Topic-Sensitive Influencer Mining in Interest-Based Social Media Networks via Hypergraph Learning

Fang, Quan; Sang, Jitao; Xu, Changsheng; Rui, Yong

doi:10.1109/tmm.2014.2298216

Cited by 97 publications

(47 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, we establish a hyperedge containing the users who commonly follow a target person. The weight of a user hyperedge is defined as the average similarity over all pairs of vertices linked with the hyperedge, which is in line with the intuition that a high weight should be assigned to a hyperedge if the users within it are close to each other [Fang et al 2014]. Similar to previous work [Cui et al 2014], for two users u ∈ I and v ∈ I, we estimate their similarity by measuring how many groups and persons are co-joined or co-followed by u and v:…”

Section: Bi-relational Hypergraph Representationmentioning

confidence: 94%

“…For the adjustment parameter η in Equation (8), we treated the grouping factor and the following factor equally and set η = 0.5. For the number of nearest neighbors k considered when constructing the POI hypergraph, empirical studies [Fang et al 2014; have showed that a relatively small number of k is sufficient to make the graph connected. We thus set k = 5 to make the POI hypergraph sparse for computational efficiency.…”

Section: Parameter Settingsmentioning

confidence: 99%

See 1 more Smart Citation

Augmented Collaborative Filtering for Sparseness Reduction in Personalized POI Recommendation

Cui

Shen

Nie

et al. 2017

ACM Trans. Intell. Syst. Technol.

View full text Add to dashboard Cite

As mobile device penetration increases, it has become pervasive for images to be associated with locations in the form of geotags. Geotags bridge the gap between the physical world and the cyberspace, giving rise to new opportunities to extract further insights into user preferences and behaviors. In this article, we aim to exploit geotagged photos from online photo-sharing sites for the purpose of personalized Point-of-Interest (POI) recommendation. Owing to the fact that most users have only very limited travel experiences, data sparseness poses a formidable challenge to personalized POI recommendation. To alleviate data sparseness, we propose to augment current collaborative filtering algorithms along from multiple perspectives. Specifically, hybrid preference cues comprising user-uploaded and user-favored photos are harvested to study users' tastes. Moreover, heterogeneous high-order relationship information is jointly captured from user social networks and POI multimodal contents with hypergraph models. We also build upon the matrix factorization algorithm to integrate the disparate sources of preference and relationship information, and apply our approach to directly optimize user preference rankings. Extensive experiments on a large and publicly accessible dataset well verified the potential of our approach for addressing data sparseness and offering quality recommendations to users, especially for those who have only limited travel experiences.

show abstract

Section: Bi-relational Hypergraph Representationmentioning

confidence: 94%

Section: Parameter Settingsmentioning

confidence: 99%

Augmented Collaborative Filtering for Sparseness Reduction in Personalized POI Recommendation

Cui

Shen

Nie

et al. 2017

ACM Trans. Intell. Syst. Technol.

View full text Add to dashboard Cite

show abstract

“…In any case, refreshing and appropriation of the street delineate is an unpredictable and high cost handle. Inquire about [10] proposed outline work for Topic-Sensitive Influencer Mining (TSIM).In which, impact estimation of clients and pictures is resolved with hypergraph learning approach. It uses visual-textual content relations to develop homogeneous hyper edges for the point appropriation learning and social connection relations to develop heterogeneous hyper edges for impact positioning in the system.…”

Section: Literature Surveymentioning

confidence: 99%

A headway methodology for online web-based social network against cybercriminal mining

Leela¹,

Jayasree²,

Raj³

et al. 2018

IJET

View full text Add to dashboard Cite

Data mining is the way toward gathering information from various setting and condenses them into helpful data. Information mining can be utilized to decide the connection between inside elements and outside elements .It permits the clients to investigate, order and decides the connections deduced in them. Content mining as a rule alluded to as content information mining can be utilized be utilized to concentrate data from content. Content mining can be utilized as a part of data recovery, design acknowledgment and information mining systems. The presentation of online networking and interpersonal organizations has changed the open doors accessible for us as well as we should be careful about the dangers. Late explores demonstrate that the quantity of wrongdoings are expanding through online web-based social networking and they may bring about enormous misfortune to associations. Existing digital advancements are not viable to secure organizations. Existing mining techniques focus on dictionaries in which they can distinguish just a predetermined number of relations. Here a hereditary calculation approach is presented in which inert ideas can be removed. Hereditary Calculation is a straight pursuit which requires just little data from vast hunt zone.. At that point these ideas are subjected to separate the semantics which construes the comparing connections. Hereditary calculation gives a superior arrangement in which exactness and time effectiveness can be moved forward. The principle commitment of the paper demonstrates that they distinguish the relating cybercriminal systems.

show abstract

“…Domestic and foreign researchers design qualitative or quantitative measurements before studying on user influence [1][2][3][4][5][6][7], and information propagation [8][9][10][11][12]. Me young et al [13] present an in-depth comparison of three measures of influence: in degree, retweets, and mention.…”

Section: Related Workmentioning

confidence: 99%

Measurement and Analysis of Burst Topic in Microblog

Dong¹,

Zou²,

Wang³

et al. 2015

IJDTA

View full text Add to dashboard Cite

show abstract

Topic-Sensitive Influencer Mining in Interest-Based Social Media Networks via Hypergraph Learning

Cited by 97 publications

References 49 publications

Augmented Collaborative Filtering for Sparseness Reduction in Personalized POI Recommendation

Augmented Collaborative Filtering for Sparseness Reduction in Personalized POI Recommendation

A headway methodology for online web-based social network against cybercriminal mining

Measurement and Analysis of Burst Topic in Microblog

Contact Info

Product

Resources

About