Rastri Prathivi
Universitas Semarang

Published : 10 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 10 Documents
Search

Performance Evaluation of Naive Bayes Algorithm for Classification of Fertilizer Types Rastri Prathivi; April Firman Daru; Sara Sharifzadeh
Telematika Vol 15, No 1: February (2022)
Publisher : Universitas Amikom Purwokerto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35671/telematika.v15i1.1410

Abstract

Determining the right fertilizer is very important to get optimal plant growth results. Each plant requires different nutrient requirements. Different soil types cause the soil's nutrient content and PH value to differ from one type to another. Regional conditions in a place will also cause the need for plant absorption of nutrient content to be more varied. By using the classification of the problems that have been mentioned, it can be solved by studying patterns from existing fertilizer use data into knowledge that can be used to determine decisions. In this study, modeling with the Naïve Bayes algorithm has been applied to the existing fertilizer use data where the probability value of each class has been calculated to get the highest probability value of a class. The measurement of the accuracy value of the modeling used is measured using the Split Validation method, where the training data will be divided into training data and testing data so that the accuracy value of the model is obtained. From the applied modeling, an accuracy value of 60% is obtained, which shows the level of accuracy of the model obtained from the classification results in the form of the name of the fertilizer, which is expected to help in determining the name of the fertilizer that needs to be used.
SISTEM PRESENSI KELAS MENGGUNAKAN PENGENALAN WAJAH DENGAN METODE HAAR CASCADE CLASSIFIER Prathivi, Rastri; Kurniawati, Yunita
Simetris: Jurnal Teknik Mesin, Elektro dan Ilmu Komputer Vol 11, No 1 (2020): JURNAL SIMETRIS VOLUME 11 NO 1 TAHUN 2020
Publisher : Universitas Muria Kudus

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (932.608 KB) | DOI: 10.24176/simet.v11i1.3754

Abstract

Kehadiran siswa di kelas merupakan hal penting saat kegiatan belajar mengajar di laksanakan. Sistem presensi yang masih menggunakan cara manual dengan memakai kertas. Memiliki permasalahan yang sering muncul antara lain: terjadinya manipulasi data kehadiran, hilangnya buku presensi, sulit dalam merekapitulasi data kehadiran. Sistem presensi dengan pengenalan wajah digunakan karena hanya memerlukan kamera untuk mengambil citra gambar. Dengan mengotomatiskan proses kehadiran akan lebih meningkatkan produktivitas guru kepada siswanya. Metode haar cascade classifier digunakan karena memiliki komputasi yang sangat cepat tergantung pada jumlah piksel dalam persegi bukan nilai piksel dari sebuah gambar. Dari hasil pendeksian wajah menggunakan metode haar cascade classifier prosentase yang telah dicapai sebesar 75%. Seluruh sistem terbukti dapat berjalan dengan baik dalam mendeteksi seluruh objek yang ada secara tepat. Sistem memudahkan dalam memantau kehadiran siswadi kelas secara akurat, efisien serta menghemat waktu serta tenaga
GAME SCORING SUPPORTING OBJECTS MENGGUNAKAN AGEN CERDAS BERBASIS ARTIFICIAL INTELLIGENCE Novita Putri, Astrid; Prathivi, Rastri
Jurnal Tr@nsForMat!ka Vol 13, No 2 (2016)
Publisher : Jurusan Teknologi Informasi Universitas Semarang

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

Abstract

Game are activity most structure, one that rdinary is done in fun and also education tool and help to develop practical skill, as training, education, simulation or psychological. On its developing current game have until 3D. In one game, include in First Person Shutter  necessary scoring  one that intent to motivate that player is more terpacu to solve game until all through,  on scoring  Super Marios game Boss, Compass does count scoring havent utilized Artifical Intelligent so so chanted, while player meet with supporting objects example ammor  ability really guns directly dead, so is so easy win. Therefore at needs a count scoring  interesting so more motivated in finishing problem Scoring accounting point for First Person Shutters game .This modelling as interesting daring in one game, since model scoring  one that effective gets to motivate that player is more terpacu in plays and keep player for back plays. Besides model scoring  can assign value that bound up with game zoom.On Research hits scoring this game will make scoring bases some criterion which is health Point, Attack point, Defence point, And  Magic  what do at have  supporting objects ,then in this research do compare two method are methodic statistic and Fuzzy. Result of this research 83,4 % on testings examination and on eventually gets to be concluded that fuzzys method in trouble finish time more long time but will player more challenging to railroad.  
Optimasi Model Transfer Learning Convolutional Neural Network Untuk Klasifikasi Citra CIFAR-10 Rastri Prathivi
Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Vol 4 No 4 (2020): Agustus 2020
Publisher : Ikatan Ahli Informatika Indonesia (IAII)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29207/resti.v4i4.2131

