Alih Bentuk Kalimat Non-Formal Menjadi Kalimat Formal Menggunakan Pendekatan Machine Translation
Abstract
This research aims to apply the Long Short-Term Memory algorithm to the conversion of informal sentences into formal sentences using Indonesian sentences. Development of LSTM model application software to convert informal sentences to formal sentences in this study uses the incremental method. The BLEU evaluation determines whether or not the predicted sentence is accepted by measuring the closeness of the context of the formal sentence from the dataset. This research produces a website-based application. The conclusion obtained is that the training model that is built produces accuracy and loss values with a value of 0.7011 and 3.7401. Based on these results, an assessment of the training model is not good enough. This is due to several factors that cause the model training to experience overfitting.