ICACSIS 2011

I have presented some papers on International Conference, but it is the first time I submit the paper for ICACSIS 2011. This conference is about Computer Science and Information System that committed by Universitas Indonesia. The conference will be held at Mercure Hotel, Ancol, North Jakarta, Indonesia. Unfortunately, at that day, I will attend the IELTS Test in PPB UGM, so I asked the second writer, Mr Haryono to present our paper. I hope he will be able to present our paper without any mistakes and there is no problem about the language.

Our research is about real time network monitoring using artificial neural network of our hotspot area. The committee of ICACSIS 2011 said that our research is has a novelty and interesting, but there are so many grammar error. The important critics is about cross validation that I did not mention in our paper. The cross validation is a testing method that use some groups. Every group must be tested after training by other group and vice versa.

If you want to join this conference as a participant, you can attend this on 17 – 18 December 2011. It costs Rp. 200.000,00 and I think it is not expensive compared to the experience that you will get there, as many researcher from overseas will come and present their interesting paper.

Update: 19/04/2017

The paper was indexed by IEEE (http://ieeexplore.ieee.org/document/6140744/) and Scopus (http://www.scopus.com/inward/authorDetails.url?authorID=55014574400&partnerID=MN8TOARS), with 2 citated (not bad 🙂 )

Competitive Research Grant Presentation

Dikti does not support the Young Lecturer Research grant anymore, but it still supports the Competitive research grant. It is difficult to get this grant because we have to present our research proposal first. On 5 December 2011, I presented my research proposal at Educational University of Indonesia (UPI) Bandung. My Research Proposal has a title “The evaluation and optimation of gas station building in Bekasi by using genetic algorithm”.

There were two group who would present the proposal: Main group and Guiding group. My university was member of guiding group that must be guided by other universities. Our group consists of 17 universities. Every leader of research presented for about 10 minutes before asked more about her/his research proposal by the interviewer, Mr Anton and Mrs Nenden.

After waiting for about 45 minutes, I presented my research proposal in front of other participants. I must reduced my power point because every presenter only needed 10 minutes. I must waited two other speakers before asked with so many questions. The first interviewer asked me whether my proposal was for competitive or fundamental grant. I was very surprised when listened that question. Fortunately, after little explanation about the content of my research proposal, he said that it was only wrong title, not wrong proposal. The second interviewer, Mrs Nenden, only asked simple and easy question about what research that I have done. I explained about my last research such as attendance system, engineering calculation, and so on. When all participants finished, I was disappointed because my watch showed that I miss my train to Yogyakarta, but I was very happy when my wife phoned me and said if our tickets were bought back by our national train department.

My Lecturer Certification

Lecturer is not only a job but also a profession like lawyer, teacher, doctor, and so on. It needs a certification too. Lecturer like a doctor, a university can not directly make a doctor, as after graduate a medical student must get a liscence from the government in order to be a doctor. After a student graduate from university he/she can give a lecturer at some universities, but in order to be said a lecturer he/she must fulfill a minimum point. In Indonesia, a lecturer must have S2 degree with functional level lektor. Of course with S2 and functional level Asisten Ahli, someone is allowed to give a lecture but the goverment will not give a lecturer allowance.

It is not difficult to get lecturer certification in Indonesia. You only just follow the rule and do some activity you must do. That activity called lecturer work load. For some lecturer candidate, they feel difficult because doing research is a point that must be fulfill. Moreover, they also must doing a community service by using their skill. After you got S2 degree and functional level lektor, you only propose to your faculty and university for getting lecturer certification. Because of budget limit, the government give a quota for every universities. Thank’s god, the government give a chance to me at batch II in 2011, so I only take nine years to get a lecturer certification.

Interested again in Pencils

After filling a portofolio form of lecturer certification, I feel free now. I took a trip around Yogyakarta, and after buying a lot of book at gramedia Yogyakarta, I try to enhance my writing skill in in English. It is about pencil, the oldest tool for modern writing. You have used it when studying at elementary school.

In nineteen century, the philosophers David Henry Thoreou developed pencil by his family manufacture. He separate pencil according to hardness and carbon contain in number 2, 3, 4, and so on. But the production stopped when England had found a big graphite resources which is a main component of pencils.

