Indosum: A New Benchmark Dataset for Indonesian Text Summarization
- Kemal Kurniawan, Samuel Louvan
- Computer ScienceInternational Conference on Asian Language…
- 12 October 2018
This paper presents INDOSUM, a new benchmark dataset for Indonesian text summarization, which consists of news articles and manually constructed summaries and is almost 200x larger than the previous Indonesian summarization dataset of the same domain.
Recent Neural Methods on Slot Filling and Intent Classification for Task-Oriented Dialogue Systems: A Survey
- Samuel Louvan, B. Magnini
- Computer ScienceInternational Conference on Computational…
- 1 November 2020
This work surveys how neural based models have rapidly evolved to address natural language understanding in dialogue systems, and introduces three neural architectures: independent models, which model SF and IC separately, and joint models,Which exploit the mutual benefit of the two tasks simultaneously.
Simple is Better! Lightweight Data Augmentation for Low Resource Slot Filling and Intent Classification
- Samuel Louvan, B. Magnini
- Computer SciencePacific Asia Conference on Language, Information…
- 8 September 2020
It is shown that lightweight augmentation, a set of augmentation methods involving word span and sentence level operations, alleviates data scarcity problems and achieves competitive performance with respect to more complex, state-of-the-art, augmentation approaches.
Extracting the Main Content from HTML Documents
- Samuel Louvan
- Computer Science
- 2009
This paper presents an approach for extracting main content from web documents which combines classification tasks and heuristic approaches and it is shown that this approach can be successful in relation to resolving noisy contents.
Multi-Task Active Learning for Neural Semantic Role Labeling on Low Resource Conversational Corpus
- Fariz Ikhwantri, Samuel Louvan, Rahmad Mahendra
- Computer ScienceDeepLo@ACL
- 5 June 2018
A Multi-Task Active Learning framework for Semantic Role Labeling with Entity Recognition (ER) as the auxiliary task to alleviate the need for extensive data and use additional information from ER to help SRL is proposed.
Leveraging Non-Conversational Tasks for Low Resource Slot Filling: Does it help?
- Samuel Louvan, B. Magnini
- Computer ScienceSIGDIAL Conferences
- 1 September 2019
It is shown that using auxiliary non-conversational tasks in a multi-task learning setup consistently improves low resource slot filling performance.
IndoNLI: A Natural Language Inference Dataset for Indonesian
- Rahmad Mahendra, Alham Fikri Aji, Samuel Louvan, Fahrurrozi Rahman, Clara Vania
- Computer ScienceConference on Empirical Methods in Natural…
- 27 October 2021
IndoNLI is designed to provide a challenging test-bed for Indonesian NLI by explicitly incorporating various linguistic phenomena such as numerical reasoning, structural changes, idioms, or temporal and spatial reasoning.
Classification of pornographic content on Twitter using support vector machine and Naive Bayes
- N. Izzah, I. Budi, Samuel Louvan
- Computer ScienceInternational Conference on Computing: Theory and…
- 27 June 2018
This study aims to determine the best model to detect the pornographic content on social media based on unigram and bigram features, classification algorithm, k-fold cross validation, and Support Vector Machine and Naive Bayes.
Unsupervised aspect-based sentiment analysis on Indonesian restaurant reviews
- Dhanang Hadhi Sasmita, A. Wicaksono, Samuel Louvan, M. Adriani
- Computer ScienceInternational Conference on Asian Language…
- 1 December 2017
This work addresses the problem of aspect-based sentiment analysis from Indonesian restaurant reviews by proposing an unsupervised approach which does not rely too much on hard-core natural language processing tools since Indonesian language is still under-resourced in terms of language technology.
Exploring Named Entity Recognition As an Auxiliary Task for Slot Filling in Conversational Language Understanding
- Samuel Louvan, B. Magnini
- Computer ScienceSCAI@EMNLP
- 1 October 2018
This work proposes a joint model of slot filling and Named Entity Recognition (NER) in a multi-task learning (MTL) setup and shows that using NER as an auxiliary task improves slot filling performance and achieve competitive performance compared with state-of-the-art.
...
...