Claim Missing Document
Check
Articles

Found 23 Documents
Search

Deep Learning Model for Sentiment Analysis on Short Informal Texts Sam Farisa Chaerul Haviana; Bagus Satrio Waluyo Poetro
Indonesian Journal of Electrical Engineering and Informatics (IJEEI) Vol 10, No 1: March 2022
Publisher : IAES Indonesian Section

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52549/ijeei.v10i1.3181

Abstract

This paper proposes a classification model to classify short informal texts. Those short informal texts were texts that were noisy, typos, irregular, and could consist of a very small number of words or even only a single word. The proposed model was trained using a dataset collected from student comments from an application called Evaluasi Dosen Oleh Mahasiswa (EDOM). This application assesses the lecturers using questionnaires filled out by students. It also records the student's comments but is not part of the evaluation calculation, therefore this work makes the data possible to be part of the assessment through sentiment analysis. This work focuses on building suitable preprocessing algorithm and building a simple deep learning network. The preprocessing algorithm was based on multiple word n-gram and Term Frequency-Inverse Document Frequency (TF-IDF) vectorization, and the network was built with a relatively shallow network. To evaluate the model in real usage, an application was built. The results were very convincing, reaching 0.979 in accuracy and 0.63 in F1-Score. Nonetheless, the imbalanced dataset was the only factor that needed to be investigated further for better overall performance.
GREEDY ALGORITHM IN GREEDY REDUCTION EDUCATION GAME BASED ON ANDROID PLATFORM Bagus Satrio Waluyo Poetro; Nurul Hidayah; Suryani Alifah
Journal of Telematics and Informatics Vol 6, No 1: March 2018
Publisher : Universitas Islam Sultan Agung

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (736.289 KB) | DOI: 10.12928/jti.v6i1.

Abstract

The greedy algorithm is one of the simplest algorithms to solve the optimization problem. The greedy algorithm is able to solve the problem quickly. Greedy Reduction game is a game of mathematical reduction that uses the theory of greedy algorithm in the process of determining the player's victory in playing the game. Players who play this game have to answer the problem by solving the reduction operation with subtracting the number by selecting available numbers on the game, until it reaches the specified number of questions. Players will either win the game or produce a serial number if the numbers the player chooses to subtract fewer numbers from the numbers used by greedy. Tests conducted on this research is to use black box testing method. Based on the tests performed, greedy algorithm can be applied in solving the existing problems in Greedy Reduction game.
Implementation of 128 bits Camellia Algorithm for Cryptography in Digital Image Bagus Satrio Waluyo Poetro
Proceeding of the Electrical Engineering Computer Science and Informatics Vol 3: EECSI 2016
Publisher : IAES Indonesia Section

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (510.83 KB) | DOI: 10.11591/eecsi.v3.1143

Abstract

Nowadays information technology requires stronger cryptographic algorithms. Camellia algorithm is also known for its suitability in terms of the implementation of both software and hardware as well as a high level of safety. The digital image is an image f (x, y) which having the spatial coordinates, and brightness levels are discrete. Unlike text messages, the image data has special features such as high redundancy and a high correlation between pixels. This research conducted a cryptographic process of the digital image using the Camellia algorithm. Comparisons were made on three digital image format .bmp, .jpg, .png with 128 bits key block Camellia algorithm. Results shows that Camellia cryptographic algorithms in digital image can successfully produce encrypted images. In addition, the same algorithm can also reproduce the image when decryption process.
PERANCANGAN SISTEM PAKAR UNTUK MENGIDENTIFIKASI GANGGUAN PADA TANAMAN JAMBU AIR MENGGUNAKAN METODE CERTAINTY FACTOR Ahmat Yusuf; Bagus Satrio Waluyo Poetro
Journal of Applied Science and Technology Vol 1, No 01 (2021): Januari 2021
Publisher : Universitas Islam Sultan Agung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30659/jast.1.01.1-5

Abstract

The state of Indonesia, especially in the area of Central Java, is one of the largest producers of water guava. Some farmers rely on the results of guava cultivation as the basis for the family's economy because the guava harvest is quite promising. However, in certain seasons, water guava plants are often attacked by various kinds of pest and disease season disturbances. So that it can reduce crop yields or even cause crop failure. To overcome this problem, fast and precise action is needed so that the guava plant is protected from various kinds of pests and diseases. The method of developing an expert system for disturbance in water guava plants is the best solution in order to identify disturbances in water guava plants as early as possible. The method used in this expert system is the certainty factor method. With this method, several experts are needed to provide data related to disturbances in water guava plants. The expert system for guava plant disorders is designed using a web-based PHP programming language. In designing the knowledge base, which is the core of this system, is made dynamically to make it easier to add new data.
Web Based Security System Academic Exam Questions Using Advanced Encryption Standard Bagus Satrio Waluyo Poetro; Sam Farisa Chaerul Haviana; Arief Budiman
Journal of Applied Science and Technology Vol 1, No 02 (2021): Juli 2021
Publisher : Universitas Islam Sultan Agung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30659/jast.1.02.13-22

