Nicholas Rayden, . (2025) IMPLEMENTASI EFFICIENTNET DAN GATED RECURRENT UNIT UNTUK MENDETEKSI POTENSI DEPRESI SERTA PENERAPAN MODEL MELALUI DESAIN USER INTERFACE. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
ABSTRAK.pdf Download (32kB) |
|
Text
AWAL.pdf Download (721kB) |
|
Text
BAB 1.pdf Restricted to Repository UPNVJ Only Download (94kB) |
|
Text
BAB 2.pdf Restricted to Repository UPNVJ Only Download (502kB) |
|
Text
BAB 3.pdf Restricted to Repository UPNVJ Only Download (333kB) |
|
Text
BAB 4.pdf Restricted to Repository UPNVJ Only Download (1MB) |
|
Text
BAB 5.pdf Download (97kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (157kB) |
|
Text
DAFTAR RIWAYAT HIDUP.pdf Restricted to Repository UPNVJ Only Download (116kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (4MB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (19MB) |
|
Text
ARTIKEL KI.pdf Restricted to Repository staff only Download (572kB) |
Abstract
This research is motivated by the increasing prevalence of depression globally, including in Indonesia, and the need for a practical and accurate technology-based early detection system. The proposed system combines facial expression analysis using EfficientNetB3 to recognize seven emotions (angry, sad, happy, disgust, fearful, neutral, and surprised) and Indonesian journaling text analysis using Gated Recurrent Unit (GRU) to detect negative language patterns. Image datasets are normalized and augmented, while text data undergo cleaning, tokenization, and stemming. The models are evaluated using K-Fold Cross Validation and integrated through a late fusion majority voting method. The combined model is implemented in a Streamlit-based interface supporting real-time camera and text input. Evaluation results show facial emotion model accuracy of 99.11% (training) and 97.14% (testing), and text model accuracy of 98.91% (training) and 96.28% (testing). The multimodal combination improves depression symptom detection accuracy. Expert validation by psychologists suggests the system has strong potential as an early screening tool, although professional assistance is still required for interpretation. This study delivers an efficient, accurate, and accessible mental health detection system with an interactive interface to support early diagnosis.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Additional Information: | [No.Panggil: 2110511149] [Pembimbing 1: Nur Hafifah Matondang] [Pembimbing 2: Kharisma Wiati Gusti] [Penguji 1: Ika Nurlaili Isnainiyah] [Penguji 2: Nurul Afifah Arifuddin] |
| Uncontrolled Keywords: | Depression Detection, EfficientNet, GRU, Multimodal Analysis, User Interface |
| Subjects: | B Philosophy. Psychology. Religion > BF Psychology Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software T Technology > T Technology (General) |
| Divisions: | Fakultas Ilmu Komputer > Program Studi Informatika (S1) |
| Depositing User: | NICHOLAS RAYDEN |
| Date Deposited: | 05 Aug 2025 07:03 |
| Last Modified: | 05 Aug 2025 07:03 |
| URI: | http://repository.upnvj.ac.id/id/eprint/37342 |
Actions (login required)
 |
View Item |