Articles

Found 16 Documents
Search

SEGMENTASI DAN PERHITUNGAN SEL DARAH PUTIH MENGGUNAKAN OPERASI MORFOLOGI DAN TRANSFORMASI WATERSHED Alham, Dwi Syamsuifin; Herumurti, Darlis
INFORMAL: Informatics Journal Vol 4 No 2 (2019): INFORMAL - Informatics Journal
Publisher : Faculty of Computer Science, University of Jember

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.19184/isj.v4i2.13347

Abstract

Perhitungan sel darah putih terkait dengan sistem kekebalan tubuh manusia karena bisa memberikan informasi tentang kondisi tubuh manusia seperti diagnosis penyakit, salah satunya leukimia. Penyakit ini ditandai dengan produksi sel darah putih berlebih yang menyebabkan fungsi normal darah menjadi terganggu dan dapat menyebabkan kematian. Untuk itu diperlukan deteksi dini penyakit ini, yang salah satunya dengan menganalisis populasi atau menghitung jumlah sel darah putih pada citra mikroskopis sel darah. Perlu dilakukan metode segmentasi yang tepat agar hasil perhitungan sel dapat maksimal. Metode Watershed merupakan metode yang dapat memisahkan objek yang saling berhimpit, tetapi memiliki kelemahan jika dalam suatu citra terdapat noise atau distribusi intensitas yang tidak merata. Oleh karena itu, untuk mengatasi masalah tersebut, maka ditambahkan operasi morfologi di awal untuk mengatasi adanya obyek yang tidak termasuk sel darah putih. Tahap terakhir adalah menghitung sel darah putih menggunakan standar objek yang dianggap sel darah putih normal berdasarkan ukuran roundness obyek. Pengujian dilakukan dengan menggunakan 35 citra dengan menggunakan nilai SE = 20 pada operasi morfologi opening. Hasilnya, akurasi perhitungan jumlah sel antara sistem dengan manual (ground truth) sebesar 97.67%.
PENJADWALAN MATAKULIAH DENGAN MENGGUNAKAN ALGORITMA GENETIKA DAN METODE CONSTRAINT SATISFACTION Buliali, Joko Lianto; Herumurti, Darlis; Wiriapradja, Giri
JUTI: Jurnal Ilmiah Teknologi Informasi Vol 7, No 1, Januari 2008
Publisher : Informatics, ITS

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (2047.249 KB) | DOI: 10.12962/j24068535.v7i1.a59

Abstract

Course scheduling problem has gained attention from many researchers. A number of methods have been produced to get optimum schedule. Classical definition of course scheduling cannot fulfill the special needs of lecture scheduling in universities, therefore several additional rules have to be added to this problem. Lecture scheduling is computationally NP-hard problem, therefore a number of researches apply heuristic methods to do automation to this problem. This research applied Genetic Algorithm combined with Constraint Satisfaction Problem, with chromosomes generated by Genetic Algorithm processed by Constraint Satisfaction Problem. By using this combination, constraints in lecture scheduling that must be fulfilled can be guaranteed not violated. This will make heuristic process in Genetic Algorithm focused and make the entire process more efficient. The case study is the case in Informatics Department, Faculty of Information Technology, ITS. From the analysis of testing results, it is concluded that the system can handle specific requested time slot for a lecture, that the system can process all the offered lectures, and that the system can produce schedules without violating the given constraints. It is also seen that Genetic Algorithm in the system has done optimation in finding the minimum student waiting time between lectures.
FEATURE SELECTION METHODS BASED ON MUTUAL INFORMATION FOR CLASSIFYING HETEROGENEOUS FEATURES Pawening, Ratri Enggar; Darmawan, Tio; Bintana, Rizqa Raaiqa; Arifin, Agus Zainal; Herumurti, Darlis
Jurnal Ilmu Komputer dan Informasi Vol 9, No 2 (2016): Jurnal Ilmu Komputer dan Informasi (Journal of Computer Science and Information)
Publisher : Faculty of Computer Science - Universitas Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (283.816 KB) | DOI: 10.21609/jiki.v9i2.384

