Geetharani, S and Vadivambigai, S (2023) A MULTIOBJECTIVE DEEP BELIEF NETWORK FOR EVENT PARTICIPANT PREDICTION IN ONLINE SOCIAL NETWORK SERVICE APPLICATIONS. Journal of Data Acquisition and Processing, 38 (3). pp. 2580-2593. ISSN 1004-9037

[thumbnail of Geetharani - CT.pdf] Text
Geetharani - CT.pdf - Published Version
Download (950kB)
Official URL: https://sjcjycl.cn/

Abstract

Event based Social Networks provides convenient platform for knowledge
enhancement to the participants from research communities and industrial communities in the
social network but nowadays numerous event has been organized by the organizer which leads
to the data overloading problem and it becomes complex to identify the suitable event and high
influencing event by participant . Most previous works in participant prediction using machine
learning and deep learning model focus on intrinsic and extrinsic properties of the user on their
behavior and preference analysis in the social context. However multiple social events are
hosted same time which it leads to high competition to obtain the influencing user to maximize
the number of participants. In this paper, multiobjective deep belief network for event
participant prediction is proposed to exploit the high influencing user to the various event.
Typical task is to identify the user features and event features on its contextual information.
Latent Dirichlet Allocation has been employed to extract the latent contextual information on
the different perspective to increase the high relevancy rate. Extracted latent contextual
information is projected to deep belief network to compute the participant prediction to event
classes on processing the latent information in hidden layer and visible layer. Each visible layer
enabled with representation learning of features. . Further influence weight has to be computed
on both long term interest representation model and short term interest representation model to
jointly represent user impact on the event. Interest model uses the multifaceted information
ranking based on knowledge level, hierarchy level and participation level on the relevant events
in the activation layer of deep belief function. Finally decision of the profile recommendation
to the events is integrated on basis of influence weight to the correlation of similar preferences
of the groups to the event in the visible layer. Evaluation of the proposed model through various
case studies has been implemented and validated across various measures such as accuracy on
precision, Recall and f measure along scalability and Execution time.

Item Type: Article
Uncontrolled Keywords: Event Based Social Network, Event Participant Prediction ,Online Events, Latent Dirichlet Allocation , Deep Belief Network
Divisions: PSG College of Arts and Science > Department of Computer Science
Depositing User: Mr Team Mosys
Date Deposited: 09 Aug 2023 09:41
Last Modified: 09 Aug 2023 09:41
URI: http://ir.psgcas.ac.in/id/eprint/2045

Actions (login required)

View Item
View Item
Altmetric