Claim Missing Document
Check
Articles

Found 36 Documents
Search

Klasifikasi Genus Tanaman Anggrek Berdasarkan Citra Kuntum Bunga Menggunakan Metode Convolutional Neural Network (Cnn) Mohammad Ikhsan Syahputra; Agung Toto Wibowo
eProceedings of Engineering Vol 7, No 2 (2020): Agustus 2020
Publisher : eProceedings of Engineering

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

Abstract

Abstrak Keindahan bunga membuat tanaman tersebut memiliki banyak peminatnya sehingga tanaman anggrek mempunyai nilai jual yang tinggi. Banyaknya genus tanaman anggrek membuat masyarakat umum sulit untuk membedakan genus tanaman anggrek yang berjumlah kurang lebih berjumlah 900 genus tanaman anggrek. Dengan membuat sistem yang dapat mengenali dan mengklasifikasi genus tanaman anggrek akan mempermudah masyarakat umum dalam mengenali genus-genus anggrek yang mempunyai ciri khasnya masing-masing sehingga tanaman anggrek dapat dibudidayakan dengan optimal sesuai dengan ciri khas genusnya. Sistem ini dikembangkan dengan metode Convolutional Neural Network (CNN) yang dibangun menggunakan K-Fold Cross Validation untuk memvalidasi struktur model CNN, memiliki dataset sebanyak 900 data citra kuntum bunga anggrek dari empat genus anggrek yang umum dibudidayakan di Indonesia, yakni genus Cattleya, Dendrobium, Oncidium, dan Phalaenopsis dengan hasil performansi akurasi pengujian sebesar 97,00%. Kata Kunci: Klasifikasi Genus Anggrek, Convolutional Neural Network (CNN), K-Fold Cross Validation, Kuntum Bunga Anggrek, Genus Tanaman Anggrek.
Klasifikasi Genus Tanaman Sukulen Menggunakan Convolutional Neural Network Hamad Fauzi Jessar; Agung Toto Wibowo; Ema Rachmawati
eProceedings of Engineering Vol 8, No 2 (2021): April 2021
Publisher : eProceedings of Engineering

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

Abstract

Abstrak Tanaman sukulen merupakan jenis tanaman hias yang banyak ditemukan jenisnya di indonesia. Tanaman sukulen mempunyai banyak jenis genus yang dimana setiap genus mempunyai ciri dan karakteristik yang beragam sehingga sulit untuk mengidentifikasi jenis genus pada tanaman sukulen.Oleh karena itu, penulis membuat sebuah sistem yang dapat mengenali jenis genus tanaman sukulen melalui gambar menggunakan metode Convolutional Neural Network (CNN). CNN merupakan salah satu teknik deep learning yang dapat digunakan untuk mengenali objek dua dimensi seperti gambar dan video. CNN memiliki banyak jenis arsitektur jaringan, arsitektur jaringan CNN yang digunakan penulis untuk membangun sistem ini adalah custom arsitektur dan penulis juga menggunakan k fold cross validation yang bertujuan untuk memastikan keakuratan akurasi yang dihasilkan oleh model sistem. Penelitian dilakukan penulis dengan membandingkan antara model yang dilatih menggunakan dataset berwarna (RGB) dan model yang dilatih menggunakan dataset grayscale. Dari hasil penelitian didapatkan bahwa model yang dilatih menggunakan dataset berwarna mempunyai akurasi testing yang lebih tinggi dibandingkan dengan model yang dilatih menggunakan dataset grayscale. Akurasi testing yang dihasilkan model yang dilatih dengan dataset berwarna sebesar 93% sedangkan model yang dilatih menggunakan dataset grayscale sebesar 64%. Kata kunci : convolutional neural network, deep learning, k fold, grayscale, RGB Abstract Succulent plants are a type of ornamental plant that are found in many species in Indonesia. Succulent plants have many types of genera, where each genus has various characteristics and characteristics making it difficult to identify the type of genus in succulent plants. Therefore, the authors created a system that can recognize the types of succulent plant genera through images using the Convolutional Neural Network (CNN) method. ). CNN is a deep learning technique that can be used to recognize two-dimensional objects such as images and videos. CNN has many types of network architectures, the CNN network architecture used by the author to build this system is a custom architecture and the author also uses k fold cross validation which aims to ensure the accuracy of the accuracy generated by the system model. The research was conducted by the author by comparing the model trained using the color dataset (RGB) and the model trained using the grayscale dataset. From the results of the study, it was found that models trained using color datasets have higher accuracy than models trained using grayscale datasets, namely 93% for models with color datasets while models with grayscale datasets have 64% accuracy. Keywords: convolutional neural network, deep learning, k fold cross validation, RGB, grayscale
Collaborative Filtering Based Food Recommendation System Using Matrix Factorization Muhammad Bayu Samudra Siddik; Agung Toto Wibowo
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 3 (2023): Juli 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i3.6049

