Articles
<![CDATA[Identifikasi Tanaman Buah Tropika Berdasarkan Tekstur Permukaan Daun Menggunakan Jaringan Syaraf Tiruan ]]>
<![CDATA[Muhammad Asyhar Agmalaro]]>;
<![CDATA[Aziz Kustiyo]]>;
<![CDATA[Auriza Rahmad Akbar]]>
<![CDATA[ Jurnal Ilmu Komputer & Agri-Informatika Vol. 2 No. 2 (2013) ]]>
Publisher : <![CDATA[Departemen Ilmu Komputer - IPB University ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (683.921 KB)
|
DOI: 10.29244/jika.2.2.73-82
<![CDATA[Indonesia merupakan salah satu negara dengan keanekaragaman tanaman buah tropika yang cukup tinggi. Keanekaragaman tanaman buah tropika tersebut merupakan satu tantangan dalam melakukan identifikasi. Identifikasi tanaman dapat dilakukan berdasarkan buah, bunga, maupun daun. Identifikasi berdasarkan daun merupakan identifikasi yang lebih mudah dilakukan karena daun akan ada sepanjang masa, sedangkan bunga dan buah mungkin hanya ada pada waktu tertentu. Identifikasi tanaman menggunakan daun dapat dilakukan berdasarkan bentuk, tekstur, maupun warna citra daun tersebut. Pada penelitian ini, ekstraksi fitur gray level co-occurrence matrix (GLCM) dari tekstur citra permukaan daun buah tropika digunakan sebagai input dari pelatihan Jaringan syaraf tiruan untuk proses identifikasi. Secara keseluruhan, pengujian dengan menggunakan hidden neuron sebanyak 7 menghasilkan hasil akurasi terbaik, yaitu 90%. Kata kunci: buah tropika, daun, GLCM, jaringan syaraf tiruan, tekstur.]]>
<![CDATA[Identifikasi Daun Shorea menggunakan KNN dengan Ekstraksi Fitur 2DPCA ]]>
<![CDATA[Erni Yusniar]]>;
<![CDATA[Aziz Kustiyo]]>
<![CDATA[ Jurnal Ilmu Komputer & Agri-Informatika Vol. 3 No. 1 (2014) ]]>
Publisher : <![CDATA[Departemen Ilmu Komputer - IPB University ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (609.587 KB)
|
DOI: 10.29244/jika.3.1.18-26
<![CDATA[Shorea adalah jenis meranti yang memiliki nilai ekonomis yang tinggi. Shorea tergolong dalam famili Dipterocarpaceae yang memiliki 194 spesies yang tumbuh di daerah tropis. Shorea merupakan jenis yang sulit untuk diidentifikasi karena memiliki banyak kemiripan. Untuk mengatasi kesulitan tersebut, penelitian ini mengidentifikasi Shorea berdasarkan citra daun. Jumlah spesies yang digunakan penelitian ini adalah 10 jenis Shorea. Metode ekstraksi fitur yang digunakan adalah 2 dimensional principal component analysis (2D-PCA) dengan metode klasifikasi KNN. Penelitian ini memiliki 4 percobaan yang dibagi menjadi komponen R, G, B, dan grayscale. Hasil rata-rata akurasi terbaik sebesar 75% pada komponen G dengan kontribusi nilai eigen 85%.Kata kunci: 2 Dimensional Principal Component Analysis, K-Nearest Neighbour, Shorea]]>
<![CDATA[Analisis Pola Produktivitas Penulis Artikel Bidang Perpustakaan Dan Informasi Di Indonesia : Suatu Kajian Bibliometrika ]]>
<![CDATA[Agus Wahyudi]]>;
<![CDATA[Aziz Kustiyo]]>;
<![CDATA[Sulistyo Basuki]]>
<![CDATA[ Jurnal Pustakawan Indonesia Vol. 14 No. 2 (2015): Jurnal Pustakawan Indonesia ]]>
Publisher : <![CDATA[Perpustakaan IPB ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (747.043 KB)
|
DOI: 10.29244/jpi.14.2.%p
<![CDATA[This study examines the productivity of the authors in the field of library and information science in Indonesia during the year of 2001-2010 by using Lotka’s law. The population of this study are all kinds of scientific articles contained in the journal library and information science in which the journal is registered in PDII-LIPI databases and published between the years 2001 to 2010. Assessment author's participation was done by using 'straight count'. The sampling technique used is saturated sample. Journals in accordance with the criteria of research as much as 24, 1085 articles written by 1018 authors. Due to this research used 'straight count' only 547 authors counted. Obtained the results of the calculation of the value of n worth 1,92 and C 0,6172. The finding conclude, in the year 2001-2010 the number of certain writers that contribute one article is 61.72% of the total number of authors. Test results showed that the value Dmaks smaller than the critical value, this means that the productivity distribution author library and information science in Indonesia year of 2001-2010 in accordance with the argument of Lotka’s law. It is known that the productive author works as a lecturer/librarians derived from the college environment.Keywords: Library and Information Science, Bibliometrics, Lotka’s Law]]>