Abstract

One way to measure the success of the academic process and achievement of student competence, is giving exams, from the smallest level namely daily tests, semester exams, to the highest level, namely the national exam. In an effort to maintain the security of exam question data, there is a data security method known as cryptography. In this research, a security system was designed that serves to protect exam questions so that data cannot be read by student before its time by using the Advanced Encryption Standard (AES) algorithm.  The AES algorithm is a type of symmetric algorithm where the encryption and decryption processes have the same key for the encryption and decryption processes. In the system that will be designed, the Caesar Cipher algorithm is used to form an additional key (seed) that is kept secret from the public. The results of this study indicate that AES encryption method can give results maximum so that the AES method can applied to virtual data storage system to protect the transmitted data.
Pendampingan Pengembangan Teknologi Informasi pada UMKM Rangkul Semarang Bagus Satrio Waluyo Poetro; Sam Farisa Chaerul Haviana
Indonesian Journal of Community Services Vol 4, No 1 (2022): May 2022
Publisher : Universitas Islam Sultan Agung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30659/ijocs.4.1.66-73

Abstract

UMKM Rangkul (Rakyat Semarang Kuliner) merupakan suatu wadah perkumpulan bagi para pelaku UMKM di bidang kuliner yang ada di Kecamatan Ngaliyan. Sasaran UMKM Rangkul adalah meningkatkan pemasaran lokal maupun global produk – produk UMKM sebesar rata-rata 5 % per tahun. Namun terbatasnya akses masyarakat di luar anggota UMKM atas informasi produk-produk yang ada pada paguyuban membuat potensi pemasaran sangatlah kecil. Oleh karena itu dibutuhkan pemanfaatan dan pengembangan teknologi informasi guna mendukung promosi pemasaran pada UMKM Rangkul ini. Salah satu solusinya adalah dengan katalog online yang dapat diakses masyarakat luas untuk lebih mudah mengetahui produk-produk apa saja yang ada pada paguyuban UMKM Rangkul semarang. Solusi dari permasalahan ini direalisasikan dalam bentuk pengembangan Katalog Online produk-produk anggota UMKM Rangkul semarang yang dapat diakses melalui internet dan handal dalam menyajikan informasi.UMKM Rangkul (Rakyat Semarang Kuliner) is an association forum for MSME actors in the culinary field in Ngaliyan District. The target of UMKM Rangkul is to increase local and global marketing of UMKM products by an average of 5% per year. However, the limited access of people outside of UMKM members to information on products in the association makes the marketing potential very small. Therefore, it is necessary to use and develop information technology to support marketing promotions in this UMKM Rangkul. One solution is to use an online catalog that can be accessed by the wider community to more easily find out what products are in the UMKM Rangkul Association, Semarang. The solution to this problem is realized in the form of developing an Online Catalog of products for members of UMKM Rangkul Semarang which can be accessed via the internet and is reliable in presenting information.
Sosialisasi Perawatan dan Pembuatan Face Shield bagi Tenaga Kesehatan Nuzulia Khoiriyah; Sri Mulyono; Bagus Satrio Waluyo Poetro
Abdimas: Jurnal Pengabdian Masyarakat Universitas Merdeka Malang Vol 6, No 3 (2021): August 2021
Publisher : University of Merdeka Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26905/abdimas.v6i3.5288

Abstract

Face shield is one of the important PPE (Personal Protective Equipment) for health workers in health services in the era of the COVID-19 pandemic as it is today. The limited number of face shield will certainly hamper the health services provided by the hospital. This Community Services activity was carried out at RSI Sultan Agung and RSIGM Sultan Agung to provide training on making face shield independently and also maintenance, so that face shield is more durable and not easily damaged. The tool used is a 3D printer which is easy to operate. 3D Printing technology will produce solid objects, and not like printing a sheet of paper on a printer that is commonly used. Design 3D models using CAD software. This activity was attended by 80 participants. 91.25% of participants are interested in making their own face shields as an effort to meet the need for face shields From the Community Services activities, around 100 face shields were produced which were handed over to the hospital. 
Sistem Deteksi Penyakit Mata Pterigium Berbasis Android Mengunakan Metode Convolutional Neural Network Nur Muhammad Syaifuddin; Dedy Kurniadi; Bagus Satrio Waluyo Poetro
TRANSISTOR Elektro dan Informatika Vol 5, No 1 (2023)
Publisher : Universitas Islam Sultan Agung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30659/ei.5.1.%p

