A Novel PageRank Based Fault Handling Strategy for Workflow Scheduling in Cloud Data Centers

Unexpected faults result in unscheduled cloud outage, which negatively affects the completion of workflow tasks in the cloud. This paper presents a novel PageRank based fault handling strategy to rescue workflow tasks at the faulty data center. The proposed approach uses a holistic view and considers the task attributes, the timeline scenario, and the overall cloud performance. A priority assignment system is developed based on the modified PageRank algorithm to prioritise workflow tasks. A Min-Max normalization method is applied to select the target data center and match the timeline at this data center. Additionally, a dynamic PageRank-constrained task scheduling algorithm is proposed to generate the task scheduling solution. The simulation results show that the proposed approach can achieve better fault handling performance, measured by task resilience ratio, workflow resilience ratio and workflow continuity ratio, in both the traditional 3-replica and the image backup cloud environment.

Verification of composed web service using Synthesized Nondeterministic Turing Model (SNTMM) with Multiple Tapes and Stacks

To verify the composed Web services, a general view of what traits of a service need to be identified is still lacking. The existing verification model did not address any mechanism for getting alternative services if we failed to reach the desired service and partially concentrated on the reachability problem for a deterministic and non-deterministic system in sequential. This paper proposes a Synthesised Non-deterministic Turing Machine Model (SNTMM) by combining the Multistacked Non-deterministic Turing Machine (MSNTM) model and Multitaped Non-deterministic Turing Machine (MTNTM) model to verify the composed Web services for both deterministic and non-deterministic systems in parallel. The deceased transition and departed service marking algorithm have been proposed to address each participated service’s reachability in composing service for all possible input in parallel. This article shows an example to demonstrate the meticulousness of the model. The experimental results show that the performance of the proposed model is measured efficiently

A graph neural network based algorithm for Point-of-Interest recommendation using social relation and time series

POI recommendation has gradually become an important topic in the field of service recommendation, which is always achieved by mining user behavior patterns. However, the context information of the collaborative signal is not encoded in the embedding process of traditional POI recommendation methods, which is not enough to capture the collaborative signal among different users. Therefore, a POI recommendation algorithm is presented by using social-time context graph neural network model (GNN) in Location-based social networks. First, it finds similarities between different social relationships based on the users' social and temporal behavior. Then, the similarity among different users is calculated by an improved Euclidean distance. Finally, it integrates the graph neural network, the interaction bipartite graph of users and social-time information into the algorithm to generate the final recommendation list in this paper. Experiments on real datasets show that the proposed method is superior to the state-of-the-art POI recommendation methods.

Intelligent and Adaptive Web Page Recommender System

In this manuscript, an Intelligent and Adaptive Web Page Recommender System is proposed that provides personalized, global and group mode of recommendations. The authors enhance the utility of a trie node for storing relevant web access statistics. The trie node enables dynamic clustering of users based on their evolving browsing patterns and allows a user to belong to multiple groups at each navigation step. The system takes cues from the field of crowd psychology to augment two parameters for modeling group behavior: Uniformity and Recommendation strength. The system continuously tracks the user’s responses in order to adaptively switch between different recommendation-criteria in the group and personalized modes. The experimental results illustrate that the system achieved the maximum F1 measure of 83.28% on CTI dataset which is a significant improvement over the 70% F1 measure reported by Automatic Clustering-based Genetic Algorithm, the prior web recommender system.

Bottleneck-Aware Resource Allocation for Service Processes

In 5G systems and beyond, traditional generic service models are no longer appropriate for highly customized and intelligent services. The process of reinventing service models involves allocating available resources, where the performance of service processes is determined by the activity node with the lowest service rate. This paper proposes a new bottleneck-aware resource allocation approach by formulating the resource allocation as a max-min problem. The approach can allocate resources proportional to the workload of each activity, which can guarantee that the service rates of activities within a process are equal or close-to-equal. Based on the business process simulator (i.e., BIMP) simulation results show that the approach is able to reduce the average cycle time and improve resource utilization, as compared to existing alternatives. The results also show that the approach can effectively mitigate the impact of bottleneck activity on the performance of service processes.

