A WordBased Intelligent Automated English Essay Scoring System: LSTM and 2Vec Deep Learning Models

The growing need for effective and reliable assessment of student writing has made automated essay scoring a significant subject of study in the field of educational technology. In addition to taking a lot of time and effort from teachers, traditional manual essay assessment may result in subjective grading bias. This study suggests an online automated English essay scoring system that uses machine learning approaches to address these issues. A number of linguistic and structural characteristics, including word count, sentence count, average word length, spelling mistakes, and part-of-speech data like nouns, verbs, adjectives, and adverbs, are extracted by the system in order to analyse essays. Additionally, Word2Vec embeddings, which aid in representing words in a meaningful vector space, are used to collect semantic information from the essay content. After that, a Long Short-Term Memory (LSTM) deep learning model is used to identify patterns in the textual data and forecast each essay's score. The Django framework is used to construct the suggested system as a web-based application that enables users to submit essays and promptly receive automatic scoring results. The algorithm can accurately predict essay results while preserving consistency and minimising manual grading labour, according to experimental evaluation. For automated assessment in contemporary educational settings, the system offers a scalable and effective alternative.