Abstract

A recommendation system is a method that provides suggestions of items that might users like. There are many domains that can be recommended, one of the most demanded domains by users today is food. In the era of big data, food choices from the large amount of data make it difficult for users to choose the right food for them. The collaborative filtering (CF) approach is considered capable of providing accurate and high quality item suggestions. One of the algorithms that can provide good performance results from the CF approach is Matrix Factorization (MF). This study aims to test a dataset that contains product ratings of food items using three MF algorithms, which are Singular Value Decomposition (SVD), SVD with Implicit Ratings (SVD++), and Non-Negative Matrix Factorization (NMF). Different latent factors are also used for the purpose of improving the performance of the proposed recommendation system algorithm. The dataset used is Amazon Fine Food Reviews. The study shows NMF and SVD++ as the best algorithm for generating user rating predictions for items. NMF has the smallest average prediction error as measured by MAE which is 0.7311. While SVD++ obtains the smallest prediction error value of 1.0607 as measured using RMSE. In addition to these results, the top-n evaluation also shows that the proposed algorithm performs quite well. The hit ratio value for each different n-item always increases proportionally to the number of recommended n-items. The highest hit ratio value is generated from the SVD++ algorithm of 0.0025 on n-item recommendations of 25 items. Overall it can be said that the proposed algorithm has good performance in providing item recommendations.
Music Recommender System Based on Play Count Using Singular Value Decomposition++ Muhamad Elang Ramadhan; Agung Toto Wibowo
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 3 (2023): Juli 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i3.6424

Abstract

The availability of digital music content on various music streaming services, which is constantly growing, has increased the need for recommender systems (RS) to assist users in finding music that suits their taste without the need of searching manually. One of the commonly used paradigms is Collaborative Filtering (CF). In CF, the input used to predict ratings can take the form of explicit or implicit input from user feedback. In the music domain, implicit feedback such as the number of music plays can be utilized to predict a user's music preferences. Singular Value Decomposition++ is one of the Matrix Factorization (MF) algorithms that can leverage implicit feedback and address the sparsity issue. In this research, a music recommender system is built using the Million Song Dataset (MSD) Subset from The Echo Nest, utilizing SVD++ algorithm. Additionally, the performance of the built system is measured through k-fold cross-validation using the evaluation metrics RMSE and NDCG. The performance measurement results using RMSE and NDCG in 5-fold cross-validation yield an RMSE of 0.4423, NDCG@5 of 0.8232, and NDCG@10 of 0.8231 for the top 10 items.
Pelatihan Berpikir Komputasional untuk Peningkatan Kompetensi Guru Telkom Schools sebagai Bagian dari Gerakan PANDAI Muhammad Arzaki; Selly Meliana; Ema Rachmawati; Ade Romadhony; Agung Toto Wibowo; Bambang Pudjoatmodjo; Bedy Purnama; Dodi Wisaksono Sudiharto; Fat'hah Noor Prawira; Fazmah Arif Yulianto; Putu Harry Gunawan; Rimba Whidiana Ciptasari
I-Com: Indonesian Community Journal Vol 3 No 3 (2023): I-Com: Indonesian Community Journal (September 2023)
Publisher : Fakultas Sains Dan Teknologi, Universitas Raden Rahmat Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33379/icom.v3i3.2988