Model-Driven Open Ecological Cloud Enterprise Resource Planning

Cloud enterprise resource planning (Cloud ERP) is an internet- and cloud computing-based enterprise information system developed on the cloud platform. Cloud ERP has lower costs and shorter development time compared with traditional ERP system, but it remains in a state of information isolated island. To maximize the advantages of cloud computing and make up the deficiency of traditional ERP systems, it is necessary to break down the "wall" between enterprises, making cloud ERP enter a more open and interconnected ecological environment. The model-driven development approach contributes to a better resilient scheduling capability of ERP system, leading to faster development and deployment of it. In this article, the authors propose a “knowledge + data” model-driven open ecological cloud ERP and explain the definition and functions of each model layer. Finally, the effectiveness of model layers is demonstrated in the open ecological cloud ERP reference architecture.

Web Service Clustering Approach Based on Network and Fused Document-Based and Tag-Based Topics Similarity

To produce a web services clustering with values that satisfy many requirements is a challenging focus. In this article, the authors proposed a new approach with two models, which are helpful to the service clustering problem. Firstly, a document-tag LDA model (DTag-LDA) is proposed that considers the tag information of web services, and the tag can describe the effective information of documents accurately. Based on the first model, this article further proposes an efficient document weight and tag weight-LDA model (DTw-LDA), which fused multi-modal data network. To further improve the clustering accuracy, the model constructs the network for describing text and tag respectively and then merges the two networks to generate web service network clustered. In addition, this article also designs experiments to verify that the used auxiliary information can help to extract more accurate semantics by conducting service classification. And the proposed method has obvious advantages in precision, recall, purity, and other performance.

Probabilistic Graph Model Mining User Affinity in Social Networks

In social networks, discovery of user similarity is the basis of social media data analysis. It can be applied to user-based product recommendations and inference of user relationship evolution in social networks. In order to effectively describe the complex correlation and uncertainty for social network users, the accuracy of similarity discovery is improved theoretically for massive social network users. Based on the Bayesian network probability map model, network topological structure is combined with the dependency between users, and an effective method is proposed to discover similarity in social network users. To improve the scalability of the proposed method and solve the storage and computation problem of mass data, Bayesian network distributed storage and parallel reasoning algorithm is proposed based on Hadoop platform in this paper. Experimental results verify the efficiency and correctness of the algorithm.

Rider Chaotic Biography Optimization-driven Deep Stacked Auto-encoder for Big Data Classification Using Spark Architecture

The novelty in big data is rising day-by-day in such a way that the existing software tools face difficulty in supervision of big data. Furthermore, the rate of the imbalanced data in the huge datasets is a key constraint to the research industry. Thus, this paper proposes a novel technique for handling the big data using Spark framework. The proposed technique undergoes two steps for classifying the big data, which involves feature selection and classification, which is performed in the initial nodes of Spark architecture. The proposed optimization algorithm is named rider chaotic biography optimization (RCBO) algorithm, which is the integration of the rider optimization algorithm (ROA) and the standard chaotic biogeography-based optimisation (CBBO). The proposed RCBO deep-stacked auto-encoder using Spark framework effectively handles the big data for attaining effective big data classification. Here, the proposed RCBO is employed for selecting suitable features from the massive dataset.

ResumeVis Interactive Visualization of Resumes Based on Multi-Source Data

Resumes are critical for individuals to find jobs and for HR to select staffs. To explore the career patterns and demographic information correlation, 372,829 Chinese resumes working in Beijing in 2015 are collected with rich attributes. Besides, 1,837,281 documents in the People's Daily from May 1946 to December 2015 and the national college entrance examination scores of 42 majors in 27 Beijing universities from 2005 to 2015 are collected to build the multi-source dataset to assist resume data mining. The decade characteristics and major characteristics are explored from the multi-source dataset. Based on the data observation, an interactive visualization system called ResumeVis is developed to explore career patterns in the context of the times, especially the correlations among the resume attributes. The system is helpful for both job seekers and human resources.