Abstract

The low accuracy when performing the image classification process is a problem that often occurs. The image classification process requires the completeness of the features of the image which form an informative image pattern so that information from the image can be displayed. The purpose of this study is to classify images in the CIFAR-10 image dataset using the CNN method. Initially the CNN method gave an accuracy of 79.4% but had a long computation time of 12 hours with 10,000 iterations. The optimization process for the CNN method is carried out by combining the CNN method, the PCA algorithm and the t-SNE algorithm. The algorithm is used to reduce the length of the image matrix in the initial transfer of learning without reducing the information in the image so that the classification process can be done correctly. The final result obtained from the optimization has an accuracy of 90.5%. With an optimization rate of 11%. The resulting time is more efficient, namely 3 hours for the feature transfer-value process and 6 minutes for the testing process with 10,000 iterations.
FEATURE RECOGNITION BERBASIS CORNER DETECTION DENGAN METODE FAST, SURF DAN FLANN TREE UNTUK IDENTIFIKASI LOGO PADA AUGMENTED REALITY MOBILE SYSTEM Rastri Prathivi
Jurnal Transformatika Vol 11, No 2 (2014): January 2014
Publisher : Jurusan Teknologi Informasi Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/transformatika.v11i2.96

Abstract

Logo is a graphical symbol that is the identity of an organization, institution, or company. Logo is generally used to introduce to the public the existence of an organization, institution, or company. Through the existence of an agency logo can be seen by the public. Feature recognition is one of the processes that exist within an augmented reality system. One of uses augmented reality is able to recognize the identity of the logo through a camera.The first step to make a process of feature recognition is through the corner detection. Incorporation of several method such as FAST, SURF, and FLANN TREE for the feature detection process based corner detection feature matching up process, will have the better ability to detect the presence of a logo. Additionally when running the feature extraction process there are several issues that arise as scale invariant feature and rotation invariant feature. In this study the research object in the form of logo to the priority to make the process of feature recognition. FAST, SURF, and FLANN TREE method will detection logo with scale invariant feature and rotation invariant feature conditions. Obtained from this study will demonstration the accuracy from FAST, SURF, and FLANN TREE methods to solve the scale invariant and rotation invariant feature problems.
Optimasi Algoritme Naive Bayes Untuk Klasifikasi Data Gempa Bumi di Indonesia Berdasarkan Hiposentrum Rastri Prathivi
Telematika Vol 13, No 1: Februari (2020)
Publisher : Universitas Amikom Purwokerto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35671/telematika.v13i1.928

