Implementasi Algoritma Hamming Distance Dan Text Similarity Pada Asesmen Komputerisasi Anak Disleksia

Amri, Arni Muarifah (2022) Implementasi Algoritma Hamming Distance Dan Text Similarity Pada Asesmen Komputerisasi Anak Disleksia. Masters thesis, Institut Teknologi Sepuluh November.

[thumbnail of 6022201012-Master_Thesis.pdf] Text
6022201012-Master_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2024.

Download (2MB) | Request a copy

Abstract

Disleksia merupakan gangguan belajar spesifik yang dapat memberikan pengaruh negatif terutama terhadap keberhasilan belajar anak. Deteksi dini dan pengobatan berkelanjutan dapat membantu anak-anak mencapai keberhasilan yang lebih baik di sekolah. Anak disleksia memerlukan beberapa evaluasi dan pengobatan selama program rehabilitasi dari terapis dan psikolog. Selama ini penilaian dilakukan secara manual dan direkam menggunakan dokumen kertas, selain itu anak harus bertemu dengan terapis secara fisik. Metode assessment komputerisasi untuk menilai anak Disleksia dengan menggunakan pemrosesan teks adalah pendekatan baru untuk mencatat kemajuan rehabilitasi dengan lebih baik selain itu menyediakan aplikasi evaluasi di rumah sehingga pembelajaran jarak jauh juga dapat dilakukan terutama di era Pandemi. Dalam penelitian ini, data diambil dari 24 anak Disleksia selama program terapi 3 bulan mereka. Program penilaian dilakukan setelah anak mengikuti program rehabilitasi dari terapis. Selama kurun waktu 3 bulan tersebut, terdapat 2 kali jadwal rehabilitasi dan 2 kali jadwal penilaian.Penilaian menggunakan tiga kriteria yaitu Writing Letter Series (WLS), Arranging Sentences (AS), dan Copying Sentences (CS) yang harus dilakukan oleh setiap anak Disleksia. Aplikasi komputer untuk pengujian dikembangkan dengan sistem penilaian yang diimplementasikan menggunakan teknik pemrosesan teks. Sistem Hamming distance dan Text similarity diterapkan untuk menilai hasil tes dan kemudian dibandingkan dengan metode penilaian manual yang dilakukan oleh terapis. Pada penelitian ini implementasi algoritma Hamming distance dilakukan pada kriteria WLS dengan mencapai akurasi 83% dibandingkan penilaian manual. Untuk kriteria AS, dilakukan teknik Text Similarity dengan mencapai akurasi 96% dibandingkan penilaian manual, dan pada CS mencapai 88%. Hasil penelitian menunjukkan tingginya potensi penerapan teknik pemrosesan teks dalam menilai anak Disleksia.
================================================================================================
Dyslexia is a specific learning disorder that can have a negative influence, especially on children's learning success. Early detection and continuous treatment can help children to achieve better success in school. Children with Dyslexia need several evaluations and treatment during the rehabilitation programs from therapists and psychologists. Until recently, the assessment is carried out manually and recorded by using paper documents, besides the children must meet the therapist physically. A computerized method for assessing Dyslexia children by using text processing is a new approach to better record the rehabilitation progress and provide a home evaluation application so that some social distancing then can also be done especially in the era of the Pandemic. In this study, the data was taken from 24 Dyslexic children during their 3 months therapy program. The assessment program is done after the children follow the rehabilitation program from the therapist. During that 3 months period, there were 2 times scheduled for rehabilitation and 2 times scheduled for assessment. The assessment uses criteria, namely writing letter series, arranging sentences, and copying sentences that must be done by each Dyslexia child. Computer application for testing is developed with a scoring system is implemented using text processing technique. Hamming distance and Auto Score system were applied to score the test result and then be compared with the manual scoring method performed by the therapists.In this study Hamming distance done for the criteria of the series of letters reaching an accuracy of 83%, then auto score on the criteria of compiling sentences reaches 96% and compose sentences reaches 88%. The results showed that auto scores showed high accuracy on one criterion, so this automated scoring can be used as a practical and efficient assessment tool to help dyslexic children.

Item Type: Thesis (Masters)
Uncontrolled Keywords: text mining, text similarity, algoritma hamming distance, hamming distance algorithm
Subjects: Q Science > QA Mathematics
Q Science > QA Mathematics > QA76.6 Computer programming.
Q Science > QA Mathematics > QA9.58 Algorithms
T Technology > T Technology (General) > T57.6 Operations research--Mathematics. Goal programming
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User: Arni Muarifah Amri
Date Deposited: 08 Jun 2022 07:25
Last Modified: 31 Oct 2022 01:53
URI: http://repository.its.ac.id/id/eprint/94903

Actions (login required)

View Item View Item