<![CDATA[Evaluasi Penggunaan Layanan Koleksi E-Resources Menggunakan Standar Indikator Kinerja (ISO 11620:2014) Di Perpustakaan Nasional RI ]]>
<![CDATA[Indreswari Nurmalia]]>;
<![CDATA[Aziz Kustiyo]]>;
<![CDATA[Sulistyo Basuki]]>
<![CDATA[ Jurnal Pustakawan Indonesia Vol. 15 No. 1-2 (2016): Jurnal Pustakawan Indonesia ]]>
Publisher : <![CDATA[Perpustakaan IPB ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (717.884 KB)
|
DOI: 10.29244/jpi.15.1-2.%p
<![CDATA[One of library’s most important element of the library is collection. The rise information and technology, has improved library needs to transforms into digital era. Library users preferred e-resources collections. The National Library of Indonesia (NLI) realized e-resources collection has become one of the primary collection. During these five years, there are not enough research about usage of e-resources collection services evaluation. This research took evaluation of the usage of e-resources collection services with a quantitative approach of the ISO 11620:2014 as general standard for library performance indicators. ISO 11620:2014 is a statement of symbol, numeric, and verbal that is obtained through the library statistics and data that is used to characterized the library performance indicators. There are 6 (six) performance indicators that are used as benchmark in assesing how far e-resources collection is used in The NLI on 2014-2015. Those 6 (six) indicators are : (1) The percentage of e-resources collection that is not used); (2) the number of content unit downloaded per capita; (3) the number of visitors that join the e-resources training; (4) the expense of the e-resources collection procurement; (5) the percentage of expenditure on the provision of information for the collection of e-resources; (6) Percentage of library staff who provide the guidance of the e-resources collection services for. This research found that e-resources usage collections services in The NLI is not optimized. E-resources collections service has not maximized for following reasons. First, The NLI’s e-resources is in balanced referring to covered subjects. The high e-resources collections that were not used by user, only 3% of the whole e-resources collections are used by users. Although, the level of content downloaded unit per capita for each e-resources collections decreased from the previous year. E-journals downloaded are more higher than the e-books and e-videos downloaded from whole e-resources collections. In 2015, the number e-books and e-video downloaded are less than 5 per 1000 user. PROQUEST download decreased from 2.7 to 1.9 for all users served. These conditions was caused by lack of e-resources development policy that becomes controlling in the process of collection management at the National Library of Indonesia. NLI needs to improve the e-resources development a primary missions to create a great form of national collections. Second, NLI need to set priority in providing technology based services. E-resources training and promotion are also need to set up as priority to colleges that have limited infrastructure and internet access. There are a large number user can be a potential gateway to increasing e-resources usage. This research found users were trained at the NLI’s e-resources has decreased in 2015. There was about 3 users per 1000 the NLI’s users in the previous year. Third, the level of the percentage of staff who provide training increased by 12.34% from the previous year. Although there was improvement the percentage of the GCC staff training is still very small compared