Klasifikasi gangguan depresi pada Twitter menggunakan algoritma Synenketch

Nurjannah, Nurjannah (2019) Klasifikasi gangguan depresi pada Twitter menggunakan algoritma Synenketch. Diploma thesis, Uin Sunan Gunung Djati Bandung.

[img]
Preview
Text (COVER)
1_cover.pdf

Download (49kB) | Preview
[img]
Preview
Text (ABSTRAK)
2_abstrak.pdf

Download (48kB) | Preview
[img]
Preview
Text (DAFTAR ISI)
3_daftarisi.pdf

Download (1MB) | Preview
[img]
Preview
Text (BAB 1)
4_bab1.pdf

Download (1MB) | Preview
[img] Text (BAB 2)
5_bab2.pdf
Restricted to Registered users only

Download (1MB) | Request a copy
[img] Text (BAB 3)
6_bab3.pdf
Restricted to Registered users only

Download (1MB) | Request a copy
[img] Text (BAB 4)
7_bab4.pdf
Restricted to Registered users only

Download (1MB) | Request a copy
[img] Text (BAB 5)
8_bab5.pdf
Restricted to Registered users only

Download (1MB) | Request a copy
[img] Text (DAFTAR PUSTAKA)
9_daftarpustaka.pdf
Restricted to Registered users only

Download (1MB) | Request a copy

Abstract

INDONESIA : Menurut data dari Word Health Organization (WHO) terdapat 35 juta orang yang mengalami depresi, 60 juta orang yang terkena bipolar dan 21 juta terkena skizofrenia, serta 47,5 juta terkena dimensia [1]. Tidak sedikit orang meluapkan segala isi hati dan perasaannya melalui media sosial yang merujuk kepada gejala gangguan depresi. Untuk mengenali gejala gangguan depresi seseorang pada media sosial tidak mudah karena penggunaan bahasa yang nonformal, singkatan serta multi bahasa. Klasifikasi gejala gangguan depresi ini berdasarkan tujuh gejala dasar (tidur terlalu banyak atau sedikit, agitasi, nafsu makan, kehilangan energi, perasaan bersalah, sulit konsentrasi, pemikiran tentang bunuh diri). Pengambilan data dilakukan menggunakan crawler. Selanjutnya pelabelan data dilakukan secara manual dari hasil penyaringan 3.056 tweet menjadi 369 tweet digunakan dalam klasifikasi. Terdapat tiga tahap yang dilakukan yaitu pelabelan dataset, preproses kemudian klasifikasi. Tahap klasifikasi menggunakan beberapa rule dari Algoritma Synesketch. Berdasarkan hasil eksperimen ini mengeluarkan hasil gejala gangguan depresi yang dominan mengenai tweet bunuh diri dengan akurasi sebesar 86,91 %. Penggunaan kamus kata ini sesuai dari indeks pengetauan yang telah diverifikasi oleh data ahli. ENGLISH : According to data from the Word Health Organization (WHO) there are 35 million people who are depressed, 60 million people are affected by bipolar disorder and 21 million are affected by schizophrenia, and 47.5 million are affected by dementia [1]. Not a few people pour all their hearts and feelings through social media that refers to symptoms of depressive disorders. To recognize the symptoms of depression in a person on social media is not easy because of the use of non-formal language, abbreviations and multi-language. This classification of depressive symptoms is based on seven basic symptoms (sleeping too much or too little, agitation, appetite, loss of energy, feelings of guilt, difficulty concentrating, thoughts about suicide). Data retrieval is done using a crawler. Furthermore, data labeling is done manually from the filtering results of 3,056 tweets to 369 tweets used in classification. There are three stages to be carried out, namely labeling the dataset, preprocessing then classification. The classification stage uses several rules from the Synesketch algorithm. Based on the results of this experiment, the results of dominant depressive symptoms regarding suicide tweets with an accuracy of 86.91%. The use of this word dictionary is in accordance with the index of knowledge that has been verified by expert data.

Item Type: Thesis (Diploma)
Uncontrolled Keywords: Depresi; Rule; Synesketch; gejala gangguan depresi
Subjects: Differential and Developmental Psychology > Psychology of Young People Twelve to Twenty
Differential and Developmental Psychology > Psychology of Adults
Differential and Developmental Psychology > Stress
Divisions: Fakultas Sains dan Teknologi > Program Studi Teknik Informatika
Depositing User: Nurjannah Nurjannah
Date Deposited: 10 Jan 2020 08:18
Last Modified: 10 Jan 2020 08:18
URI: http://digilib.uinsgd.ac.id/id/eprint/28678

Actions (login required)

View Item View Item