Summarization algorithms performance for topic clustered twitter microblogs

Author

JOHN SIXTO G. SANTOS, Ateneo de Manila University

Date of Award

2018

Document Type

Thesis

Degree Name

Master of Science in Computer Science

Department

Information Systems & Computer Science

First Advisor

Estuar, Ma. Regina Justina E., Ph.D.

Abstract

This paper discusses an approach that would allow for the condensation of a bodyof Twitter microblogs into a wieldy size by extracting the topics being discussed in acorpus of tweets using Latent Dirichlet Allocation (LDA). The approach presents theoutput into a human readable summary using the Phrase Reinforcement (PR)algorithm. The average F-measure score of this method exceeds those of othermethods when evaluated against human-made summaries. Results also suggest thatLDA together with PR is more robust against noisier datasets than the other testedmethods. This solution would help utilize Twitter into a tool not only for sharing ofexperiences but also a tool for gathering the state of the population. Decision makerscan use this solution to make informed action.

Comments

The C7.S258 2018

Recommended Citation

SANTOS, J. (2018). Summarization algorithms performance for topic clustered twitter microblogs. Ateneo de Manila University.
https://archium.ateneo.edu/theses-dissertations/58