with the number of staff at service center. Librarian’s competencies also needed to improve e-resources training in NLI. Librarian must have competenciies as much skill in providing information services in technology, has skill in searching strategy using e-resources. The research also found evaluation of e-resources usage at NLI will increase. Fourth, in terms of budget indicators issued for the provision of information in the form of a collection of e-resources shows the results are quite positive. Costs incurred for the provision of e-resources collections always increase every year. The research recommended further research related to the application of e-resources system performance measurement standards. It is important for the evaluation of performance of e-resources development at NLI to improve performance even better in the future.Key words: e-resources, usage evaluation, National Library of Indonesia (NLI), ISO 11620]]>
<![CDATA[Diagnosis Penyakit Demam Berdarah Dengue Menggunakan Voting Feature Intervals 5 ]]>
<![CDATA[Irman Hermadi]]>;
<![CDATA[Aziz Kustiyo]]>;
<![CDATA[Aristi Imka Apniasari]]>
<![CDATA[ KOMPUTASI Vol 5, No 1 (2008): Komputasi: Jurnal Ilmiah Ilmu Komputer dan Matematika ]]>
Publisher : <![CDATA[Ilmu Komputer, FMIPA, Universitas Pakuan ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (1247.724 KB)
|
DOI: 10.33751/komputasi.v5i9.1425
<![CDATA[Tingkat kematian akibat penyakit Demam Berdarah Dengue relatif masih tinggi. Salah satu penyebab tingginya tingkat kematian tersebut adalah keterlambatan diagnosis. Semakin cepat diagnosis dapat dilakukan, semakin cepat pula pertolongan bisa diberikan sehingga dapat mengurangi angka kematian tersebut. Penelitan ini akan menerapkan algoritma Voting Feature Intervals 5 (VF15) untuk mendiagnosa penyakit DBD. Pada pertelitian ini dilakukan 4 tahap pengujian. Rata-rata akurasi yang dihasilkan pada pengujian tahap pertama terhadap data sebelum validasi adalah 65,66%. Rata-rata akurasi pada pengujian tahap kedua ini adalah 92,86%. Rata-rata akurasi yang dihasilkan pada pengujian tahap ketiga ini mencapai 97,62%. Selanjutnya, pada pengujian tahap keempat akurasi yang dihasilkan untuk data setelah validasi adalah 100%. Akurasi tersebut jauh lebih tinggi bila dibandingkan dengan penelitian yang telah dilakukan oleh Syafii pada tahun 2006 dengan menggunakan model ANFIS yang hanya mencapai 86,67%. Kata Kunci : Demam Berdarah Dengue, Diagnosis, Voting Feature Intervals.]]>
<![CDATA[PENGARUH INCOMPLETE DATA TERHADAP AKURASI VOTING FEATURE INTERVALs-5 (VFI5) ]]>
<![CDATA[Aziz Kustiyo]]>;
<![CDATA[Agus Buono]]>;
<![CDATA[Atik Pawestri Sulistyo]]>
<![CDATA[ KOMPUTASI Vol 4, No 8 (2007): Vol. 4, No. 8, Juli 2007 ]]>
Publisher : <![CDATA[Ilmu Komputer, FMIPA, Universitas Pakuan ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (1202.648 KB)
|
DOI: 10.33751/komputasi.v4i8.1783
<![CDATA[Permasalahan mengenai data hilangan merupakan masalah umum yang tejadi pada lingkungan medis. Data hilangan dapat disebabkan beberapa hal yaitu salah memasukan data, data nya tidak valid dan peralatan yang di gunakan untuk mengambil data tidak berfungsi dengan baik. Voting Feature Intervals merupakan algoritma klasifikasi yang di kembangkan oleh Gulsen Demiroz dan H.Altay Guvenir pada tahun 1997. Algoritma ini dapat mengatasi data hilang dengan mengabaikan data hilang tersebut . Pada penelitian ini dilakukan penerapan algoritma Voting Feature Intervals-5 (VFI5) sebagai algoritma klasifikasi pada kasus data hilang. Data yang di gunakan adalah data ordinal (data Dermatology) dan data interval (data lonosphere). Untuk mengatasi data hilang di gunakan tiga metode yaitu mengabaikan data hilang dengan mean atau modus. Rata-rata tingkat akurasi data ordinal tertinggi sebesar 93.81% dan Rata-rata tingkat interval tertinggi sebesar 79.89%. Hasil penelitian menunjukan rata-rata tingkat akurasi yang tertinggi dicapai ketika data hilang dengan mean atau modus. ]]>
