Leveraging Social Media Analytics for Student Project Allocation Using Deep Learning

Document Type

Conference Proceeding

Publication Title

Procedia Computer Science

Abstract

The integration of data-driven initiatives and technology has received substantial interest in the educational domain. Capstone projects significantly influence the career trajectories of students. Rather than merely allocating projects, the utilization of social media data to determine the domain interests of students in student project allocation has gained significant attention. This study explores the potential of social media analytics to revolutionize student project allocation using LinkedIn profiles. The proposed CNN model with Word2Vec embeddings demonstrated a superior performance of 96% in classifying the students based on LinkedIn profiles. The effectiveness of CNN utilizing the power of Word2Vec word embedding is employed to to capture syntactic and semantic relationships within text data. This high efficiency underscores the potential of Convolutional Neural Networks(CNN) as a text classification tool for facilitating smooth collaboration in project-based learning environments for quality education.

First Page

386

Last Page

397

DOI

10.1016/j.procs.2025.04.275

Publication Date

1-1-2025

This document is currently not available here.

Share

COinS