KEBI 1.0: Indonesian Spelling Error Detection System for Scientific Papers using Dictionary Lookup and Peter Norvig Spelling Corrector

Abstract

Many Indonesian spelling errors occur in research papers published to the public, closely related to academics in all institutions such as research institutions, government, schools, and universities. The spelling errors usually writing punctuation, writing letters, writing words, writing words originating from foreign or regional languages (uptake words), using affixed words, and writing ineffective sentences. The mistakes made by the academics then become a cycle in the academic environment. They usually provide guidance for writing an undergraduate thesis, thesis, dissertations to students, or the other forms of documents and scientific papers. Therefore, the research proposed the application to facilitate all authors of scientific papers in producing quality scientific works based on the General Guidelines for Indonesian Spelling published by the Agency for Development and Language Development. The application is named KEBI 1.0 Checker (Indonesian Spelling Error 1.0 Checker), a web-based application with a built-in algorithm to detect and correct Indonesian Spelling in scientific papers. The experiment result shows that the application has given the best accuracy performance to correct the non-standard words, and typographical errors reached 100% and 55,52%, respectively. The application also has been detected 209 meaningless words. The application processing time is relatively low, the average time needed to correct non-standard words is 0.016 seconds, and typo words are 14.58 seconds. KEBI 1.0 Checker is helpful for the end-user in academics but needs to improve the vocabulary of the large corpus in various fields of science for correcting typo words.

A Rule-based Spelling Checker for Correcting Punctuation Errors in Indonesia Text using KEBI 1.0 Checker

“…This study continues from previous research by Fahrudin et al (2021) that discusses the ability of the error detection and correction of Indonesian non-standard words and typos in an application called KEBI 1.0 Checker. Therefore, this study proposes adding a feature for a spelling checker for correcting punctuation errors in Indonesian text using a rule-based.…”

The Effect of Information Quality and Service Quality on User Satisfaction of the Government of Kabupaten Malang

“…These tests include: (1) Multicollinearity test: Assess the presence of multicollinearity, it a rises when the independent variables in the regression model exhibit a strong correlatio n or interdependence with one another. [21], (2) Autocorrelation test: Evaluate whether there is autocorrelation, which refers to the presence of correlation among the residuals (errors) of the regression model [22], (3) Heteroscedasticity test: Determine if there is heteroscedasticity, which indicates unequal variances in the residuals across different levels of the independent variables [23], (4) Normality test: Check whether the residuals of the regression model follow a normal distribution [24], (5) Hypothesis testing: Hypothesis tests can be conducted [25] to assess the significance of coefficients and determine the relationships between independent variables and the dependent variable.…”

SPECIL: Spell Error Corpus for the Indonesian Language

“…Non-word and real-word errors are the most common causes of spelling errors in Indonesians. Real-word error means that the word is correct but used in the wrong context, whereas non-word error means no meaning in the dictionary [10]. Non-word errors include transposition, substitution, insertion, and deletion [11].…”