Classifying and extracting data from Facebook posts for online persona identification

College

College of Computer Studies

Department/Unit

Software Technology

Document Type

Conference Proceeding

Source Title

Proceedings of the 32nd Pacific Asia Conference on Language, Information and Computation, PACLIC 2018

First Page

63

Last Page

72

Publication Date

1-1-2018

Abstract

Large amount of user-generated data areposted online in social media platforms, including user preferences, dining and leisureactivities, events, news and personal blogs.This resulted in varying efforts to process social media data using NLP and ML algorithmsfor topic classification, sentiment analysis anddetection, and events classification. Such information are problematic to process, as theytend to be short, informal, inconsistent, andare highly contextualized. A series of tasks isinvolved from collecting, pre-processing, classification and extraction before social mediadata can be used. In this study, we built amulti-class classifier model to process Facebook posts in order to identify a user's onlinepersona based on his/her preferences. Information extraction is then applied to find relevant data from the classified posts that can beused to generate a description of the user's online persona. The classifier currently achievesan accuracy of 76.02% and an F1 score of73.10% using 10-fold cross validation from adataset containing 16,682 posts. © 2018 by the authors.

html

Disciplines

Computer Sciences

Keywords

Data mining; Facebook (Electronic resource)

Upload File

wf_no

This document is currently not available here.

Share

COinS