I Made Suwija Putra’s research while affiliated with Sepuluh Nopember Institute of Technology and other places

Recognizing textual entailment (RTE) is an essential task in natural language processing (NLP). It is the task of determining the inference relationship between text fragments (premise and hypothesis), of which the inference relationship is either entailment (true), contradiction (false), or neutral (undetermined). The most popular approach for RTE is neural networks, which has resulted in the best RTE models. Neural network approaches, in particular deep learning, are data-driven and, consequently, the quantity and quality of the data significantly influences the performance of these approaches. Therefore, we introduce SNLI Indo, a large-scale RTE dataset in the Indonesian language, which was derived from the Stanford Natural Language Inference (SNLI) corpus by translating the original sentence pairs. SNLI is a large-scale dataset that contains premise-hypothesis pairs that were generated using a crowdsourcing framework. The SNLI dataset is comprised of a total of 569,027 sentence pairs with the distribution of sentence pairs as follows: 549,365 pairs for training, 9,840 pairs for model validation, and 9,822 pairs for testing. We translated the original sentence pairs of the SNLI dataset from English to Indonesian using the Google Cloud Translation API. The existence of SNLI Indo addresses the resource gap in the field of NLP for the Indonesian language. Even though large datasets are available in other languages, in particular English, the SNLI Indo dataset enables a more optimal development of deep learning models for RTE in the Indonesian language.

E-learning is an online learning system that applies information technology in the teaching process. E-learning used to facilitate information delivery, learning materials and online test or assignments. The online test in evaluating students’ abilities can be multiple choice or essay. Online test with essay answers is considered the most appropriate method for assessing the results of complex learning activities. However, there are some challenges in evaluating students essay answers. One of the challenges is how to make sure the answers given by students are not the same as other students answers or 'copy-paste'. This study makes a similarity detection system (Similarity Checking) for students' essay answers that are automatically embedded in the e-learning system to prevent plagiarism between students. In this paper, we use Artificial Neural Network (ANN), Latent Semantic Index (LSI), and Jaccard methods to calculate the percentage of similarity between students’ essays. The essay text is converted into array that represents the frequency of words that have been preprocessed data. In this study, we evaluate the result with mean absolute percentage error (MAPE) approach, where the Jaccard method is the actual value. The experimental results show that the ANN method in detecting text similarity has closer performance to the Jaccard method than the LSI method and this shows that the ANN method has the potential to be developed in further research.