Abstract

Datasets with heterogeneous features can affect feature selection results that are not appropriate because it is difficult to evaluate heterogeneous features concurrently. Feature transformation (FT) is another way to handle heterogeneous features subset selection. The results of transformation from non-numerical into numerical features may produce redundancy to the original numerical features. In this paper, we propose a method to select feature subset based on mutual information (MI) for classifying heterogeneous features. We use unsupervised feature transformation (UFT) methods and joint mutual information maximation (JMIM) methods. UFT methods is used to transform non-numerical features into numerical features. JMIM methods is used to select feature subset with a consideration of the class label. The transformed and the original features are combined entirely, then determine features subset by using JMIM methods, and classify them using support vector machine (SVM) algorithm. The classification accuracy are measured for any number of selected feature subset and compared between UFT-JMIM methods and Dummy-JMIM methods. The average classification accuracy for all experiments in this study that can be achieved by UFT-JMIM methods is about 84.47% and Dummy-JMIM methods is about 84.24%. This result shows that UFT-JMIM methods can minimize information loss between transformed and original features, and select feature subset to avoid redundant and irrelevant features.
LEAST SQUARES SUPPORT VECTOR MACHINES PARAMETER OPTIMIZATION BASED ON IMPROVED ANT COLONY ALGORITHM FOR HEPATITIS DIAGNOSIS Husain, Nursuci Putri; Arisa, Nursanti Novi; Rahayu, Putri Nur; Arifin, Agus Zainal; Herumurti, Darlis
Jurnal Ilmu Komputer dan Informasi Vol 10, No 1 (2017): Jurnal Ilmu Komputer dan Informasi (Journal of Computer Science and Information
Publisher : Faculty of Computer Science - Universitas Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (205.767 KB) | DOI: 10.21609/jiki.v10i1.428

Abstract

Many kinds of classification method are able to diagnose a patient who suffered Hepatitis disease. One of classification methods that can be used was Least Squares Support Vector Machines (LSSVM). There are two parameters that very influence to improve the classification accuracy on LSSVM, they are kernel parameter and regularization parameter. Determining the optimal parameters must be considered to obtain a high classification accuracy on LSSVM. This paper proposed an optimization method based on Improved Ant Colony Algorithm (IACA) in determining the optimal parameters of LSSVM for diagnosing Hepatitis disease. IACA create a storage solution to keep the whole route of the ants. The solutions that have been stored were the value of the parameter LSSVM. There are three main stages in this study. Firstly, the dimension of Hepatitis dataset will be reduced by Local Fisher Discriminant Analysis (LFDA). Secondly, search the optimal parameter LSSVM with IACA optimization using the data training, And the last, classify the data testing using optimal parameters of LSSVM. Experimental results have demonstrated that the proposed method produces high accuracy value (93.7%) for  the 80-20% training-testing partition.
Optimasi Naive Bayes Dengan Pemilihan Fitur Dan Pembobotan Gain Ratio Socrates, I Guna Adi; Akbar, Afrizal Laksita; Akbar, Mohammad Sonhaji; Arifin, Agus Zainal; Herumurti, Darlis
Lontar Komputer : Jurnal Ilmiah Teknologi Informasi Vol. 7, No. 1 April 2016
Publisher : Research institutions and Community Service, University of Udayana

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (533.93 KB) | DOI: 10.24843/LKJITI.2016.v07.i01.p03

Abstract

Naïve Bayes is one of data mining methods that are commonly used in text-based document classification. The advantage of this method is a simple algorithm with low computation complexity. However, there is weaknesses on Naïve Bayes methods where independence of Naïve Bayes features can’t be always implemented that would affect the accuracy of the calculation. Therefore, Naïve Bayes methods need to be optimized by assigning weights using Gain Ratio on its features. However, assigning weights on Naïve Bayes’s features cause problems in calculating the probability of each document which is caused by there are many features in the document that not represent the tested class. Therefore, the weighting Naïve Bayes is still not optimal. This paper proposes optimization of Naïve Bayes method using weighted by Gain Ratio and feature selection method in the case of text classification. Results of this study pointed-out that Naïve Bayes optimization using feature selection and weighting produces accuracy of 94%.
Efektifitas Aturan Main Untuk Game Edukasi Kosakata Bahasa Arab Berbasis Mobile Sani, Dian; Herumurti, Darlis; Kuswardayan, Imam
INTEGER: Journal of Information Technology Vol 2, No 2 (2017): September 2017
Publisher : INTEGER: Journal of Information Technology

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1376.256 KB)

