Essay Rating System Using Machine Learning

Pathak, Pooja; Raghav, Shilpy; Jain, Shubhanshu; Jalal, Sunita

doi:10.1109/iscon52037.2021.9702504

Cited by 3 publications

(5 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This highlights the potential of NLP and machine learning for creating reliable and efficient scoring systems for writing tasks. The momentum continued with the work of [31], who developed an Essay Scoring System that primarily uses linear regression and the Naïve Bayes classifier algorithm for evaluation. In the next year, [15] investigated the use of BERT and RoBERTa models for AES, finding RoBERTa superior in terms of accuracy and efficiency.…”

Section: The Role Of Advanced Language Models In Aes Systemsmentioning

confidence: 99%

“…They devised an AES approach that combined rule-based grammar and consistency tests with semantic sentence similarity, yielding promising results in score prediction compared to human graders. In a different approach concerning grammar and spelling, [31] developed an Essay Scoring System that assesses essays based on these elements, primarily using linear regression and the Naïve Bayes classifier algorithm.…”

Section: Analytical Measures and Their Impact On Aes Systemsmentioning

confidence: 99%

“…Several studies have shown the advancements in natural language processing (NLP) and transformer-based neural language model have significantly influenced the development and effectiveness of automated essay scoring (AES) systems [6,9,13,15,18,[30][31][32][33][34][35][36][37][38].…”

Section: The Role Of Advanced Language Models In Aes Systemsmentioning

confidence: 99%

See 2 more Smart Citations

Evaluation of Transformer-Based Neural Language Models for Writing Feedback and Automated Essay Scoring

Abraha,

Nazir

2024

Preprint

View full text Add to dashboard Cite

Writing remains a challenging skill for many students due to inadequate feedback tools. There is a need to develop more effective tools for supporting students’ writing skill development. This study aims to evaluate the effectiveness of transformer-based neural language models for assessing and automatically scoring argumentative essays written by 8th-12th grade English Language Learners (ELLs). The students’ English essays were assessed and scored based on six criteria, including cohesion, syntax, vocabulary, phraseology, grammar, and conventions. The models were trained on real teacher feedback from 2700 scored essays. We also compared various transformer-based neural language models to find the most effective model. Several metrics were used for evaluation, with the root mean square error (RMSE) as the primary measure. The results show that a specific model, DeBERTa-v3-large, outperforms others in most categories. In conclusion, this study suggests that transformer-based neural language models, especially when using the DeBERTa-v3-large model, hold significant promise in improving automated essay scoring and feedback, potentially leading to enhanced writing skills among English language learners.

show abstract

Section: The Role Of Advanced Language Models In Aes Systemsmentioning

confidence: 99%

Section: Analytical Measures and Their Impact On Aes Systemsmentioning

confidence: 99%

See 1 more Smart Citation

Evaluation of Transformer-Based Neural Language Models for Writing Feedback and Automated Essay Scoring

Abraha,

Nazir

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Selama tahap penyaringan, ada langkah yang disebut penghapusan stopwords. Stopwords adalah kata-kata yang kurang memiliki makna penting atau memiliki bobot kurang signifikan, dan sering muncul dalam kumpulan kata-kata [9].…”

Section: ) Penyaringanunclassified

Penerapan Algoritma TF-IDF dan Cosine Similarity untuk Query Pencarian Pada Dataset Destinasi Wisata

Al Rasyid,

Ningsih

2024

jtik

View full text Add to dashboard Cite

This research aims to improve the search for tourist destinations in 50 datasets by using search queries to find relevant documents. By optimizing the search process, the goal is to create an accurate list of tourist destinations based on a given query. To achieve this, researchers used the TF-IDF and Cosine Similarity algorithms to retrieve and compare information, measuring similarity scores between search queries and tourist destinations in the dataset. Finally, the list of tourist destinations is ranked based on the similarity score measurement. The methods used are TF-IDF and Cosine Similarity. The fifty datasets containing text content documents were normalized through pre-processing stages, namely Case Folding, Stopword Removal, and Tokenization. Documents that have been normalized are then processed again through TF-IDF weighting. TF-IDF weighting is also applied to search queries. The similarity calculation between the TF-IDF vector from the document and the TF-IDF vector from the search query is carried out using Cosine Similarity to obtain a similarity score for each document based on the search query. Testing was carried out on 5 different queries, and precision testing results were obtained with an average value of 83%

show abstract

“…In the filtering process, stopword disposal. Stopwords are words that have no meaning or words that are less meaningful and often appear in a collection of words [25].…”

Section: Filteringmentioning

confidence: 99%

Automatic Scoring Using Term Frequency Inverse Document Frequency Document Frequency and Cosine Similarity

Yulita

Untoro

Praseptiawan

et al. 2023

SJI

View full text Add to dashboard Cite

Purpose: In the learning process, most of the tests to assess learning achievement have been carried out by providing questions in the form of short answers or essay questions. The variety of answers given by students makes a teacher have to focus on reading them. This scoring process is difficult to guarantee quality if done manually. In addition, each class is taught by a different teacher, which can lead to unequal grades obtained by students due to the influence of differences in teacher experience. Therefore the purpose of this study is to develop an assessment of the answers. Automated short answer scoring is designed to automatically grade and evaluate students' answers based on a series of trained answer documents.Methods: This is ‘how’ you did it. Let readers know exactly what you did to reach your results. For example, did you undertake interviews? Did you carry out an experiment in the lab? What tools, methods, protocols or datasets did you use The method used is TF-IDF-DF and Similarity and scoring computation. Theword weight used is the term Frequency-Inverse Documents Frequency -Document Frequency (TF-IDF-DF) method. The data used is 5 questions with each question answered by 30 students, while the students' answers are assessed by teachers/experts to determine the real score. The study was evaluated by Mean Absolute Error (MAE).Result: The evaluation results obtained Mean Absolute Error (MAE) with a resulting value of 0.123.Value: The word weighting method used is the Term Frequency Inverse Document Frequency DocumentFrequency (TF-IDF-DF) which is an improvement over the Term Frequency Inverse Document Frequency (TF-IDF) method. This method is a method of weighting words that will be applied before calculating the similarity of sentences between teachers and students.

show abstract

Essay Rating System Using Machine Learning

Cited by 3 publications

References 0 publications

Evaluation of Transformer-Based Neural Language Models for Writing Feedback and Automated Essay Scoring

Evaluation of Transformer-Based Neural Language Models for Writing Feedback and Automated Essay Scoring

Penerapan Algoritma TF-IDF dan Cosine Similarity untuk Query Pencarian Pada Dataset Destinasi Wisata

Automatic Scoring Using Term Frequency Inverse Document Frequency Document Frequency and Cosine Similarity

Contact Info

Product

Resources

About