VIETNAMESE ABSTRACTIVE TEXT SUMMARIZATION USING TRANSFORMER

Abstract

In recent years, there has been an explosion of the amount of text data from variety sources. This volume of text is an invaluable source of information and knowledge which needs to be effectively summarized to be useful. Producing a text summarization technique which helps reducing the amount of text data is essential. Thus, as the evolution of machine learning, a lot of Natural Languages Processing (NLP) model were introduced. This thesis will apply Transformer – one of state at the art techniques, and a pre-trained model for implementing a new machine learning model which will receive long sentences as input and take it into shorter ones with the same meaning. The model will be trained and tested on both Vietnamese and English online articles. For evaluation, there have been many methods to evaluate the quality of summaries. However, in NLP field, ROUGE method is the most commonly used. Hence, ROUGE method will be used in this thesis to evaluate the summaries. The results of the model after evaluated is considered to be good and acceptable compared to other text summarization models. The result of the thesis introduces an abstractive Vietnamese summarization model. The model enables users to generate summary that remains the original meaning of the source text in sentence level.