Abstract

Bagi manusia, mata adalah salah satu indera yang sangat penting. Berbagai kegiatan dilakukan manusia melalui mata yang berfungsi menyerap informasi visual. Namun, mata adalah organ yang cukup rentan terhadap gangguan penglihatan, baik itu gangguan ringan maupun berat yang dapat menyebabkan kebutaan. Dengan terganggungnya fungsi penglihatan, maka berbagai aktivitas yang dilakukan seseorang akan terganggu, dan dari berbagai macam gangguan kesehatan mata, salah satunya adalah pterigium. Pterigium adalah suatu penyakit yang mengganggu fungsi penglihatan dan apabila kondisinya sudah sangat parah bisa menyebabkan kebutaan, oleh karena itu, deteksi dini pterigium dapat dijadikan sebagai solusi dalam upaya pencegahan pterigium, dan deteksi pterigium dapat dilakukan menggunakan algoritma Convolutional Neural Network. Penelitian ini menggunakan algoritma Convolutional Neural Network untuk mendeteksi pterigium yang direpresentasikan ke dalam sistem berbasis android. Hasil penelitian dieavaluasi berdasarkan akurasi pada tahap training dan validation, serta testing yang dilakukan secara manual pada smartphone android dan hasilnya akan dimasukkan ke dalam tabel confusion matrix. Dari 5 konfigurasi, model dengan kinerja terbaik pada tahap training dan validation adalah konfigurasi 4, yaitu nilai loss sebesar 0,2294, akurasi 88,89%, validasi loss 0,1052, dan validasi akurasi 100%. Dan pada tahap pengujian konfigurasi 1 dan 4, masing-masing nilai akurasi, precission, recall, dan F1 adalah 1 atau 100%.
SISTEM PENGAMANAN DOKUMEN PENGAJUAN HAK PATEN DENGAN MENGGUNAKAN METODE STEGANOGRAFI LINE SHIFTING Ahmad Saifurrohman; Bagus Satrio Waluyo Poetro; Moch Taufik
Prosiding Seminar Riset Mahasiswa Vol 1, No 1: Maret 2023
Publisher : Universitas Islam Sultan Agung

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Dokumen digital merupakan salah satu jenis data yang mudah dikirim dan diduplikasi melalui internet, sehingga rawan terkena serangan. Kemudahan ini bisa dimanfaatkan oleh pihak-pihak yang tidak bertanggung jawab untuk melakukan tindakan yang dapat merugikan hak cipta dokumen tersebut. Oleh karena itu, dibutuhkan teknik yang dapat menyembunyikan data dalam media sebagai watermark tanpa mengurangi atau merubah informasi yang adapada dokumen tersebut, teknik ini disebut steganografi. Dengan steganografi, data tersebut tidak terlihat oleh pihak yang tidak berwenang dan tidak menimbulkan kecurigaan terkait dengan keberadaan data tersebut. Saat melakukan uji sistem menggunakan white box, program ini tidak memiliki error saat dijalankan. Dengan percobaan pada lima kasus uji, hanya satu percobaan yang mengalami kegagalan saat di deocde ulang. Yaitu pengujian terhadap dokumen pdf hanya memiliki satu halaman dan pesan disisipkan sebanyak 35 karakter. Dalam penerapan steganografi line shifting untuk dokumen, masih ada karakter yang bukan merupakan karakter pesan yang tercetak pada saat proses decode.Keyword: Dokumen digital, Steganografi, Line Shifting.
Deteksi Ketersediaan Tempat Parkir menggunakan Mask R-CNN (Studi Kasus : Fakultas Teknologi Industri Universitas Islam Sultan Agung) Ahmad Ayudhawara; Bagus Satrio Waluyo Poetro; Muhammad Qomaruddin
Prosiding Seminar Riset Mahasiswa Vol 1, No 1: Maret 2023
Publisher : Universitas Islam Sultan Agung

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Tempat parkir merupakan sarana yang dibutuhkan hampir oleh setiap orang, dan lagi jumlah pemilik kendaraan bermotor di Indonesia terus meningkat setiap tahunnya. Hal tersebut dapat menjadi suatu permasalahan untuk pengguna kendaraan untuk dapat menemukan lahan parkir yang semakin sedikit. Di zaman sekarang ini sudah banyak metode yang digunakan untuk dapat memudahkan sistem perparkiran, salah satunya dengan menggunakan ilmu citra digital. Sistem ini menggunakan metode deteksi objek pada citra digital yaitu Mask R-CNN dimana metode tersebut dapat mendeteksi banyak objek pada satu frame dengan hasil keluaran yang lebih lengkap. Sistem ini dibangun agar memudahkan pengguna kendaraan mobil untuk mencari lahan parkir yang masih kosong. Pengujian sistem ini menggunakan tiga sampel video dengan tempat dan kondisi yang berbeda. Hasil pengujian menunjukkan untuk rata-rata akurasi ketersediaan tempat parkir sebesar 96%, namun akurasi untuk deteksi objek kendaraan masih belum bagus.Keyword: mask r-cnn, parkir, deteksi objek.