Network data are produced automatically by everyday interactions - social networks, power grids, and links between data sets are a few examples. Such data capture social and economic behavior in a form that can be analyzed using powerful computational tools. This book is a guide to both basic and advanced techniques and algorithms for extracting useful information from network data. The content is organized around 'tasks', grouping the algorithms needed to gather specific types of information and thus answer specific types of questions. Examples include similarity between nodes in a network, prestige or centrality of individual nodes, and dense regions or communities in a network. Algorithms are derived in detail and summarized in pseudo-code. The book is intended primarily for computer scientists, engineers, statisticians and physicists, but it is also accessible to network scientists based in the social sciences. Matlab/Octave code illustrating some of the algorithms will be available at: http://www.cambridge.org/9781107125773.

Algorithms and Models for Network Data and Link Analysis

The studies dealing with the problem of predicting scientific impacts in the scientific world mostly focus on predicting citation count of papers (PCCP). However, in the literature, only a little bit of research has been conducted on estimating the future influence of scientists individually. Estimating the impact of scientists individually is a worthwhile task for the following scientific research and cooperatives. From this point of view, a new supervised link prediction method is proposed to predict the citation count of scientists (PCCS). Many PCCP studies employ document-based attributes, such as titles, abstracts, and keywords of papers; institutions of scientists; impact factors of publishers; etc. and they do not take advantage of any topological features of complex networks formed with citations among papers. However, citation networks include valuable features for PCCP and PCCS. Therefore, we formulate the problem of PCCS as a link prediction problem in directed, weighted, and temporal citation networks. The proposed approach predicts not only links but also its weights. Our supervised link prediction method is tested on two citation networks in Experiment 1. The results of Experiment 1 confirm that our method achieves promising performances when considering prediction links with its weights are addressed for the first time in terms of link prediction in directed, weighted, and temporal networks. In Experiment 2, the performance of the proposed link prediction metric and five well-known link prediction metrics are compared in terms of prediction new links in complex networks. The results of Experiment 2 demonstrate that the proposed link prediction metric outperforms all baseline link prediction metrics.

Predicting Citation Count of Scientists as a Link Prediction Problem

Effectively ranking patents in outlierness in a patent citation network is a crucial task for patent analysis, including as it relates to technological opportunity discovery (TOD). Previous studies in the area of TOD focus on patent textual data. In this paper, we introduce a new approach that addresses TOD via patent outlierness, leveraging both patent attributes and citations. We propose the following characteristics for patent outliers: 1) not highly clustered with other patents; 2) low node centrality within the citation network; and 3) low similarity to other patents in the network. Existing outlier ranking approaches have the drawback of not leveraging the unique characteristics of attributed patent citation networks. We propose new outlier ranking methods developed specifically for patents in attributed patent citation networks. Attribute data independently describe a patent, while citation network data relate patents to each other, thus capturing patent outlierness from two different aspects. The contributions of this paper are, given an attributed patent citation network: 1) patent clustering algorithm, and 2) method for scoring and ranking patents in outlierness. Developed methods are validated using artificial datasets. Proposed outlier ranking methods are evaluated using U.S. patents in the area of digital information and security.

Patent Clustering and Outlier Ranking Methodologies for Attributed Patent Citation Networks for Technology Opportunity Discovery

In today's competitive business environment, the timely identification of potential technology opportunities is becoming increasingly important for the strategic management of technology and innovation. Existing studies in the field of technology opportunity discovery (TOD) focus exclusively on patent textual information. In this article, we introduce a new method that tackles TOD via technology convergence, using both patent textual data and patent citation networks. We identify technology groups with high convergence potential by measuring connectivity between clusters of patents. From such technology groups we select pairs of core patents based on their technological relatedness, on their past involvement in convergence, and on the impact of their new potential convergence. We finally carry out TOD by extracting representative keywords from the text of the selected patent pairs and organizing them into the basic description of a new invention, which the potential convergence of the patent pair might produce. We illustrate our proposed method using a data set of U.S. patents in the field of digital information and security.

Inter-cluster connectivity analysis for technology opportunity discovery

This article introduces a method for identifying potential opportunities of innovation arising from the convergence of different technological areas, based on the presence of edge outliers in a patent citation network. Edge outliers are detected via the assessment of their centrality; pairs of patents connected by edge outliers are then analyzed for technological relatedness and past involvement in technological convergence. The pairs with the highest potential for future convergence are finally selected and their keywords combined to suggest new directions of innovation. We illustrate our method on a data set of US patents in the field of digital information and security.

Two-phase edge outlier detection method for technology opportunity discovery

Centrality measures such as degree centrality have been utilized to identify influential and important patents in a citation network. However, no existing centrality measures take into consideration information from the change of the similarity matrix. This paper presents a new centrality measure based on the change of a node similarity matrix. The proposed approach gives more intuitive understanding of the finding of the influential nodes. The present study starts off with the assumption that the change of matrix that may result from removing a given node would assess the importance of the node since each node make a contribution to the given similarity matrix between nodes. The various matrix norms using the singular values such as nuclear norm which is the sum of all singular values, are used for calculating the contribution of a given node to a node similarity matrix. In other words, we can obtain the change of matrix norms for a given node after we calculate the singular values for the case of the nonexistence and the case of existence of the node. Then, the node resulting in the largest change (i.e., decrease) of matrix norms can be considered as the most important node. Computation of singular values can be computationally intensive when the similarity matrix size is large. Therefore, the singular value update technique is also developed for the case of the network with large nodes. We compare the performance of our proposed approach with other widely used centrality measures using U.S. patents data in the area of information and security. Experimental results show that our proposed approach is competitive or even performs better compared to existing approaches.

Automated Detection of Influential Patents Using Singular Values

In today's business environment, competition within industries is becoming more and more intense. To survive in this fast-paced competitive environment, it's important to know what the core patents are and how the patents can be grouped. This study focuses on discovering core patents and clustering patents using a patent citation network in which core patents are represented as an influential node and patent groups as a cluster of nodes. Existing methods have discovered influential nodes and cluster nodes separately, especially in a citation network. This study develops a method used to detect influential nodes (that is, core patents) and clusters (that is, patent groups) in a patent citation network simultaneously rather than separately. The method allows a core patent in each patent group to be discovered easily and the distribution of similar patents around a core patent to be recognized. For this study, kernel k-means clustering with a graph kernel is introduced. A graph kernel helps to compute implicit similarities between patents in a high-dimensional feature space.

Sang Pil Lee

Papers

Automated Detection of Influential Patents Using Singular Values

A Graph Kernel Approach for Detecting Core Patents and Patent Groups