Abstract

Berpikir komputasional (BK) atau computational thinking (CT) merupakan salah satu keahlian esensial yang diperlukan sumber daya manusia Indonesia dalam rangka menghadapi revolusi industri 4.0 dan masyarakat 5.0. Gerakan PANDAI (Pengajar Era Digital Indonesia) merupakan suatu gerakan nasional yang merupakan kolaborasi nirlaba antara komunitas Bebras Indonesia, Kementerian Pendidikan dan Kebudayaan Indonesia, dan Google Indonesia dalam rangka meningkatkan kompetensi BK yang dimiliki oleh guru sekolah dasar dan menengah. Pada tahun 2022, Biro Bebras Universitas Telkom mengadakan pelatihan BK kepada lebih dari 60 guru Telkom Schools sebagai bagian dari gerakan ini. Pelatihan ini terdiri dari lima tahapan besar yang meliputi lokakarya luring, pembelajaran mandiri, lokakarya daring, dan dua kegiatan microteaching. Hasil analisis kuantitatif menunjukkan peningkatan kemampuan konseptual peserta terkait BK, meskipun masih banyak hal yang perlu dibenahi dari sisi kemampuan teknis dalam pengerjaan soal-soal BK.
Fashion Recommendation System using Collaborative Filtering Muhammad Khiyarus Syiam; Agung Toto Wibowo; Erwin Budi Setiawan
Building of Informatics, Technology and Science (BITS) Vol 5 No 2 (2023): September 2023
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v5i2.3690

Abstract

Collaborative Filtering is an method used to build a recommendation system with the concept that conclusions from different clients are used to anticipate things that may be of interest to users. This research uses data from Rent the Runway and the method used is Item-based Collaborative filtering, where the system will look for similarities in products that have been purchased by customers and then look for predictive values. Fashion requires recommendations because it plays a crucial role in helping individuals express their identity, personal style, and personality through clothing choices, accessories, and dressing styles.The recommendation system uses the item method based on analyzing the number of purchases or sales and grouping according to each product category so that it can help consumers in choosing fashion products. It was found that the use of Adjusted Cosine Similarity produces better recommendations with an average MAE value of 0.2750, while Cosine Similarity with an average MAE difference of 0.3989. This proves that the use of adjusted cosine similarity can produce better recommendations because the adjustment algorithm not only considers user behavior, but also produces lower performance errors.
Movie Recommendation System Based on Synopsis Using Content-Based Filtering with TF-IDF and Cosine Similarity Armadhani Hiro Juni Permana Juni Permana; Agung Toto Wibowo
International Journal on Information and Communication Technology (IJoICT) Vol. 9 No. 2 (2023): Vol.9 No. 2 Dec 2023
Publisher : School of Computing, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21108/ijoict.v9i2.747

Abstract

Recommendation systems have become an interesting topic in the field of artificial intelligence and data analysis. In the current era of technological advancement, the entertainment industry is rapidly growing, particularly the film industry, which is highly popular among the public due to their enthusiasm for watching movies. The increasing number and variety of films with various genres and titles have made it challenging for users to choose a film. To assist them in selecting movies, the presence of a recommendation system is necessary to provide information or film recommendations based on user interests and preferences. In this research, the development of the recommendation system will utilize the content-based filtering method, employing the TF-IDF algorithm and cosine similarity. The dataset used in this study is derived from publicly available data (MovieLens). The results of this research demonstrate that the TF-IDF and cosine similarity algorithms provide recommendations that align with the viewers' interests, as measured by precision, recall, and f1-score calculations.
MOVIE RECOMMENDATION SYSTEM USING HYBRID FILTERING WITH WORD2VEC AND RESTRICTED BOLTZMANN MACHINES Pradana, Muhammad Aryuska; Wibowo, Agung Toto
JIPI (Jurnal Ilmiah Penelitian dan Pembelajaran Informatika) Vol 9, No 1 (2024)
Publisher : STKIP PGRI Tulungagung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29100/jipi.v9i1.4306

