China
Africa
Explore chapters and articles related to this topic
Collective Intelligence in Networking
Published in Phan Cong Vinh, Nature-Inspired Networking: Theory and Applications, 2018
SimRank [37] is used to calculate pair-wise similarity between objects in a network based on the link information. The intuition of the similarity model is based on the idea that “two objects are similar if they are related to similar objects.” In other words, the similarity between objects can be propagated from pair to pairvia links. SimRank can also be applied to bipartite networks, where similarity between one type enhances the quality of the other type alternatively. The time complexity of computing SimRank is high, as the similarity score between a pair of objects is dependent on the similarity between every other pair of objects.
Point-of-interest recommendation using extended random walk with restart on geographical-temporal hybrid tripartite graph
Published in Journal of Spatial Science, 2023
Mozhgan Taheri, Mahdi Farnaghi, Abbas Alimohammadi, Parham Moradi, Samira Khoshahval
Several approaches can be perceived to develop further and improve the accuracy of GHTG-ERWR. Considering both temporal and spatial distances in calculating the weight of session-session edges may influence the recommendation’s accuracy. Incorporating the information derived from different location categories can improve the quality of the recommendation process. Additionally, accounting for the sequence of check-ins in the graph construction phase could improve the algorithm’s performance. Utilization and assessment of other graph-based recommendation algorithms, such as SimRank, to calculate the user similarity in GHTG can be considered future work of this study. A combination of clustering techniques with the graph-based mechanism for POI recommendation is also an open area for future studies. Last but not least, comparing the output of this graph-based POI recommendation approach with other classes of POI recommendation that work based on CF and CB could provide valuable input for further improvements.
An Analysis of Distributed Programming Models and Frameworks for Large-scale Graph Processing
Published in IETE Journal of Research, 2022
Alejandro Corbellini, Daniela Godoy, Cristian Mateos, Silvia Schiaffino, Alejandro Zunino
Some studies have applied MR in graph recommendation algorithms such as Personalized PageRank [9] or SimRank [10]. These studies reported some issues regarding the communication overhead and disk I/O imposed by MR. In particular, when processing iterative algorithms that converge to an stationary value, the results which have already found their final value must be processed repeatedly until the whole computation is completed. This behaviour is caused by the lack of support of MR for iterative algorithms, forcing the user to process all values from the previous iteration to keep them for the next iteration.
Universal manufacturing: enablers, properties, and models
Published in International Journal of Production Research, 2022
Zager and Verghese (2008) proposed a graph similarity measure based on the structural similarity of local neighbourhoods to determine similarity scores for the nodes of two different graphs. This measure was applied to the graph matching problem. The issues facing representation designs by graphs and their assessment were discussed in Strug (2013). Kernel functions were used to compute the similarity of designs. The proposed approach was approach was applied to evaluate layout designs. A graph similarity approach for the detection of bearing faults was proposed by Sun et al. (2020). Hamedani and Kim (2017) proposed a similarity measure for graphs, JacSim, that had overcome the shortcomings of the SimRank measure. The measure has been extended to weighted graphs. Contextual similarity between pairs of nodes was introduced in Dutta et al. (2018). Subsequently, graph matching was formulated as a node and edge selection problem. The Tanimoto index measuring the topological similarity of graphs was studied in Dehmer and Varmuza (2015). The properties of the index applied to chemical alkane trees were studied. Bopche and Mehtre (2017) proposed graph distance metrics based on the maximum common subgraph and graph edit distance (GED) for assessment of security risk of networks. Computational results for 11 different metrics tested on a set of three different network models were provided. Theoretical considerations of graph similarity measures based on three graph matrices, the adjacency matrix, the Laplacian matrix, and the Markov matrix were reported in Avrachenkov, Chebotarev, and Rubanov (2019). Additional graph similarity and distance metrics were analysed in Chartrand, Kubicki, and Schultz (1998). Sabarish, Karthi, and Kumar (2020) developed a graph-based model for trajectory graphs of moving objects such as vehicles, humans, animals, or phenomena. The similarity between the graphs was computed using the edge and vertex-based metrics. The problem of searching graphs with noisy and incomplete data was considered in Zheng et al. (2015). Computational experience has confirmed the performance advantage of the approach proposed in the paper over the existing graph similarity approaches. Kwon et al. (2006) proposed a similarity metric to address interoperability issue in semantic web ontologies represented with graphs. The metric was applied to compute similarity across multiple web ontologies. Digital models of universal manufacturing will take different forms, including process models enriched with additional information imposed by application-specific model requirements. The additional information could include resources such as machine tools, their process characteristics and time-based availability, control software, and edge solution. The widely discussed in the literature digital twin is an instance of digital model.