight for the corresponding object. As for this weight value, we propose to utilize the probability of reaching from the target object to the specific object, computed using the “Random Walk with Restart” strategy. Then, we define the similarity between two objects as the cosine similarity of the two vectors. In this paper, we provide examples to show that our approach does not suffer from the aforementioned problems. We also evaluate the performance of the proposed methods in comparison with existing link-based measures, qualitatively and quantitatively, with respect to two kinds of data sets, scientific papers and Web documents. Our experimental results indicate that the proposed methods significantly outperform the existing measures.

Measuring Similarity Based on Link Information: A Comparative Study

Measuring similarity between objects is a fundamental task in domains such as data mining, information retrieval, and so on. Link-based similarity measures have attracted the attention of many researchers and have been widely applied in recent years. However, most previous works mainly focus on introducing new link-based measures, and seldom provide theoretical as well as experimental comparisons with other measures. Thus, selecting the suitable measure in different situations and applications is difficult. In this paper, a comprehensive analysis and critical comparison of various link-based similarity measures and algorithms are presented. Their strengths and weaknesses are discussed. Their actual runtime performances are also compared via experiments on benchmark data sets. Some novel and useful guidelines for users to choose the appropriate link-based measure for their applications are discovered.

An Accurate Revelation of the Similarity between Clusters

The structure of the data set playing a vital role in datamining. In concept of datamining information recovery and pattern identification nothing but data clustering. There are multiple clustering algorithms have been commenced to clustering categorical data. Unfortunately these algorithms created an incomplete information. In recent times cluster ensembles have come out as an essential solution to overcome these limitations and to get the excellence results for clustering. A Link-Based similarity measure is proposed to guess unknown values from a link network of clusters and bridges the gap among the task of data clustering and that link examination. It also improves the ability of ensemble methodology for categorical data. A new LinkBased cluster ensemble approach is commenced which is well-organized than the previous model, where a binary cluster association matrix, like matrix is used to create the cluster ensembles. These cluster ensembles have impurity information, to overcome these problem Link-Based similarity algorithm is used to generate an accurate pure clusters.

An improved clustering ensemble method based link analysis

Clustering Ensemble aggregates several base clustering analyses into a consensus clustering result, which is more accurate, stable and meaningful than standard clustering algorithm. In this paper, the ensemble information is described by data cluster association matrix. However, most data cluster association matrix overlooks an important type of information about the relationship between clusters. This paper proposes a new method WETU to refine the data cluster association matrix with link-based similarity measure. The refined data cluster association matrix is obtained according to the similarity of clusters among all base clustering results, not in one base clustering result. In addition, WETU can provide more discriminative information than CSM and WTU. The data cluster association matrix is refined into high level real-valued matrix, which can be aggregated by real-valued method, such as Global k-means. Experiments on synthetic dataset and UCI datasets show that the proposed method outperforms standard K-means, base clustering algorithm and CSM+Global k-means and WTU+Global k-means.T