Tujuan dari penelitian ini menerapkan metode Levenshtein Distance untuk mendeteksi kesalahan kata dalam penulisan dokumen skripsi. Penelitian ini menggunakan model proses data mining Cross-Industry Standard Process for Data Mining (CRISP-DM) yaitu Business Understanding Phase, Data Understanding Phase, Data Preparation Phase, Modelling Phase, Evaluation Phase dan Deployment Phase. Hasil penelitian ini berupa aplikasi deteksi kesalahan kata pada dokumen skripsi untuk memvisualisasi proses kerja metode Levenshtein Distance pada salah satu tahap preprocessing yaitu tokenisasi yaitu menghilangkan angka, simbol, emoticon serta pengubahan setiap huruf menjadi huruf kecil. Berdasarkan evaluasi menggunakan metode Confusion Matrix diperoleh tingkat akurasi sebesar 93,25%.

The purpose of this study is to apply the Levenshtein Distance method to detect word errors in the writing of thesis documents. This study uses Cross-Industry Standard Process for Data Mining (CRISP-DM) data mining process model, namely Business Understanding Phase, Data Understanding Phase, Data Preparation Phase, Modelling Phase, Evaluation Phase and Deployment Phase. The result of this study is a word error detection application in the thesis document to visualize the working process of levenshtein distance method at one of the preprocessing stages, namely tokenization that eliminates numbers, symbols, emoticons and converts each letter into lowercase letters. Based on the evaluation using confusion matrix method obtained an accuracy rate of 93,25%.