Abstract

The game is ways to remove saturation from various kinds of affairs. The game also used as learning media. This educational game is usually used to invite users to play while learning. Languages and games must be complementary. Because of the language is important, it needs an effective and efficient way to increase interest in language learning. So the focus of this research is to create three mobile gaming applications with educational content of Arabic vocabulary which gameplay are different and analyze which applications are effective for users in recognizing Arabic vocabulary. Testing is done to find out if the application is effective for players to know and learn Arabic language. Samples taken are second grade students Madrasah Ibtidaiyyah who know the hijaiyah letters. Pre-test, treatment and post-test are part of the testing phase. Test results were analyzed using ANOVA one way method. The results of this study indicate that there is an increase in learning ability of participants through the medium of Arabic educational games with a value of 3.65 beyond the critical value with a significance level of 5% or 0.05 of 3.10. When Hypothesis 0 is rejected, then the comparison test between groups (games) with Scheffe method with critical value 2.48. The result is a third game with a value of 2.57 is said to be effective than the first game (0.58) and the second (2.00). The second game (2.05) is said to be more effective than the first game (0.51).Keywords: Educational Game, Game Effectiveness, Mobile games, Arabic vocabulary, ANOVA.
Analisis Pengaruh Penggunaan Game Edukasi pada Penguasaan Kosakata Bahasa Asing dengan Studi Kasus Game Edukasi Bahasa Arab Khairy, Muhammad Shulhan; Herumurti, Darlis; Kuswardayan, Imam
Khazanah Informatika Vol. 2 No. 2 Desember 2016
Publisher : Universitas Muhammadiyah Surakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23917/khif.v2i2.2137

Abstract

Pemanfaatan game saat ini telah merambah ke ranah edukasi, ditambah dengan berkembangnya teknologi saat ini, maka hal tersebut dapat dimanfaatkan untuk kepentingan edukasi. Pada penelitian ini akan dilakukan analisis pengaruh game edukasi pada kemampuan dalam menguasai kosakata bahasa asing, dengan studi kasus bahasa Arab. Game edukasi tersebut menggunakan perangkat bergerak dan salah satunya menggunakan teknologi realitas virtual dengan kakas Google Cardboard. Game edukasi diujikan pada pengguna berusia 10-15 tahun dan dibagi menjadi dua kelompok, berdasarkan teknologi yang digunakan dan genre game. Pengguna melakukan pre-test dan post-test? untuk mengukur kemampuan mereka sebelum dan sesudah mengujikan game. Hasil pengujian tersebut dianalisis dengan metode uji hipotesis ANOVA. Dari kedua kelompok tersebut didapatkan kesimpulan bahwa perbedaan teknologi tidak berpengaruh secara signifikan terhadap kemampuan pengguna. Begitu pula pada kelompok kedua, didapatkan kesimpulan bahwa faktor jenis game, faktor jenis kelamin pengguna, dan hubungan kedua faktor tersebut tidak berpengaruh secara signifikan terhadap perubahan kemampuan pengguna dalam menguasai perbendaharaan kosakata bahasa Arab.
Adaptive Non Playable Character in RPG Game Using Logarithmic Learning For Generalized Classifier Neural Network (L-GCNN) Mabruroh, Izza; Herumurti, Darlis
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Vol 4, No 2, May 2019
Publisher : Universitas Muhammadiyah Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (286.645 KB) | DOI: 10.22219/kinetik.v4i2.755

Abstract

Non-playable Character (NPC) is one of the important characters in the game. An autonomous and adaptive NPC can adjust actions with player actions and environmental conditions. To determine the actions of the NPC, the previous researchers used the Neural Network method but there were weaknesses, namely the action produced was not in accordance with the desired so the accuracy was not good. This study overcomes the problem of poor accuracy by using the Logarithmic Learning for Generalized Classifier Neural Network (L-GCNN) method with 6 input parameters, NPC health, distance from players, other NPCs involved, attack power, number of NPCs and NPC levels. While the output is to attack itself, attack in groups and move away. For testing, this study was tested on RPG games. From the results of the experiments conducted, it shows that the L-GCNN method has better accuracy than the 3 methods compared to 7% better than NN and SVM and 8% better than RBFNN because in the L-GCNN method there is an encapsulation process that is data have the same class will. Whereas the L-GCNN training time is 30% longer than the NN method because on L-GCNN one neuron consists of one data where there are fewer NNs in the hidden layer.
KOMBINASI FITUR BENTUK, WARNA DAN TEKSTUR UNTUK IDENTIFIKASI KESUBURAN TELUR AYAM KAMPUNG SEBELUM INKUBASI Dijaya, Rohman; Suciati, Nanik; Herumurti, Darlis
Jurnal Buana Informatika Vol 7, No 3 (2016): Jurnal Buana Informatika Volume 7 Nomor 3 Juli 2016
Publisher : Universitas Atma Jaya Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (428.984 KB) | DOI: 10.24002/jbi.v7i3.659

