Caesarardhi, Muhammad Rasyad (2023) Implementasi Aplikasi Peringkasan Teks Otomatis Untuk Atlas Penyakit Menular Pada Manusia Menggunakan Metode Ordered Abstractive Summarization. Other thesis, Institut Teknologi Sepuluh Nopember.
Text
05211940000077-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (5MB) | Request a copy |
Abstract
Tingkat kematian dari pandemic kedua setelah pandemic pertama The Justinian Plague (A.D. 541 hingga 544) telah mencapai 15-40%. Depopulasi manusia dengan loss rate 50-60% telah diperkirakan terjadi karena pandemi tersebut. Atlas of Human Infectious Disease (AHID) telah mengumpulkan distribusi dan faktor penentu untuk sebagian besar penyakit menular pada manusia. AHID dilengkapi dengan informasi mengenai infectious agents, clinical, dan epidemiological. Sekarang, setiap penyakit di dalam AHID masih belum memiliki visualisasi dengan ringkasan kunci di dalamnya. Hal tersebut dapat membantu orang-orang yang tidak memiliki latar belakang medis untuk memahami tentang penyakit menular secara lebih cepat dan tepat. Menariknya, data di dalam AHID telah memiliki bentuk yang semi-structured, sehingga kita harus mengotomasi proses untuk merepresentasikannya kedalam model pengetahuan. Penelitian sebelumnya telah menunjukkan perkembangan yang cukup baik dalam model untuk peringkasan teks. Seq2Seq model telah mencapai skor ROUGE-1 sebesar 28,42 dan model Bringing in Order to Abstractive Summarization (BRIO) telah mencapai skor ROUGE-1 sebesar 49,07 dalam dataset Extreme Summarization (XSum). Maka dari itu, dalam tugas akhir ini, telah dilakukan peringkasan teks untuk data AHID menggunakan model BRIO dan mendapatkan hasil skor ROUGE-1 sebesar 43,86 . Hasil model tersebut kemudian dapat digunakan untuk menambahkan penyakit baru yang belum dicantumkan dalam AHID seperti Covid-19 dan mampu menghasilkan ringkasan teks secara otomatis untuk setiap atributnya. Hasil dari peringkasan teks telah disampaikan melalui aplikasi kamus berbasis web.
================================================================================================================================
The death rate for the second pandemic after the first pandemic The Justinian Plague (A.D. 541 to 544) had reached 15-40%. Human depopulation with a loss rate of 50-60% also has been estimated due to the pandemics. Atlas of Human Infectious Disease (AHID) has captured the distribution and determinants for most infectious diseases for humans. AHID is complemented with information about infectious agents, clinical, and epidemiological. Now, every single disease that AHID explains still does not have a visualization with key-point summary on it. It helps people with no medical backgrounds to understand the infectious diseases faster and properly. Interestingly, AHID is semi-structured in terms of data, so we need to automate the process of representing it to a knowledge model. Previous research has shown a rather good improvement in the model of text summarization. Seq2Seq model has reached a ROUGE-1 score of 28,42 and the latest model Bringing in Order to Abstractive Summarization (BRIO) has reached a ROUGE-1 score of 49,07 on the extreme summarization dataset (XSum). Therefore, in this research, we will do text summarization for AHID data using the BRIO model and reached score of ROUGE-1 of 43,86. The resulting model therefore can be used to add another disease that is not included yet in AHID such as Covid-19 and output automated text summarization for each one of the attributes. The result of text summarization will be delivered as a web-based dictionary.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | AHID, BRIO, Peringkasan Teks, Text Summarization |
Subjects: | Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. |
Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis |
Depositing User: | Muhammad Rasyad Caesarardhi |
Date Deposited: | 25 Jul 2023 02:31 |
Last Modified: | 25 Jul 2023 02:31 |
URI: | http://repository.its.ac.id/id/eprint/99361 |
Actions (login required)
View Item |