Abstract

Recommender systems are designed to provide interesting information to users and assist them in making choices. With the help of a recommender system, users can feel more comfortable using an application. In this final project, we will implement a hybrid filtering method using two techniques: Word2Vec as the algorithm for content-based filtering and Restricted Boltzmann Machine for collaborative filtering. The Word2Vec algorithm will utilize a pre-trained model provided by Google, while the Restricted Boltzmann Machine algorithm will utilize the TensorFlow library. The dataset used for this project will be Movie Lens. The goal of this final project is to evaluate the accuracy and performance of the recommender system using various metrics such as Precision and Normalized Discounted Cumulative Gain.
Sistem Rekomendasi Produk Elektronik Berbasis Collaborative Filtering Manggunakan Matrix Factorization Rajib , Ustami; Wibowo, Agung Toto
eProceedings of Engineering Vol. 12 No. 3 (2025): Juni 2025
Publisher : eProceedings of Engineering

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

Abstract

Sistem rekomendasi adalah suatu program yangmelakukan prediksi suatu item, dalam pembuatan sistemrekomendasi terdapat Beberapa metode yang dapat digunakandiantaranya Collaborative Filtering karena dianggap mampumemberikan saran item yang lebih akurat. pendekatanCollaborative Filtering karena dianggap mampu memberikansaran item yang lebih akurat. Pada penelitian ini akan dibuatsistem rekomendasi menggunakan 3 Algoritma Turunan MFyaitu Singular Value Decomposition (SVD), SVD++, NonNegative Matrix Factorization NMF terhadap dataset AmazonReview dengan Studi Kasus Elektronik, ini perlu diaplikasikandalam penelitian sistem rekomendasi, karena data Elektronikini mempunyai jumlah data yang sangat besar. Dalampenelitian ini akan dilakukan uji coba terhadap beberapaparameter yang meliputi n-epochs, n-factor dalam mekanisme5-fold cross-validation. Untuk menangani data yang terlalubesar, penulis melakukan random sampling sebesar 25% daritotal dataset untuk mengurangi beban komputasi. Dari hasil ujicoba didapatkan performansi rata-rata terbaik MAE = 1.0384dan RMSE = 1.3139 yaitu pada Algoritma SVD. Kata kunci— Produk Elektronik, Sistem Rekomendasi,Collaborative Filtering, Matrix Factorization, Cross Validation
Sistem Pemberi Rekomendasi Artis Berdasarkan Jumlah Interaksi Menggunakan Metode Collaborative Filtering Yuliarta, Chara Maria Emmanuel; Wibowo, Agung Toto
eProceedings of Engineering Vol. 11 No. 4 (2024): Agustus 2024
Publisher : eProceedings of Engineering

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

Abstract

Sistem rekomendasi adalah suatu sistem penyaringan yang bertujuan untuk memprediksi preferensi yang diberikan oleh pengguna terhadap suatu elemen tertentu, pada penelitian ini terhadap sebuah artis. Penelitian ini memiliki tujuan untuk meningkatkan hasil performansi sistem rekomendasi artis menggunakan metode collaborative filtering. Metode collaborative filtering menggunakan informasi pengguna dan artis dalam membangun rekomendasi. Dataset yang digunakan mencakup jumlah pemutaran lagu oleh pengguna. Metode collaborative filtering diimplementasikan dengan melakukan perhitungan similarity antar pengguna dan antar artis. Perhitungan similarity yang digunakan, menggunakan cosine similarity. Setelah dilakukan perhitungan kesamaan, dilakukan perhitungan weighted sum dan menghasilkan prediksi. Evaluasi performansi dihitung menggunakan MAE (Mean Absolute Error), MSE (Mean Squared Error), dan RMSE (Root Mean Squared Error). Hasil evaluasi yang didapatkan pada penelitian ini adalah MAE 9,474, MSE 52.653,40 dan RMSE 229,208 pada perbandingan 70:30. Sedangkan pada perbandingan 75:25 menghasilkan MAE 9,902, MSE 45.914,85 dan RMSE 210,017. Pada perbandingan 80:20 hasil yang didapatkan adalah MAE 10,486, MSE 48.764,51 dan RMSE 217,416. Hasil tersebut menunjukkan bahwa, semakin besar rasio data train terhadap data test, nilai MAE, MSE dan RMSE cenderung meningkat. Kata kunci: Sistem Rekomendasi, Collaborative filtering, Cosine similarity, MAE, MSE, RMSE.