Social media platforms generate vast amounts of data, necessitating efficient content classification and community detection methods. This study addresses this challenge through the utilization of VGGNet analytics, a powerful deep learning architecture. We employed a two-step approach, beginning with VGGNet-based content classification to categorize social media posts. Subsequently, a community detection algorithm was applied to identify distinct user groups based on their interactions and content preferences. This research contributes an novel framework that seamlessly integrates VGGNet for content analysis and community detection, enhancing the understanding of user behavior in social media platforms. The proposed method aims to provide more accurate and insightful results compared to traditional approaches. Our experiments on diverse social media datasets demonstrate the effectiveness of the VGGNet-based approach. The content classification accurately assigns posts to relevant categories, while the community detection algorithm identifies cohesive user groups. The results highlight the potential for improved content recommendation systems and targeted marketing strategies.
Vishal Gangadhar Puranik1, V. Saravanan2, P. Selvaraju3, R. Janaki4
MIT Academy of Engineering, India1, Johns Hopkins University, United States of America2, Excel Engineering College, India3, Roever Engineering College, India4
Community Detection, Content Classification, Social Media, VGGNet Analytics, Deep Learning
| January | February | March | April | May | June | July | August | September | October | November | December |
| 0 | 0 | 12 | 4 | 1 | 2 | 2 | 1 | 0 | 4 | 1 | 0 |
| Published By :
ICTACT
Published In :
ICTACT Journal on Image and Video Processing
( Volume: 14 , Issue: 3 , Pages: 3181 - 3186 )
Date of Publication :
February 2024
Page Views :
455
Full Text Views :
27
|