<![CDATA[Prediksi Kandungan Lignin pada Dedak Padi Bercampur Sekam Menggunakan Tekstur Statistik dan KNN ]]>
<![CDATA[Eylen Desy Novita]]>;
<![CDATA[Aziz Kustiyo]]>;
<![CDATA[Anuraga Jayanegara]]>;
<![CDATA[Toto Haryanto]]>;
<![CDATA[Hari Agung Adrianto]]>
<![CDATA[ Jurnal Ilmu Komputer dan Agri-Informatika Vol 9 No 1 (2022) ]]>
Publisher : <![CDATA[Departemen Ilmu Komputer, Institut Pertanian Bogor ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.29244/jika.9.1.58-69
<![CDATA[Adulteration in rice bran happens quite high due to the expensive price of rice bran. Mixing the rice bran with husk could decrease the rice bran quality because the content of crude fiber and lignin cointained in husk are anti-nutrients. Lignin content can be estimated by the texture of rice bran mixed with husk image. This study aimed to analyze the texture of rice bran mixed with husk image using run length feature extraction method with k-nearest neighbour (KNN) classification. The images of rice bran mixed with husk were taken using Dino Capture digital microscope with magnification 200 times. The images were generated with the spatial resolution of 640×480 pixels in a bitmap format. Those images were converted from RGB into grayscale in preprocessing phase, then the result of grayscale images were enhanced using histogram equalization as image enhancement method. The training and testing was determined using 5-fold cross validation with 3 repetition. The result of KNN classification with 7 features showed the highest accuracy of 74.55%.]]>
<![CDATA[Model Recurent Neural Network untuk Peramalan Produksi Tebu Nasional ]]>
<![CDATA[Aziz Kustiyo]]>;
<![CDATA[Mukhlis Mukhlis]]>;
<![CDATA[Aries Suharso]]>
<![CDATA[ Bahasa Indonesia Vol 9 No 1 (2022): Bina Insani ICT Journal (Juni) 2022 ]]>
Publisher : <![CDATA[Lembaga Penelitian dan Pengabdian Masyarakat Universitas Bina Insani ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.51211/biict.v9i1.1744
<![CDATA[Abstrak: Produksi tebu di Indonesia tersebar di beberapa wilayah yang mengakibatkan variabilitas yang tinggi dari variabel-variabel yang mempengaruhi produksi tebu nasional. Di samping itu, tidak mudah untuk mendapatkan data-data tersebut dalam waktu yang cukup panjang. Oleh karena itu peramalan produksi tebu nasional berdasarkan variabel-variabel tersebut sangat sulit dilakukan. Sebagai solusi dari masalah tersebut, maka peramalan produksi tebu nasional dilakukan berdasarkan data historisnya. Penelitian ini bertujuan untuk mengembangkan model recurrent neural networks (RNN) untuk peramalan produksi tebu nasional berdasarkan data historisnya. Data yang digunakan adalah data produksi tebu nasional dari tahun 1967 sampai dengan tahun 2019 dalam satuan ton. Sebagai data latih digunakan data tahun 1967 sampai dengan tahun 2006 dan sisanya dipakai sebagai data uji. Pada penelitian ini dilakukan percobaan untuk mengetahui pengaruh panjang deret waktu dan ukuran batch terhadap kinerja model RNN dengan tiga ulangan. Hasil penelitian menunjukkan bahwa model RNN dengan panjang deret waktu 4 dan ukuran batch 16 menghasilkan nilai mean absolut percentage error (MAPE) sebesar 9.0% dengan nilai korelasi 0.77. Secara umum, model RNN yang dibangun mampu menangkap pola produksi tebu nasional dengan tingkat kesalahan yang masih dapat ditoleransi. Kata kunci: deret waktu, peramalan, produksi tebu, recurrent neural networks Abstract: Sugarcane production in Indonesia is spread over several regions. This condition results in high variability of the variables that affect national sugarcane production. In addition, it is not easy to obtain these data over a long period. As a result, it is very difficult to forecast the production of national sugarcane based on the influencing variables. Therefore, the forecasting was based on historical data of the national sugarcane production. This study aims to develop a recurrent neural networks (RNN) model for forecasting national sugarcane production based on historical data. The data used is national sugarcane production data from 1967 to 