loder

Prosa.ai’s NMT Research Awarded as Best Paper at The ICIC APTIKOM 2019

Semarang (24/10), the research paper titled, “Combining Word and Character Vector Representation on Neural Machine Translation”, written by Khaidzir Muhammad Shahih (Prosa.ai Text Engineer) and Ayu Purwarinati (Prosa.ai Chief Scientist of Text and Chairman of AI Center ITB), and presented by Khaidzir, was awarded as Best Paper by the International Conference on Informatics and Computing Asosiasi Pendidikan Tinggi Informatika dan Komputer 2019 (ICIC APTIKOM 2019). The conference was held at Universitas Dian Nuswantoro (UDINUS), Semarang.

The paper describes combinations of word vector representation and character vector representation in English-Indonesian neural machine translation (NMT). The results from the experiment showed that NMT models with a concatenation of word and character representation obtained BLEU score higher than baseline model (word only based model), ranging from 9.14 points to 11.65 points.

Khaidzir Prosa AI Best Paper ICIC APTIKOM 2019.jpg

The paper was awarded as Best Paper by ICIC APTIKOM 2019. ICIC APTIKOM is part of an annual event, Rapat Koordinasi Nasional APTIKOM (RAKORNAS APTIKOM), held by Asosiasi Pendidikan Tinggi Informatika dan Komputer (APTIKOM) collaborating with UDINUS. The conference aims to provide a platform for researchers and academics to exchange information, knowledge, skills, and experiences in the field of Computer Science, Information Science, and Computer Engineering. This year, the conference introduced the topic “Memperkuat Masyarakat Ekonomi Digital Melalui Inovasi Industri Kreatif Di Era Revolusi Industri 4.0”.

Moreover, the paper presented in the conference by Khaidzir focuses on Natural Language Processing (NLP) specifying in Bahasa Indonesia, one of Prosa.ai’s research and business focus along with Speech Processing for Bahasa Indonesia and Computer Vision.

None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None