PENERAPAN OPTICAL CHARACTER RECOGNITION MENGGUNAKAN ALGORITMA CONVOLUTION NEURAL NETWORK UNTUK SISTEM PENCATATAN KEUANGAN PRIBADI BERBASIS ANDROID

Pranarendra Dwikurnia, . (2024) PENERAPAN OPTICAL CHARACTER RECOGNITION MENGGUNAKAN ALGORITMA CONVOLUTION NEURAL NETWORK UNTUK SISTEM PENCATATAN KEUANGAN PRIBADI BERBASIS ANDROID. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.

[img] Text
ABSTRAK.pdf
Download (778kB)
[img] Text
AWAL.pdf
Download (3MB)
[img] Text
BAB 1.pdf
Restricted to Repository UPNVJ Only
Download (1MB)
[img] Text
BAB 2.pdf
Restricted to Repository UPNVJ Only
Download (6MB)
[img] Text
BAB 3.pdf
Restricted to Repository UPNVJ Only
Download (2MB)
[img] Text
BAB 4.pdf
Restricted to Repository UPNVJ Only
Download (16MB)
[img] Text
BAB 5.pdf
Download (506kB)
[img] Text
DAFTAR PUSTAKA.pdf
Download (1MB)
[img] Text
RIWAYAT HIDUP.pdf
Restricted to Repository UPNVJ Only
Download (108kB)
[img] Text
LAMPIRAN.pdf
Restricted to Repository UPNVJ Only
Download (734kB)
[img] Text
HASIL PLAGIARISME.pdf
Restricted to Repository staff only
Download (177kB)
[img] Text
ARTIKEL KI.pdf
Restricted to Repository staff only
Download (680kB)

Abstract

Personal financial recording is becoming increasingly important in today's digital era. Data from the Financial Services Authority (OJK) shows a significant increase in financial literacy among Indonesians, positively impacting individuals' ability to manage their finances. Daily financial recording is an effective method for organizing personal finances, yet most current financial recording applications rely on manual input, which is less practical for users. This research aims to implement Optical Character Recognition (OCR) based on Convolutional Neural Network (CNN) to scan receipts in an Android-based financial recording application. This study employs methods including problem identification, literature review, data collection, OCR implementation, data extraction from receipts, OCR result testing, Android UI design, application development, OCR integration, and application testing. The results indicate that the developed financial recording application functions well. The CNN algorithm successfully recognizes and extracts text from receipts with adequate accuracy. OCR integration into the application was successfully achieved using NGROK tunneling techniques. Evaluations show that the application can detect date, item list, convenience store name, and total expenditure from the receipt, and fulfills all test scenarios. The conclusion of this research is that the use of CNN algorithm-based OCR in Android-based financial recording applications makes it easier for users to record expenses by scanning shopping receipts automatically with an accuracy value of 66.9. This provides a practical and efficient solution in recording personal finances.

Item Type: Thesis (Skripsi)
Additional Information: [No.Panggil: 2010511097] [Pembimbing: Indra Permana Solihin, Novi Trisman Hadi] [Penguji 1: Bayu Hananto] [Penguji 2: Nurul Afifah Arifuddin]
Uncontrolled Keywords: Convolutional Neural Network, Optical Character Recognition, financial recording application, Android.
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Q Science > QA Mathematics > QA76 Computer software
Divisions: Fakultas Ilmu Komputer > Program Studi Informatika (S1)
Depositing User: PRANARENDRA DWIKURNIA
Date Deposited: 05 Sep 2024 07:33
Last Modified: 05 Sep 2024 07:33
URI: http://repository.upnvj.ac.id/id/eprint/31655

Actions (login required)

View Item View Item