One advantage in using pencil is we can erase if we have an error on our writing. Of course we can use eraser for pen, but it leaves a stain in the paper, and sometime can break the paper. The pencil is appropriate for children who learn writing. It also has no dangerous liquid in ink as a mixture.

Now I always use pencil when taking an IELTS course at PPB UGM Yogyakarta. I do not know why I love pencil now. When I write something using a pencil, I can express my emotion and of course I use it for TOEFL and IELTS exam.

About Rp. 1 Billion/Person for Ph.D Dikti Scholarship

On 6-7th October 2011 some university in Australia gave presentation about their university at language education center of gadjah mada university (PPB UGM). The university are Curtin, Sidney, Adelaide and many more. After they had presented, the Dikti staff, Mrs Merry, gave an explanation about the scholarship. She explained why Dikti, the government institution which has a highest authority of high education, give 1,000 place for scholarship to study in Australia.

The English course at PPB UGM now is for second time (Batch – 2). From this course Dikti hope all of participants pass the IELTS score above 6.0, because some university require IELTS score 7.0. Fortunately, some subject (it my subject :)), only require minimum IELTS scor 6.0 and Mrs Merry, as an agent for scholarship in Australia, give dispensation if some one who participate in this course have IELTS score only 5.5, but he/she have to attand to 10 weeks IELTS course held by INSEARCH. Fortunately it’s free. At the end of speaking Mrs merry said that if you have finished your study by Dikti scholarship, you must go back to Indonesia, because one person for Ph.D degree, our government needs Rp. 1 Billion.

Checking Your Writing

Writing something in your blog will make your writing ability improve. For someone, writing is a difficult thing. How to make our writing easy to understand, interresting, coheren and grammatically right is the reason for someone to avoid writing. I thinks we must try to write somethings about everything for example, your own idea about hot news, your teaching’s modul or just write about your expression.

In order our writing skill improve, of course we must check our writing. In UK or USA, may be you must pay someone to check your writing, but I suggest you to check your writing by using information technology. You must be know google (www.translate.google.com), the famous sites, that can be help you correcting the wrong words. If you want to check your writing for errors or plagiarism, you should use another site for example http://www.grammarly.com

This site is very interesting for me as an IT lecturer because it include some soft computing algorithms. They can know weather our spelling, punctuation, style are good enough. It also know if our writing is a plagiarism. Here is the answer about my last writing, what a bad result !

Compiling the m-file

Compiling the m-file means we convert the source code of matlab into executable file, so the other computer can run it although there is no matlab software being installed. Of course if the computer which run the application has matlab software we don’t need to compile it.

Because there are a lot of file which build matlab-based application, we must prepare them first. Application with Graphic User Application (GUI) need at least two files: the fig-file for interface and m-file for source code. If you use some toolbox on it, you must include that files to your deployment project. For example, if you create fuzzy-based application, you mast include the fis-file to your deployment project.

Let’s open your matlab software and I suggest you to change the current directory into the same directory as your programming file. Matlab provide project for compiling by clicking the file, new and deployment project in the menu bar. This figure shows you deployment project window.

Please try to compile your own matlab application. As an information, matlab use Matlab Compile to C (mcc) to convert code in matlab into c++ first before converting to exe.

Artificial Neural Network Tutorial for Matlab (Presentation)

Artificial Neural Network (ANN) is my favorit algoritm because it use analogy of biological neural system. It is very interesting, neurons and axons convert to algorithm of transfer function and weight that can be adjust by activity called “learning”. When doing learning, ANN use some iteration to match the target with the input by changing it weight and some correction number, called bias.

Classification, Pattern Matching, signature identification, and other similar usage usually use ANN with some modification on input data and target. Because ANN has an ability to store memory on it neurons, some vendor use it for Pattern Matching application. There are some problem when using ANN, because when training, a lot of memory must be available if you don’t want your system being “hang”. The learning algorithm usually use the famous one, called backpropagation (from John Hopfield invention). But some research add other soft computing algorithms to enhanced the speed of learning and accuracy.

The language for creating ANN may be C++, Java or the easiest one, Matlab. I use Matlab for most of my application. Of course there is a classic problem by using Matlab, the lisence. But some language, for example, Octave, Scilab, etc, are open source and free for download. And the most important is the same syntax with Matlab. Of course we must build our function of ANN in open source language, because a lot of function (called M-file) does not available. If you want to know more about ANN with Matlab you can see www.mathworks.com or other millist, for example www.techsource.com.sg and free for sign up. This is a presentation form of explanation about ANN using Matlab you can download from techsource. Please find by your self in www.google.com.