Abstract

Abstract. In the chicken nursery industry (doc) hatching efficiency is obtained by observing the eggs through candling before the incubation process. To sort out infertile eggs the use of fertility image identification thought egg candling is needed before incubation. The focus of this study is to combine the features of shape, texture and color to the area and egg yolk to determine the most dominant features in the image representing firtile egg candling. Features used in this study are the feature of forms: roundness, elongation, Index, Ellips Varriance and Circularity Ratio, moment invariant texture features of the area and the egg yolk, and features HSI color in egg yolks area. The test results show that the highest accuracy is on the features of the new forms of egg yolk with an accuracy of 76.67%. The second highest is shown by the combination of form features (Circularity Ratio, Ellips Varriance) and texture features in the area moment yolk color features HSI with 81.67% accuracy using SVM classification method.Keywords: Egg candling imagery, fertile, infertile, incubation Abstrak. Pada industri pembibitan ayam (doc) efisiensi penetasan telur ayam didapatkan dengan melakukan candling (peneropongan telur) sebelum proses inkubasi menggunakan mesin tetas. Untuk mengklasifikasikan telur fertile dan infertile dibutuhkan identifikasi kesuburan telur menggunakan citra candling sebelum inkubasi. Fokus dari penelitian ini adalah mengkombinasikan fitur bentuk, tekstur dan warna pada area kuning telur dan telur untuk mengetahui fitur yang paling dominan dalam merepresentasikan citra candling telur ayam kampung. Fitur yang digunakan dalam penelitian ini adalah fitur bentuk (Roundness, Elongation, Index, Ellips Varriance dan Circularity Ratio), fitur tektur moment invarian dari area telur dan kuning telur dan fitur warna HSI pada area kuning telur. Hasil pengujian menunjukkan akurasi tertinggi pada fitur bentuk kuning telur baru dengan akurasi 76,67% dan kombinasi fitur bentuk (Circularity Ratio, Ellips Varriance), fitur tekstur moment pada area kuning telur dengan fitur warna HSI dengan akurasi 81,67 % menggunakan metode klasifikasi SVM. Kata Kunci: Citra candling telur, fertile, infertile, inkubasi.
SISTEM REKOMENDASI INDEKS WEB DENGAN METODE FREQUENT TERMS BERBASIS MULTI INSTANCE LEARNING Herumurti, Darlis; Buliali, Joko Lianto; Andriana, Ria
Jurnal Informatika Vol 8, No 1 (2007): MAY 2007
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (152.253 KB) | DOI: 10.9744/informatika.8.1.pp. 10-17

Abstract

Web index page is well known as page that arranges information by giving the title and short explanation about the information, where the complete information will be presented in other page. However since the amount of information become accumulate, the existence of a lot of index page exactly cause difficulty on getting information because it is possible to direct users into a mount of irrelevant information. Without a system which can help user navigation, the process of seeking the expected information is equal to a trial and error processing. In this paper, web index recommendation system is investigated which involved the activity of user on accessing the index page. This system will arrange the frequent term in index page and then implement Multi Instance Learning to give recommendation of the new index page automatically. The algorithm is citation kNN that will be adapted into fretCit kNN by implementing the minimal Hausdorff distance in measuring the distance. The experiments show that from the several test of users, the system give performance in average recommendation until 82,41% accuracy with 66,71% recall. Abstract in Bahasa Indonesia : Halaman indeks dikenal sebagai halaman yang mengelompokkan informasi-informasi, dengan memberikan judul serta penjelasan singkat tentang suatu informasi, dimana informasi lengkap akan dipresentasikan pada halaman-halaman lain. Namun dengan ketersediaan informasi yang menjadi semakin menumpuk, keberadaan halaman indeks yang semakin banyak justru menyebabkan kesulitan dalam mendapatkan informasi karena mungkin akan mengarahkan pada banyak informasi yang tidak relevan. Tanpa adanya sebuah sistem yang dapat membantu navigasi user, untuk mencari informasi yang diinginkan sama saja dengan sebuah kegiatan trial dan error. Dalam penelitian ini, dirancang sebuah sistem rekomendasi indeks web yang melibatkan aktifitas user dalam mengakses halaman indeks. Sistem ini mengelompokkan frequent terms pada halaman indeks dan kemudian mengimplementasikan metode Multi Instance Learning untuk memberikan rekomendasi secara otomatis dari halaman-halaman indeks baru. Algoritma yang digunakan adalah algoritma Citation kNN yang diadaptasi menjadi fretCit-kNN dengan mengaplikasikan minimal Hausdorff distance dalam pengukuran jaraknya. Dalam hasil proses dan analisis disimpulkan bahwa dengan beberapa macam uji coba data dari beberapa user sistem menampilkan performa hingga rata-rata 82,41% akurasi dan nilai kembalian sebesar 66,71%. Kata kunci: halaman indeks, sistem rekomendasi, multi instance learning, citation kNN, hausdorff distance.