Narrow Results

It is clear that the topological structure of a neural network somehow determines the activity of the neurons within it. In the present work, we ask to what extent it is possible to examine the structural features of a network and learn something about its activity? Specifically, we consider how the centrality (the importance of a node in a network) of a neuron correlates with its firing rate. To investigate, we apply an array of centrality measures, including In-Degree, Closeness, Betweenness, Eigenvector, Katz, PageRank, Hyperlink-Induced Topic Search (HITS) and NeuronRank to Leaky-Integrate and Fire neural networks with different connectivity schemes. We find that Katz centrality is the best predictor of firing rate given the network structure, with almost perfect correlation in all cases studied, which include purely excitatory and excitatory–inhibitory networks, with either homogeneous connections or a small-world structure. We identify the properties of a network which will cause this correlation to hold. We argue that the reason Katz centrality correlates so highly with neuronal activity compared to other centrality measures is because it nicely captures disinhibition in neural networks. In addition, we argue that these theoretical findings are applicable to neuroscientists who apply centrality measures to functional brain networks, as well as offer a neurophysiological justification to high level cognitive models which use certain centrality measures.

Network topology is the basic for the development of traffic management and control. In a road network, bayonets with installation of surveillance facilities are key components to recognize traffic congestion from time to time. Therefore, identifying the essential bayonets in a road network becomes one of the most efficient ways to alleviate traffic congestion for traffic engineers and transport department. To do so, this paper aims to propose a novel sorting algorithm based on similarity measurements and traffic flow information to efficiently identify key bayonets in road networks. Our research results show that by analyzing the bayonet data in a fixed period of time in a medium-sized city of China, we have successfully identified the location of key bayonet points. Most of these key bayonet points are closed to residential areas and important traffic stations. The rank of these bayonet points can help the city managers better understand the topological characteristics of the road network as well as the propagation of congestion so as to make the traffic policies or control strategies for traffic congestion alleviation.

Understanding the structural characteristics of urban traffic network comprehensively can provide references for improving road utilization rate and alleviating traffic congestion. This paper focuses on the spatial-temporal correlations between different pairs of traffic series and proposes a complex network-based method of constructing the urban traffic network. In the network, the nodes represent road segments, and an edge between a pair of nodes is added depending on the result of significance test for the corresponding spatial-temporal correlation. Further, a modified PageRank algorithm, named the geographical weight-based PageRank algorithm (GWPA), is proposed to analyze the spatial distribution of important segments in the road network. Finally, experiments are conducted by using three kinds of traffic series collected from the urban road network in Beijing. Experimental results show that the urban traffic networks constructed by three traffic variables all indicate both small-world and scale-free characteristics. Compared with the results of PageRank algorithm, GWPA is proved to be valid in evaluating the importance of segments and identifying the important segments with small degree.

In this paper, we present a model to classify users of Social Networks. In particular, we focus on Social Network Sites. The model is based on the PageRank algorithm. We use the personalization vector to bias the PageRank to some users. We give an explicit expression of the personalization vector that allows the introduction of some typical features of the users of SNSs. We describe the model as a seven-step process. We illustrate the applicability of the model with two examples. One example is based on real links of a Facebook network. We also indicate how to take into account real actions of Facebook users to implement the model.

During software maintenance, bug reports are widely employed to improve the software project’s quality. A developer often refers to stowed bug reports in a repository for bug resolution. However, this reference process often requires a developer to pursue a substantial amount of textual information in bug reports which is lengthy and tedious. Automatic summarization of bug reports is one way to overcome this problem. Both supervised and unsupervised methods are effectively proposed for the automatic summary generation of bug reports. However, existing methods disregard the significance of duplicate bug reports in summarizing bug reports. In this study, we propose a PageRank-based Summarization Technique (PRST), which utilizes the textual information contained in bug reports and additional information in associated duplicate bug reports. PRST uses three variants of PageRank-based on Vector Space Model (VSM), Jaccard, and WordNet similarity metrics. These variants are utilized to calculate the textual similarity of the sentences between the master bug reports and their duplicates. PRST further trains a regression model and predicts the probability of sentences belonging to the summary. Finally, we combine the values of PageRank and regression model scores to rank the sentences and produce the summary for the master bug reports. In addition, we construct two corpora of bug reports and duplicates, i.e. MBRC and OSCAR. Empirical results suggest that PRST outperforms the state-of-the-art method BRC in terms of Precision, Recall, F-score, and Pyramid Precision. Meanwhile, PRST with WordNet achieves the best results against PRST with VSM and Jaccard.

Programming education is becoming increasingly popular in universities. However, due to a lack of debugging experience, novices often encounter numerous difficulties in the programming process. Automatic fault localization techniques have emerged as a promising solution to address this issue. Among these techniques, Spectrum-Based Fault Localization (SBFL) and Mutation-Based Fault Localization (MBFL) have been widely used in industrial programs. However, there is a significant difference between industrial and novice programs and the performance of these methods on novice programs has not been extensively studied.

