Mining subtopics from weblogs and analyzing their spatiotemporal patterns have applications in multiple domains. In this paper, we define the novel problem of mining theme propose a probabilistic approach to model subtopic themes simultaneously. The proposed discovers by (1) extracting common weblogs; (2) generating life cycles for each given location; (3) snapshots time period. Evolution can be discovered comparative analysis snapshots. Experiments on three different data sets show that...

Previous chapter Next Full AccessProceedings Proceedings of the 2005 SIAM International Conference on Data Mining (SDM)Mining Behavior Graphs for "Backtrace" Noncrashing BugsChao Liu, Xifeng Yan, Hwanjo Yu, Jiawei Han, and Philip S. YuChao Yupp.286 - 297Chapter DOI:https://doi.org/10.1137/1.9781611972757.26PDFBibTexSections ToolsAdd to favoritesExport CitationTrack CitationsEmail SectionsAboutAbstract Analyzing executions a buggy software program is essentially data mining process. Although...

Fast and noise-suppressed incoherent coded aperture correlation holographic imaging is proposed, which utilized by employing an annular sparse phase mask together with adaptive phase-filter cross-correlation reconstruction method. Thus the proposed technique here coined as interferenceless holography (AI-COACH). In AI-COACH, first designed generated Gerchberg-Saxton algorithm for suppressing background noise during reconstruction. order to demonstrate three-dimensional sectional capabilities...

With the rapid depletion of fossil fuels and increasing concern about climate change, photovoltaic systems have become a key technology option for realizing decarbonized power sector sustainable energy supply. The global trade in cells has increased dramatically last two decades deserves dedicated investigation. First, from static analysis perspective, this study builds cell network competition 2000 to 2019 analyzes characteristics pattern trade. Second, dynamic potential impacts coronavirus...

Evolutionary algorithms have shown their promise in coping with many-objective optimization problems. However, the strategies of balancing convergence and diversity effectiveness handling problems irregular Pareto fronts (PFs) are still far from perfect. To address these issues, this paper proposes an adaptive sorting-based evolutionary algorithm based on idea decomposition. First, we propose environmental selection strategy. Solutions each subpopulation (partitioned by reference vectors)...

With respect to multicriteria group decision‐making (MCGDM) problems in which the experts have different priority levels, criteria values are form of Pythagorean fuzzy uncertain linguistic variables (PFULVs), and information about weights is completely unknown, a novel method developed. Firstly, concept PFULV defined, some operational laws, score function, accuracy normalized Hamming distance PFULVs presented. Then, aggregate given by all experts, prioritized weighted averaging aggregation...

Drawing on the identity perspective, we investigate roles of organizational and moral identification in relationship between abusive supervision subordinates' deviance. By conducting a multiwave study 182 subordinate–supervisor dyads, found that partially mediated deviance after controlling for perceived leader social support as an alternative mediator. In addition, demonstrated interacted with affecting workplace moderated indirect effect via identification. However, did not find mediating...

Lwamo et al. recently proposed a robust and efficient remote single multi-server biometric authentication scheme using smart card RSA. The is vulnerable to the lost attacks; therefore, cannot resist offline guessing attacks user impersonation attacks, provide forward security anonymity. To address these issues, we propose new privacy-preserving ring learning with errors (RLWE)-based (RRBAS) for environments. RRBAS first lattice-based Security analysis show that can satisfy authenticated key...

Purpose This paper aims to explore the characteristics of knowledge diffusion library and information science reveal its development trend influence on other disciplines. Design/methodology/approach Based ESI discipline classification, this measures from disciplines over last 24 years using indicators in four dimensions: breadth, intensity, speed theme diffusion. Findings The results show that breadth is wide, spreading 21 disciplines; spread mainly concentrates soft or applied disciplines,...

Purpose Steered by upper echelon theory, this study aims to scrutinize the prevalence of project manager demographic factors (age, education and experience) in sustainability management performance. Design/methodology/approach We used a sample 209 managers/supervisor/team leaders who were working projects China-Pakistan Economic Corridor (CPEC). Findings The results indicate that have significant influence on performance (except management. Moreover, partially mediates relationship between...

Promoting close and sustainable cooperation between schools, enterprises, government has become an important concern in many countries. Based on evolutionary game theory, this paper constructs a tripartite model of order to analyze the stability strategies different players. The results show that main factor influences schools enterprises is reward positive from sources other than government, strategy benefit under scenario where cooperate with enterprises. Therefore, should focus...

Conventional research on P2P incentive mechanism is based cooperative or non-cooperative game theory. This paper assumes the node in system limited rationality, using evolutionary theory to study mechanism, we first investigate individual population model and replicator dynamics. Mathematics analysis numerical result shows that stable when plays as designed. Then condition with migration ability. We assume divided into different subpopulation according dynamic property, dynamics, prove even...

Virtual instrument system has received amounts of attention and become a common practice in many fields. is program that implements functions an by software which could replace the work real to save resources. The these systems are limited they commonly have little related information about sensors, navigation, monitoring objects due special requirements. procedure development integration often suffers from low efficiency because non-standard technologies. In this paper, we propose complex...

There have been more than 12 million open source software in GitHub so far, but the existing studies about sentiments analysis are not sufcient for GitHub. This paper proposes an approach to analyze correlation of comment and bug fixing speed. It's proved existence certain factors among some relevance after experiments. The experimental results also give suggestions on development process.

In this paper, we establish a Bernstein-type inequality for widely orthant dependent random variables, and obtain the rates of strong convergence kernel estimators density hazard functions, under some suitable conditions.

Named Entity Recognition (NER) is a basic task of natural language processing and an indispensable part machine translation, knowledge mapping other fields. In this paper, fusion model Chinese named entity recognition using BERT, Bidirectional LSTM (BiLSTM) Conditional Random Field (CRF) proposed. model, BERT generates word vectors as embedding model. Word through BiLSTM can learn the label distribution. Finally, uses Fields to make syntactic restrictions at sentence level get annotation...