Ok .. if you could not find, this is the link for download from my own storage: http://www.ziddu.com/download/16472066/neuralnetworkinmatlabtutorialpresentation.rar.html

Security vs Educating User

Security is main concern for microsoft coorporation. Since windows Xp sp 2, it included firewalls on its product, even though in sp 1 windows had used firewalls even must be set manually. After launch windows 7, microsoft introduce anti virus which have a life time lisence. But microsoft sometimes could not educate user well, for example in windows 7 there is suggestion when entering the hot spot area to push the button on the router. Can you imagine if your staff do this at your office?

Perkembangan Disain

Berbicara mengenai perencanaan produk tidak lepas dari kebutuhan dan keinginan pasar. Sesungguhnya kita tidak bisa secara tepat meramalkan keinginan pasar, tetapi pergerakan dari perubahan keinginan pasar dapat dilihat gejalanya. Walaupun secara teknis suatu produk sangat baik, tetapi jika tidak memiliki disain yang kurang diminati akan gagal di pasaran. Tentu saja jika gagal di pasaran, produk tersebut dikatakan gagal.

Sebagai studi kasus kita lihat motor honda yang di tahun 80-an sangat terkenal. Bahkan di tahun itu orang sudah menyebut naik motor dengan naik honda karena brand honda yang sudah merakyat. Coba bayangkan jika honda tidak melakukan riset pasar terhadap bentuk disainnya yang baru, tentu saja konsumen akan beralih ke merek baru karena takut dikatakan naik motor babeh.

Seiring dengan perkembangan teknologi dan kondisi jalan di Indonesia, perlu untuk merancang mesin dengan kapasitas besar. Akhirnya muncul motor-motor dengan cc besar seperti honda megapro dengan cc 145 dan 160. Honda mega pro di atas hasil modifikasi buntut yang lama yang banyak diprotes karena berbentuk seperti perahu.

Seiring dengan perkembangan jaman, perhatikan bentuk honda tiger 2000 yang mengikuti perkembangan jaman dari velg racing (walaupun ada yg suka model klasik velg jari-jari) dengan tutup tangki tambahan sebagai asesoris. Di awal tahun 2000 motor ini sempat merajai jalan.

Dan ternyata bentuk tedeng tambahan di tangki (lihat gambar di atas diambil dari dapurpacu.com .. gambar cewek cuma penyedap aja lho ..) ternyata banyak dipakai oleh motor-motor sport saat ini seperti yamaha, honda dan bajaj (pulsar). Dan lumayan, meningkatkan penjualan karena memiliki disain yang baik. Bagaimana disain yang akan datang, tentu saja kita harus mengikuti arah pasar.

Perancangan Produk Berbasis Komputer

Tak sadar waktu terus berlalu dan tahun ajaran baru mulai kita masuki. Seperti biasa semester ini tugas mengajar mata kuliah perancangan produk menjadi tanggung jawab saya. Saatnya evaluasi terhadap materi ajar tahun lalu yang agak kurang maksimal karena dasar menggambar komputer berbasis komputer yang kurang. Mudah-mudahan semester ini bisa diperbaiki, apalagi siswa sudah mengenal software CATIA yang cukup baik untuk merancang produk.

Masalah utama bagi siswa untuk mampu belajar disain adalah sulitnya menguasai perangkat lunak pendukung, terutama untuk fitur-fitur yang rumit. Dua tahun yg lalu sempat diperkenalkan AutoCAD hingga ke 3D ternyata sangat rumit. Dan akhirnya dialihkan ke CATIA yang agak mendingan walaupun sedikit lebih mudah, terbukti beberapa siswa menghasilkan gambar yang lumayan baik. Salah satu kelemahannya adalah CATIA tidak mampu menganalisa panas, sehingga membutuhkan tool tambahan misalnya CFD.

Seperti yang diucapkan oleh Prof Habibie bahwa dengan sebongkah besi, kita bisa membuat pacul seharga puluhan ribu, tetapi dengan otak kita, kita bisa membuat komponen berharga puluhan juta. Tentu saja didukung oleh kemampuan menggambar dan berimajinasi yang baik. Jika pada semester lalu (Mata kuliah SIMBAD CAD) kita masih dalam tarap “mencontek” gambar, dalam mata kuliah ini diharapkan mahasiswa mampu menemukan disain baru.

