Implementation of Neural Networks in the Detection of Equality of Answer Student Essays

E-learning is an online learning system that began to be used by utilizing information technology in the teaching and learning process. The function of e-learning is to facilitate the delivery of information, sharing of learning, and work on questions or assignments. The application of giving questions or tasks in the evaluation can be a multiple-choice method and essay answers. Tasks with essay characteristics are considered the most appropriate method used in assessing the results of complex learning activities. But there are some challenges in evaluating student essay answers. One challenge is how to be sure the answers given by students are not the same answers or "copy-paste" with the answers of other students. This research makes a similarity detection system (Similarity Checking) answers to student essays automatically so that it helps prevent plagiarism in evaluating learning outcomes. The automated text similarity detection system will be embedded in the e-learning system with a text similarity detection technique using one of the Neural Network methods. This method is highly developed in the world of machine learning with various reference features and is widely used in text information processing both information matching and grouping information. This research will create and design a similarity detection system for student essay answers and measure the performance of this method. The measurement results of these methods become a benchmark for the success of the system in detecting the similarity of students' answers in e-learning.