Semantic role labeling in conversational chat using deep bi-directional long short-term memory networks with attention mechanism

Valdi Rachman, Rahmad Mahendra, Alfan Farizki Wicaksono, Ahmad Rizqi Meydiarso, Fariz Ikhwantri

Research output: Contribution to conferencePaperpeer-review

Abstract

Semantic Role Labeling (SRL) has been extensively studied, mostly for understandingEnglish formal language. However, only a fewreports exist for informal conversational text,especially for language being used in the chatbot system. The challenges of informal textinglanguage include a wide variety of slangs andabbreviations, short sentences, as well as disorganized grammars. In this work, we proposean attention mechanism on top of Long ShortTerm Memory Networks architecture for solving SRL task on informal conversations. Thetask is evaluated on informal language usedon an Indonesian chatting platform. Our proposed model achieves F1 score of 82.68%.

Original languageEnglish
Pages558-566
Number of pages9
Publication statusPublished - 2018
Event32nd Pacific Asia Conference on Language, Information and Computation, PACLIC 2018 - Hong Kong, Hong Kong
Duration: 1 Dec 20183 Dec 2018

Conference

Conference32nd Pacific Asia Conference on Language, Information and Computation, PACLIC 2018
Country/TerritoryHong Kong
CityHong Kong
Period1/12/183/12/18

Fingerprint

Dive into the research topics of 'Semantic role labeling in conversational chat using deep bi-directional long short-term memory networks with attention mechanism'. Together they form a unique fingerprint.

Cite this