To fill this gap, we conducted an empirical study to evaluate the fault localization performance and execution overhead of SBFL and MBFL in a typical novice programming environment. Our study specifically examined how different program characteristics, including code coverage and mutation score, affect the accuracy of these localization methods. Additionally, during the study, we identified the tie problem in both methods and further investigated its impact on fault localization techniques in novice programs.

To remove the impact of the tie problem, we proposed using PageRank scores as weights for the suspiciousness, sorting, and locating faults based on the weighted suspiciousness. The PageRank algorithm is based on statement coverage information and constructs a directed graph. From the directed graph, a transition matrix generates the weight scores (PageRank scores) for each statement.

Our research demonstrates that both SBFL and MBFL are effective for fault localization in novice programs, with MBFL showing significantly better performance in our tests. In TOP-$N(N=1,3,5)$, MBFL accurately locates 67, 96 and 114 faults, respectively, indicating superior performance. Additionally, calculating weighted suspiciousness significantly alleviates the tie problem.

The massive and heterogeneous Web exacerbates IR problems and short user queries make them worse. The contents of web pages are not enough to find target pages. PageRank compensates for the insufficiencies of content information. The content information and PageRank are combined to get better results. To get up-to-date results in the dynamic environment of the Web, an index database must be updated continuously. Since PageRank deals with the full link structure of the web pages, this task is time consuming and expensive. In this paper, an incremental PageRanking scheme is proposed. This scheme assigns PageRanks to newly crawled web pages based on their structure, not on the full link structure of the total web pages. This approach avoids the expensive task of updating the link structure of web pages when changes occur. The basic assumption here is that the existing set of web pages is sufficiently large so that the newly crawled pages do not make any big changes to it. PageRanks of the previous pages are propagated using the links of the current pages. The proposed incremental PageRanking has been tested for various crawling methods. Test results indicate that crawling by the order of the PageRank achieves the best result.

This paper explores citation-based metrics, how they differ in ranking papers and authors, and why. We initially take as example three main metrics that we believe significant; the standard citation count, the more and more popular h-index, and a variation we propose of PageRank applied to papers (called PaperRank), that is appealing as it mirrors proven and successful algorithms for ranking web pages. As part of analyzing them, we develop generally applicable techniques and metrics for qualitatively and quantitatively analyzing indexes that evaluate content and people, as well as for understanding the causes of their different behaviors. Finally, we extend the analysis to other popular indexes, to show whether the choice of the index has a significant effect in how papers and authors are ranked. We put the techniques at work on a dataset of over 260 K ACM papers, and discovered that the difference in ranking results is indeed very significant (even when restricting to citation-based indexes), with half of the top-ranked papers differing in a typical 20-element long search result page for papers on a given topic, and with the top researcher being ranked differently over half of the times in an average job posting with 100 applicants.

Google employs PageRank to rank web pages, determining the order in which search results are presented to users based on their queries. PageRank is primarily utilized for directed networks, although there are instances where it is also applied to undirected networks. In this paper, we have applied PageRank to undirected networks, showing that a vertex’s PageRank relies on its initial value, often referred to as an intrinsic, non-network contribution. We have analytically proved that when the initial value of vertices is either proportional to their degrees or set to zero, the PageRank values of the vertices become directly proportional to their degrees. Simulated and empirical data are employed to bolster our research findings. Additionally, we have investigated the impact of initial values on PageRank localization.

The different web mining methods and techniques can help to solve some typical issues of the contemporary websites, contribute to more effective personalization, improve a website structure and reorganize its web pages. However, only several papers tried to combine web structure and web usage mining (WUM) methods with this aim. The paper researches if and how the combination of selected web structure and WUM methods can identify misplaced web pages and how they can contribute to improving the website structure. The paper analyzes the relationship between the estimated importance of the web page from the web page creator’s point of view using the web structure mining method based on PageRank and visitors’ real perception of the importance of that individual web page using the WUM method based on sequence patterns analysis, which eliminates the problem with repeated visits of the same web page during one session. The results prove that the expected probability of accesses to the individual web page correlates with the observed visit rate obtained from the log files using the WUM method. Furthermore, the website can be improved based on the consequent application of the residual analysis on the obtained results. The applicability of the proposed combination of the web structure and WUM methods is presented on two case studies from different application domains of the contemporary web. As a result, the web pages, which are underestimated or overestimated by the web page creators, are successfully identified in both cases.

Search engines based on content similarity algorithm prevail in Web retrieval field, but they return large volume of Web pages, and therefore, make the user wait longer time. Search engines based on Web hyperlink structure analysis algorithm play a significant role in improving the precision of searching results, but they prone to cause the polarization towards Web pages. To fix these drawbacks, this paper proposes a new searching algorithm that combines the Web content analysis and Web linkage analysis. Based on the PageRank, we introduce an accelerating factor, which can reflect the changeable speed for the importance of a Webpage. Moreover, according to the importance of the two algorithms described above, we present two parameters expressing their weight values. The experiments show that the new integrated algorithm is improved.