Community Detection by Fuzzy Relations

Luo, Wenjian; Yan, Zhenglong; Bu, Chenyang; Zhang, Daofu

doi:10.1109/tetc.2017.2751101

Cited by 35 publications

(15 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover SOMs have been employed for a hierarchical clustering scheme for discovering latent gene expression patterns [25] and gene regulatory networks [26]. Given that the trained fuzzy cognitive maps can be represented as a fuzzy graph, clustering can be performed by fuzzy community discovery algorithms [27][28][29]. In Reference [30] fuzzy graphs have been used in a technique for estimating the number of clusters and their respective centroids.…”

Section: Previous Workmentioning

confidence: 99%

See 1 more Smart Citation

Extending Fuzzy Cognitive Maps with Tensor-Based Distance Metrics

et al. 2020

View full text Add to dashboard Cite

Cognitive maps are high level representations of the key topological attributes of real or abstract spatial environments progressively built by a sequence of noisy observations. Currently such maps play a crucial role in cognitive sciences as it is believed this is how clusters of dedicated neurons at hippocampus construct internal representations. The latter include physical space and, perhaps more interestingly, abstract fields comprising of interconnected notions such as natural languages. In deep learning cognitive graphs are effective tools for simultaneous dimensionality reduction and visualization with applications among others to edge prediction, ontology alignment, and transfer learning. Fuzzy cognitive graphs have been proposed for representing maps with incomplete knowledge or errors caused by noisy or insufficient observations. The primary contribution of this article is the construction of cognitive map for the sixteen Myers-Briggs personality types with a tensor distance metric. The latter combines two categories of natural language attributes extracted from the namesake Kaggle dataset. To the best of our knowledge linguistic attributes are separated in categories. Moreover, a fuzzy variant of this map is also proposed where a certain personality may be assigned to up to two types with equal probability. The two maps were evaluated based on their topological properties, on their clustering quality, and on how well they fared against the dataset ground truth. The results indicate a superior performance of both maps with the fuzzy variant being better. Based on the findings recommendations are given for engineers and practitioners.

show abstract

Section: Previous Workmentioning

confidence: 99%

“…Perhaps the most important figure of merit of this category is the topological error. The latter is defined in Equation (29). Figure 3 shows the average topological error as a percentage of the total number of data points after R 0 runs for each distance metric using the cosine rate.…”

Section: Topological Errormentioning

confidence: 99%

Extending Fuzzy Cognitive Maps with Tensor-Based Distance Metrics

et al. 2020

View full text Add to dashboard Cite

show abstract

“…The CDFR and the CDFR-U algorithm based on the fuzzy relations were proposed by Luo et al [30], which the NGC node (Nearest node with Greater Centrality) plays an important role that determines the community label of each node. The centrality of each node is calculated according to the degree of each node and the degree of the direct neighbors, and then a special algorithm is used to find the NGC nodes and calculate the fuzzy relation between each node and its NGC node.…”

Section: Motivation a Related Workmentioning

confidence: 99%

“…To compare the performance of these algorithms, experiments are carried out by using the aforementioned generated networks, and the comparison of the experimental results given by each algorithm are shown in Table 1. Meanwhile, we also compare the NMI index given by the global community detection algorithms that include Louvain [15], SA [53], EO [54], LPA [27], EM [55], CDFR-U [30] (it is necessary to set a delta threshold to enlarge the gap between the central node of the community and the non-central node), and LIDGC. We can see that our algorithm guarantees good performance in most cases, especially in the first two networks, the performance of the algorithm reaches the highest value.…”

Section: A Random Networkmentioning

confidence: 99%

Community Detection Based on Local Information and Dynamic Expansion

et al. 2019

View full text Add to dashboard Cite

Mining the community structure in the real-world networks has been a hot topic in the field of complex networks, and has emerged as a prominent research area and continues to grow with the introduction of requirements for personalized recommendation. However, most of the existing community detection algorithms are based on global information, fewer works are devoted to detecting the communities hidden in the network by using local information. To this end, in this paper, we propose two improved signed modularity functions to evaluate the community properties in complex network, and then we apply these indicators to identify the community structure by using the local information since it is difficult to obtain the global information in practice. During the dynamic expansion, each local community will absorb the neighboring node with the highest positive energy; in addition, a new local community is generated when all local communities cannot contain the neighboring node. Finally, the algorithm has been applied to unsigned networks and signed networks, respectively. The experimental results show that the division results given by our proposed algorithm are in line with the actual ones in artificial and real-life networks.

show abstract

“…There are also some researchers apply DP algorithm to image processing [21], community detection [22]- [25], extracting multi-document abstracts [26] and noise removal [27].…”

Section: Algorithm 1 Dpmentioning

confidence: 99%