Semester baru, tentu saja ada hal-hal baru yang akan saya berikan. Karena kebanyakan produk-produk yang tersedia di pasaran berupa disain-disain yang menarik minat konsumen, tentu saja penggunakan software yg fleksibel dan mudah untuk disain mutlak diperlukan. Mungkin ada yang sudah mengenal 3Ds Max dari vendor Autodesk? Tampilannya adalah seperti gambar di bawah ini (versi 2011).

Dengan software ini diharapkan mahasiswa mampu merancang bentuk yang rumit dengan mudah, mengenai gambar tekniknya (drafting) kita tinggal mengkonversi gambar rumit kita ke format AutoCAD, oleh karena itu mau tidak mau harus mengulang kembali mata kuliah AutoCAD.

Metodologi dalam Pembuatan Perangkat Lunak

Seperti halnya perancangan produk di industri, merancang produk perangkat lunak juga memerlukan metodologi agar produk yang dihasilkan berkualitas baik. Karena karakteristiknya yang unik (tidak bisa aus, cepat berkembang, dll) software memiliki metodologi beragam yang telah dilakukan oleh pengembang-pengembang perangkat lunak.

Dalam bukunya, Roger S presman membagi metodologi menjadi bermacam-macam (waterfall, incremen, spiral, prototype, dll). Namun, pendekatan yang disarankan oleh Martin Fowler dalam bukunya UML Distilled cukup menarik, yakni hanya membagi metodologi menjadi waterfall dan iterasi. Sedangkan yang lainnya seperti spiral, incremen, dimasukan dalam kategori iterasi. Berikut penjelasan singkatnya.

Metodologi waterfall, sesuai dengan namanya “air terjun” merupakan metode klasik yang telah digunakan oleh analis dan disain perangkat lunak. Metode ini membagi proses pembuatan perangkat lunak dalam fase-fase seperti analisa, disain, coding, testing dan implementasi dengan urutan yang jelas. Karena memiliki kelemahan yang cukup signifikan, metode Iterasi membagi proses pembuatan perangkat lunak menjadi tahapan-tahapan yang tiap tahapan terdiri dari fase-fase yang ada pada waterfall ( analisa, disain, coding, testing dan implementasi). Sehingga kemungkinan kegagalan dalam proses pembuatan software dapat ditekan sekecil mungkin. Tahapan tersebut disusun mulai dari kebutuhan software terkecil hingga lengkap, namun tentu saja membagi menjadi tahap-tahap bukan merupakan pekerjaan yang mudah.

Sedangka dalam hal perancangan, Martin Fowler dalam buku yang sama juga menjelaskan bahwa ada dua jenis perancangan, yaitu prediktif dan adaptif. Perencanaan prediktif mengharuskan vendor pembuat perangkat lunak mampu memprediksi baik dari sisi kebutuhan software maupun hal-hal lain. Sedangkan perencanaan adaptif vendor pembuat tidak memiliki prediksi yang jelas, sehingga kebutuhan sofware selama proses pembuatan perangkat lunak bisa saja berubah (beradaptasi) mengikuti kebutuhan konsumen yang fleksibel. Oleh karena itu Martin Fowler menyarankan dalam merancang suatu sofware kita memanfaatkan tools system sebaiknya dari yang sederhana kemudian kita tambah sesuai kebutuhan dari pada memanfaatkan tools system yang kompleks dan kemudian satu persatu kita hilangkan mengikuti kebutuhan.

Namun kebanyakan kampus-kampus menggunakan metodologi yang ada di buku-buku teks klasik (Roger S Pressman dan Ian Sommerville) walaupun saat ini sudah mulai muncul metode-metode baru yang menyesuaikan dengan kebutuhan/karakter software yang dirancang seperti Agile dan Extreme Programming (XP). Menilik dari pengalaman-pengalaman yang lalu dalam membuat bahasa standar object programming UML yang banyak memakan waktu dan dana, ada baiknya para metodis (pakar di bidang metodologi) agar sedikit longgar dan mengikuti tren pasar. Bahkan sering disindir, “bedanya metodis dengan teroris hanya satu, yaitu kita bisa bernogosiasi dengan teroris”. 🙂

Automatic Neural Network-Based Network Analyzer for Hot Spot Area

After about three months, we’ve just finished the Automatic Neural Network-Based Network Analyzer for Hot Spot Area. We created it by Matlab program and by other softwares such us Wireshark (http://www.wireshark.org) and Quick Screenshot Capture.

