• Publications
  • Influence
Indosum: A New Benchmark Dataset for Indonesian Text Summarization
TLDR
We present INDOSUM, a new benchmark dataset for Indonesian text summarization. Expand
  • 12
  • 4
  • PDF
Extracting the Main Content from HTML Documents
A modern web document typically consists of many kinds of information. Besides the main content which conveys the primary information, a web document also contains noisy contents such asExpand
  • 5
  • 1
  • PDF
University of Indonesia at TREC 2011 Microblog Task
TLDR
We investigate various mechanisms to reformulate our query in order to retrieve more relevant documents. Expand
  • 8
  • PDF
Exploring Named Entity Recognition As an Auxiliary Task for Slot Filling in Conversational Language Understanding
TLDR
We propose a joint model of slot filling and Named Entity Recognition (NER) in a multi-task learning (MTL) setup which achieves competitive performance compared with state-of-the-art. Expand
  • 6
  • PDF
Unsupervised aspect-based sentiment analysis on Indonesian restaurant reviews
TLDR
We address the problem of aspect-based sentiment analysis (ABSA) from Indonesian restaurant reviews due to its usefulness for customer satisfaction applications. Expand
  • 7
Empirical Evaluation of Character-Based Model on Neural Named-Entity Recognition in Indonesian Conversational Texts
TLDR
We report an empirical evaluation of neural sequence labeling models with character embedding to tackle NER task in Indonesian conversational texts. Expand
  • 5
  • PDF
Leveraging Non-Conversational Tasks for Low Resource Slot Filling: Does it help?
TLDR
We show that using auxiliary non-conversational tasks as auxiliary tasks in a multi-task learning setup consistently improves low resource slot filling performance. Expand
  • 5
  • PDF
How secure your applications are?: Analysis of web developers awareness to application security
TLDR
We develop a framework to measure the security awareness level of web developers in Indonesia. Expand
  • 3
Classification of pornographic content on Twitter using support vector machine and Naive Bayes
TLDR
The Internet has many benefits, some of them are to gain knowledge and gain the latest information. Expand
  • 3
Cross Sentence Inference for Process Knowledge
TLDR
A cross-sentence, collective role-labeling and alignment method for harvesting process knowledge from multiple sentence-level descriptions. Expand
  • 5
  • PDF