IncOrder: Incremental density-based community detection in dynamic networks

Huang, Jianbin; Zhang, Xin; Liu, Jiao; Wang, Dong; Liu, Huailiang; Zou, Jianhua; Song, Qinbao

doi:10.1016/j.knosys.2014.07.015

Cited by 28 publications

(11 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The clustering results are summarized as a collection of multiple articles cantered on k cluster centre articles. for each element in last cluster 5) new documents ← {} 6) means ← near mean (docfeaturetable, element)// Call near mean algorithm, the most close to the sample mean vector 7) for i←0 to length(initcluster) 8) if(distance(element, i) ≥ threshold// Screening and clustering centre distance of similarity threshold or essay collections 9) add each element to new documents 10) calculate mean vector of new documents// Calculate the mean vector collection selected articles 11) insert each element of mean vector to new centre list in order// The mean vector of each element as a new clustering centre 12) last cluster ← cluster result// To regenerate the clustering results 13) return cluster result （5）Filter and recommend all types of cluster words According to the clustering results, the collected articles have been clustered into k clusters on the basis of calculated text similarity. In order to visualize the subject of each type of article, it is necessary to filter out several characteristic words with the largest eigenvalue in all the articles in each kind of article.…”

Section: Wechat Articles Recommendation Methods Based On the Icipmentioning

confidence: 99%

See 1 more Smart Citation

Social Network Recommendation Algorithm Based on ICIP

Ren³

et al. 2018

Proceedings of Information Science and Cloud Computing — PoS(ISCC 2017)

View full text Add to dashboard Cite

The recent years has witnessed rapid development of social network platforms. To enable timely and effective selection of information that is valuable and raises interests, a variety of personalized recommendation algorithms have been put into practice. The improved clustering based on the Isolation Point (ICIP) algorithm is presented based on clustering. In order to overcome the shortcomings of the traditional clustering, the process of Isolation Point has been included in the ICIP algorithm. In this paper, the ICIP algorithm is used for topic extraction of WeChat articles. According to the characteristics of the social network platform, the data noise reduction and modeling is adopted first and then text classification is achieved based on the similarity. The ICIP is applied to remove isolated points, improve clustering accuracy and reduce noise. Compared with other clustering algorithms, the ICIP algorithm has higher accuracy and efficiency.

show abstract

Section: Wechat Articles Recommendation Methods Based On the Icipmentioning

confidence: 99%

“…However, these algorithms cannot detect changes in the network structure. Sun H, et al [5] proposed an algorithm discovering dynamic network community , to analyze the change area of network topology and update dynamically by using the incremental approach. Enhong Xie divided the mixed recommendation into two ways.…”

Section: Introductionmentioning

confidence: 99%

Social Network Recommendation Algorithm Based on ICIP

Ren³

et al. 2018

Proceedings of Information Science and Cloud Computing — PoS(ISCC 2017)

View full text Add to dashboard Cite

show abstract

“…Based on the overlapping behavior of the community structures, many works are developed to extract the hidden communities via a variety of different approaches. The examples in this category include detection of connected overlapping communities based on searching the adjacent cliques [18], edge partitioning techniques [19], label propagation algorithms [20], overlapping community detection in two-mode networks [21,22,23], topic oriented community detection via a link analysis approach [24], node location analysis to detect overlapping communities [25], overlapping local neighborhood ratio [26,27], maximal subgraphs [28,29], detecting core nodes among the communities [30], and model based clustering ideas [31,32]. Due to the unsupervised essence of the problem, heuristic and meta-heuristic approaches are investigated to uncover the community patterns through various heuristic fitness functions and nature-inspired algorithms [33,34].…”

Section: Introductionmentioning

confidence: 99%

IEDC: An integrated approach for overlapping and non-overlapping community detection

Hajiabadi

Zare

Bobarshad

2017

Knowledge-Based Systems

View full text Add to dashboard Cite

Community detection is a task of fundamental importance in social network analysis that can be used in a variety of knowledge-based domains. While there exist many works on community detection based on connectivity structures, they suffer from either considering the overlapping or non-overlapping communities. In this work, we propose a novel approach for general community detection through an integrated framework to extract the overlapping and non-overlapping community structures without assuming prior structural connectivity on networks. Our general framework is based on a primary node based criterion which consists of the internal association degree along with the external association degree. The evaluation of the proposed method is investigated through the extensive simulation experiments and several benchmark real network datasets. The experimental results show that the proposed method outperforms the earlier state-of-the-art algorithms based on the wellknown evaluation criteria.

show abstract

“…However, in many practical applications, the datasets are dynamically modified which means some previously learned patterns have to be updated accordingly [3,4]. Although these approaches have been successfully applied, there are some situations in which a richer model is needed for representing a cluster [5,6].…”

Section: Introductionmentioning

confidence: 99%