Abstract

Abstract: The Hiposentrum or epicentre is the source of an earthquake which is at a certain depth on earth. The classification of earthquake powers based on the depth of Hiposentrum needed to examine the potential earthquake powers spread in Indonesian territory. The results of the classification process often experience problems, namely inaccuracy in classification. To solve that problem, then algorithms optimising classification must be increased. This research uses the Naïve Bayes algorithm, which is optimized using the Adaboost algorithm. Evaluation of the results of the optimized classification algorithm is needed to determine the level of accuracy using prescriptions and recall. In this study, the object of research is earthquake data in Indonesia which will be used as training data and testing data. The average accuracy of the Naïve Bayes algorithm is 72.3%, and the Naïve Bayes and Adaboost algorithm is 85.3%.Abstrak: Hiposentrum atau pusat gempa merupakan sumber gempa yang terdapat pada kedalaman tertentu di bumi. Klasifikasi kekuatan gempa berdasarkan kedalaman hiposentrum diperlukan untuk mengetahui potensi kekuatan gempa yang tersebar di wilayah Indonesia. Hasil dari proses klasifikasi seringkali mengalami masalah yaitu ketidaktepatan dalam klasifikasi. Untuk mengatasi masalah tersebut maka algoritme klasifikasi perlu ditingkatkan optimasinya. Penelitian ini menggunakan algoritme Naive Bayes yang dioptimasi menggunakan algoritme Adaboost. Evaluasi terhadap hasil dari algoritme klasifikasi yang telah dioptimasi diperlukan untuk mengetahui tingkat akurasi menggunakan presicion dan recall. Dalam penelitian ini objek penelitian berupa data gempa bumi di Indonesia yang akan digunakan sebagai data training  dan data testing. Hasil rata - rata akurasi algoritme Naïve Bayes sebesar 72,3% dan algoritme Naïve Bayes dan Adaboost sebesar 85,3%.
Klasifikasi Data Trafik Internet Menggunakan Metode Bayes Network (Studi Kasus Jaringan Internet Universitas Semarang) Rastri Prathivi
Jurnal Transformatika Vol 12, No 2 (2015): January 2015
Publisher : Jurusan Teknologi Informasi Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/transformatika.v12i2.81

Abstract

Pemakaian internet merupakan kebutuhan yang penting yang mendukung kinerja dan aktivitas di kampus. Bagian yang terpenting dari infrastrutur internet yang difasilitasi oleh kampus adalah tersedianya bandwidth yang cukup untuk kelancaran trafik data melalui internet. Metode klasifikasi menggunakan Bayes Network ini memanfaatkan metode klasifikasi yang dimiliki oleh data mining untuk diterapkan pada data trafik jaringan internet.Penelitian ini bertujuan untuk mengklasifikasi data pemakaian internet sehingga dari klasifikasi tersebut dapat diketahui destination network, protocol dan lebar bandwidth yang banyak diakses pada waktu tertentu.Data trafik internet diambil melalui software Wireshark. Sedangkan pengolahan data dan proses pengklasifikasian data trafik internet diolah dengan Weka
Pendekatan Model Predictive Control untuk Optimalisasi Biaya Penyediaan Vaksin dengan Reorder Point Huizen, Lenny Margaretta; Handayani, Titis; Prathivi, Rastri
JUSTIN (Jurnal Sistem dan Teknologi Informasi) Vol 9, No 4 (2021)
Publisher : Jurusan Informatika Universitas Tanjungpura

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (321.029 KB) | DOI: 10.26418/justin.v9i4.44888

Abstract

Pengendalian persediaan sangat penting bagi penyedia barang, tujuannya adalah agar terjadi keseimbangan antara persediaan dan permintaan. Pengelolaan persediaan pada rumah sakit sangat penting untuk diperhatikan agar kebutuhan kesehatan tetap terpenuhi. Salah satu kebutuhan yang harus terpenuhi adalah vaksin. Pengelolaan persediaan vaksin yang benar diperlukan agar terhindar dari kekurangan stok yang dapat menyebabkan aktifitas pelayanan menjadi terganggu dan sebaliknya jika suatu persediaan terlalu banyak dapat terjadi kelebihan biaya seperti biaya pembelian, penyimpanan serta pemeliharaan. Model Predictive Control dapat digunakan untuk menganalisa pengendalian stok berdasarkan biaya yang optimal karena Model Predictive Control merupakan sistem kendali dengan menggunakan hasil perhitungan prediksi dalam mengeluarkan kontrol input. Untuk mencari biaya yang minimal yaitu berdasarkan tingkat persediaan vaksin serta biaya vaksin sehingga dapat menjamin ketersediaan vaksin. Hasil yang didapat adalah biaya yang dikeluarkan untuk penyediaan vaksin dengan menggunakan pendekatan MPC sebesar Rp 61.886.332 berdasarkan jumlah pembelian vaksin 61 ampul lebih sedikit dibandingkan dengan rumah sakit sebesar Rp 64.007.000 atau sebesar pembelian 67 ampul. Hal ini tentu saja dapat mengurangi biaya untuk pengadaan vaksin. 
Optimasi Model Transfer Learning Convolutional Neural Network Untuk Klasifikasi Citra CIFAR-10 Rastri Prathivi
Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Vol 4 No 4 (2020): Agustus 2020
Publisher : Ikatan Ahli Informatika Indonesia (IAII)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29207/resti.v4i4.2131

