Discovering trends in twitter data

• Main Authors: Chua, Grace Joy, Naval, Nancy Reena, Rapes, Matthew Michael
• Format: text
• Language: English
• Published: Animo Repository 2014
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/10941
• Institution: De La Salle University

Twitter is one of the social media streams that has contributed to the alteration of the way people communicate and interact with one another. The one concept that makes Twitter what it is, is the concept of trends. Trends are topics that are immediately popular at a certain period of time. End-users evidently applied the use of Twitter trends to various fields such as business, marketing, politics, news reporting, weather, and the like. This paper discusses the implementation of Woodpecker, a trend detection that harnesses both a language modeler and topic modeler to detect trends from Twitter data. Furthermore, Woodpecker enables its users to dig deeper into trends by drilling down the trends that it is able to detect. These outputs are then displayed in various forms of visualization.