A text summary extracts serves as a condensed representation of a written input source where important and salient information is kept. However, the condensed representation itself suffer in lack of semantic and coherence if the summary was produced in verbatim using the input itself. Sentence Compression is a technique where unimportant details from a sentence are eliminated by preserving the sentence’s grammar pattern. In this study, we conducted an analysis on our developed Malay Text Corpus to discover the rules and pattern on how human summarizer compresses and eliminates unimportant constituent to construct a summary. A Pattern-Growth based model named Frequent Eliminated Pattern (FASPe) is introduced to represent the text using a set of sequence adjacent words that is frequently being eliminated across the document collection. From the rules obtained, some heuristic knowledge in Sentence Compression is presented with confidence value as high as 85% - that can be used for further reference in the area of Text Summarization for Malay language.

No abstract

No abstract

No abstract

Statistical Machine Translation (SMT) and Neural Machine Translation (NMT) are the state-of-the-art approaches in machine translation (MT). The translation produced by an SMT is based on the statistical analysis of text corpora, while NMT uses the deep neural network to model and to generate a translation. SMT and NMT have their strength and weaknesses. SMT may produce a better translation with a small parallel text corpus compared to NMT. Nevertheless, when the amount of parallel text available is large, the quality of the translation produced by NMT is often higher than SMT. Besides that, study also shown that the translation produced by SMT is better than NMT in cases where there is a domain mismatch between training and testing. SMT also has an advantage in long sentences. In addition, when a translation produced by an NMT is wrong, it is very difficult to find the error. In this paper, we investigate a hybrid approach that combines SMT and NMT to perform English to Malay translation. The motivation for using a hybrid machine translation is to combine the strength of both approaches to produce a more accurate translation. Our approach uses the multi-source encoder-decoder long short-term memory (LSTM) architecture. The architecture uses two encoders, one to embed the sentence to be translated, and another encoder to embed the initial translation produced by SMT. The translation from the SMT can be viewed as a "suggestion translation" to the neural MT. Our experiments show that the hybrid MT increases the BLEU scores of our best baseline machine translation in the computer science domain and news domain from 21.21 and 48.35 to 35.97 and 61.81 respectively.