2019 in tons. As training data, data from 1967 to 2006 were used and the rest was used as test data. In this study, an experiment was conducted to determine the effect of time series length and batch size on the performance of the RNN model with three replications. The results showed that the RNN model with a time series length of 4 and a batch size of 16 produced a mean absolute percentage error (MAPE) of 9.0% with a correlation value of 0.77. In general, the RNN model is able to capture the national sugarcane production pattern with a tolerable error rate. Keywords: forecasting, recurrent neural networks, sugarcane production, time series]]>
<![CDATA[Pengembangan Model Bayesian Regularization Backpropagation untuk Estimasi Nilai Nutrisi berdasarkan Komposisi Kimia Pakan Ternak Ruminansia ]]>
<![CDATA[Ulfa Nikmatya]]>;
<![CDATA[Aziz Kustiyo]]>;
<![CDATA[Anuraga Jayanegara]]>
<![CDATA[ Jurnal Ilmu Komputer dan Agri-Informatika Vol 9 No 2 (2022) ]]>
Publisher : <![CDATA[Departemen Ilmu Komputer, Institut Pertanian Bogor ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.29244/jika.9.2.168-176
<![CDATA[Perbedaan komponen kimia pakan ternak dapat memengaruhi nilai nutrisi hewan ternak ruminansia. Untuk menentukan komposisi kimia dan nutrisi yang dihasilkan oleh pakan ternak tersebut perlu dilakukan analisis di laboratorium. Sebagai alternatif, pada penelitian ini estimasi nutrisi pakan ruminansia berdasarkan komposisi kimia pakan dilakukan menggunakan bayesian regularization backpropagation menggunakan data sekunder. Data yang digunakan dalam penelitian ini diperoleh dari hasil penelitian Rowett Research Institute Prancis pada kategori main constituents dan ruminant nutritive values. Main constituents menunjukkan komposisi kimia pakan ruminansia sedangkan ruminant nutritive values menunjukkan nilai nutrisi pakan yang akan diprediksi. Model bayesian regularization backpropagation yang dibangun memiliki 12 neuron input yang berasal dari 12 komponen kimia pakan ruminansia. Jumlah maksimal output model tersebut adalah 8 neuron yang merupakan 8 nilai nutrisi pakan ruminansia. Proses pelatihan dilakukan dengan metode validasi silang dengan memvariasikan jumlah neuron lapisan tersembunyi dari 5 sampai dengan 50 dan jumlah neuron output sebanyak 8, 6 dan 3. Hasil percobaan menunjukkan model bayesian regularization backpropagation terbaik adalah model dengan 8 output dengan nilai root mean square error sebesar 3.47 dan nilai mean absolute percentage error sebesar 11.82%.]]>
<![CDATA[Ekstraksi Fitur Rantai Markov untuk Klasifikasi Famili Protein ]]>
<![CDATA[Toto Haryanto]]>;
<![CDATA[Rizky Kurniawan]]>;
<![CDATA[Sony Muhammad]]>;
<![CDATA[Aziz Kustiyo]]>;
<![CDATA[Endang Purnama Giri]]>
<![CDATA[ Jurnal Ilmiah SINUS Vol 21, No 2 (2023): Volume 21 No. 2 Juli 2023 ]]>
Publisher : <![CDATA[STMIK Sinar Nusantara ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.30646/sinus.v21i2.748
<![CDATA[As complex molecules, proteins have various roles for living things. Proteins are organic molecules formed from twenty amino acid combinations with various functions for living things, such as transportation systems, a catalyst of chemical reactions for metabolism, and food reserves. This research aims to classify proteins family based on sequences of amino acids as the primary structure. There are 300 amino acid fragments obtained from the Pfam database. The proteins family database subset with three sub-sample classes was obtained, including 1-cysPrx_C, 4HBT, and ABC_Tran. In this research, the first and second order of the Markov chain for extracting features were applied. Moreover, we use a Probabilistic Neural Network (PNN) as a classifier compared to the joint probability technique with Markov assumptions. We evaluate the results by comparing the sensitivity and specificity of both classification techniques. The evaluation results show that overall, PNN has slightly better performance than the joint probability technique for classifying protein families.]]>