Abstract

The low accuracy when performing the image classification process is a problem that often occurs. The image classification process requires the completeness of the features of the image which form an informative image pattern so that information from the image can be displayed. The purpose of this study is to classify images in the CIFAR-10 image dataset using the CNN method. Initially the CNN method gave an accuracy of 79.4% but had a long computation time of 12 hours with 10,000 iterations. The optimization process for the CNN method is carried out by combining the CNN method, the PCA algorithm and the t-SNE algorithm. The algorithm is used to reduce the length of the image matrix in the initial transfer of learning without reducing the information in the image so that the classification process can be done correctly. The final result obtained from the optimization has an accuracy of 90.5%. With an optimization rate of 11%. The resulting time is more efficient, namely 3 hours for the feature transfer-value process and 6 minutes for the testing process with 10,000 iterations.
Implementasi Algoritma Klasifikasi K-Nearest Neighbor (KNN) Untuk Klasifikasi Seleksi Penerima Beasiswa Saifur Rohman Cholil; Titis Handayani; Rastri Prathivi; Tria Ardianita
IJCIT (Indonesian Journal on Computer and Information Technology) Vol 6, No 2 (2021): IJCIT November 2021
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/ijcit.v6i2.10438

Abstract

AbstrakPemberian beasiswa kepada siswa Sekolah Menengah Atas (SMA) sudah umum dilakukan. Hal ini terjadi sejak adanya dana pendidikan 20% dari Kementrian Pendidikan dan Kebudayaan (Kemendikbud). Selain untuk batuan kepada siswa yang kurang mampu, beasiswa juga diberikan kepada siswa yang mempunyai prestasi akademik maupun prestasi non akademik. Pemberian beasiswa yang terjadi selama ini baik di SMA ataupun yang lain masih menggunakan perhitungan dan pengolahan data secara manual. Proses perhitungan secara manual memungkinkan adanya penerima  beasiswa  yang tidak tepat sasaran. Pengolahan penerimaan beasiswa bisa menggunakan sebuah algoritma data mining untuk mengklasifikasikan  calon penerima  beasiswa  berdasarkan  data yang diambil dari data siswa  penerima beasiswa sebelumnya (data training) dengan data yang diambil dari calon penerima beasiswa (data testing). Penelitian ini bertujuan membantu proses seleksi beasiswa di SMA menggunakan algoritma K-Nearest Neighbor (KNN) supaya penerima beasiswa tepat sasaran. Algoritma KNN bisa memberikan  kebutuhan data yang akurat  dan informasi yang diperlukan untuk menyeleksi  calon penerima beasiswa. Hasil dari penelitian ini adalah adalah terseleksinya 30 orang dari 89 data yang telah dilakukan klasifikasi.  Pengujian sistem menggunakan pengujian akurasi metode confusion matrix dengan hasil pengujian sebesar 90.5%. Hal ini menunjukkan bahwa algoritma KNN bisa digunakan untuk mengklasifikasikan seleksi penerimaan beasiswa.Kata Kunci: algoritma, beasiswa, data mining, KNNAbstractProviding scholarships to high school students (SMA) is common. This happened since there was a 20% education fund from the Ministry of Education and Culture (Kemendikbud). In addition to rocks to underprivileged students, scholarships are also given to students who have academic and non-academic achievements. Scholarships that have occurred so far both in high school and others still use manual calculation and data processing. The manual calculation process allows for scholarship recipients who are not on target. Processing scholarship receipts can use a data mining algorithm to classify prospective scholarship recipients based on data taken from previous scholarship recipient student data (training data) with data taken from prospective scholarship recipients (data testing). This study aims to help the scholarship selection process in high school using the K-Nearest Neighbor (KNN) algorithm so that scholarship recipients are on target. The KNN algorithm can provide accurate data and information needed to select prospective scholarship recipients. The result of this research is the selection of 30 people from 89 data that has been classified. System testing uses the accuracy of confusion matrix testing with 90.5% test results. This shows that the KNN algorithm can be used to classify scholarship acceptance selections.Keywords: algorithms, data mining, KNN, scholarship