This system works simple by analyzing the image (with Artificial Neural Network Algorithm) from network traffic graph of wireshark that has captured periodically by Quick Screenshot Capture (or other application that available in the market). The resume will cluster the network by three cluster: Normal Condition, High Traffic Condition and Unnormal Condition. The unnormal condition may be happen if the viruses or DOS attacks threat our Hot Spot. Here is the demonstration of our application.

instead of for Network Analyzer, this system can be implemented for analyzing surveillance by using web cam or other devices.

Mengambil data dari Excel ke MATLAB

Selain mengambil data dari file berekstensi DAT, Matlab juga bisa mengambil data dari Microsoft Excell (baik 2007 maupun 2002/2003). Fungsi yang digunakan adalah “open”. Untuk mempraktekannya coba buka Microsoft Excell kemudian coba buat satu field berisi dua buah field (kolom) berikut ini.

Perhatikan Nama Sheet perlu diganti karena nama ini akan menjadi nama variabel data di workspace Matlab. Ganti Sheet1 menjadi Data misalnya. Simpan dengan nama bebas, misalnya tabel, tidak perlu disave as menjadi word 2002/2003. Buka Command Window Matlab, lakukan instruksi:

>>Open tabel.xlsx

Klik “Finish” saat jendel “Import Wizard” terbuka, centang isian M-Code, jika akan diaplikasikan dalam bentuk Script M-File. Klik tombol radio “Other” terlebih dahulu.

Berikutnya akan muncul satu variabel baru Data, yang jika kita ketik variabel tersebut akan memunculkan data yang sama dengan data excell.

>> Data

Data =

1 11

2 12

3 111

4 14

5 25

6 56

7 67

8 86

9 54

10 67

Berikut ini hasil generati kode –M.

function importfile(fileToRead1)
%IMPORTFILE(FILETOREAD1)
% Imports data from the specified file
% FILETOREAD1: file to read
% Auto-generated by MATLAB on 08-Aug-2011 13:50:26
DELIMITER = ' ';
HEADERLINES = 0;
% Import the file
newData1 = importdata(fileToRead1, DELIMITER, HEADERLINES);
% Create new variables in the base workspace from those fields.
vars = fieldnames(newData1);
for i = 1:length(vars)
assignin('base’, vars{i}, newData1.(vars{i}));
end

Coba sendiri untuk akses ke GUI-nya ya.

Mengarahkan Fuzzy C-Mean ke “Jalan Yang Benar”

Clustering merupakan masalah yang lumayan rumit. Misalkan kita punya serangkaian data yang terdiri dari dua kategori yaitu nilai IPA dan nilai IPS. Jika kita lakukan klasifikasi langsung dengan FCM kita tidak serta merta mendapatkan hasil klasifikasi nilai yang rata-rata besar di IPA dan yang rata-rata besar di IPS. Mengapa demikian? Karena ada kemungkinan data tersebut terklasifikasi menjadi nilai yang rata-rata IPA dan IPS kecil dan nilai rata-rata IPA dan IPS besar. Berikut ini grafik hasil olah data yang dilakukan oleh contoh help matlab (ketik help fcm). Coba lihat … data tercluster menjadi siswa yang pinter IPA dan IPS dengan yang bodoh IPA dan IPS, padahal yang kita cari siswa yang cenderung IPA (nilai IPA > IPS) dengan yang cenderung IPS (nilai IPS>IPA).

Oleh karena itu agar kita memperoleh klasifikasi antara kelompok IPA dan kelompok IPS kita harus menambahkan satu kategori di kolom berikutnya. Katakanlah jika kelompok IPA (besar nilai IPA-nya) kita kategorikan “1” dan sebaliknya IPS dengan “0”. Pindahkan ke Excel lalu beri satu kolom baru dengan instruksi IF: “=IF(A1<B1;0;1)”. Sehingga diperoleh data baru yang akan diklasifikasi oleh fcm.