Articles

Found 30 Documents
Search

MINING MULTIDIMENSIONAL FUZZY ASSOCIATION RULES FROM A DATABASE OF MEDICAL RECORD PATIENTS

Jurnal Informatika Vol 9, No 1 (2008): MAY 2008
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Original Source | Check in Google Scholar | Full PDF (273.576 KB)

Abstract

Mining association rules is one of the important tasks in the process of data mining application. In general, the input as used in the process of generating rules is taken from a certain data table by which all the corresponding values of every domain data have correlations one to each others as given in the table. A problem arises when we need to generate the rules expressing the relationship between two or more domains that belong to several different tables in a normalized database. To overcome the problem, before generating rules it is necessary to join the participant tables into a general table by a process called Denormalization Process. This paper shows a process of generating Multidimensional Fuzzy Association Rules mining from a normalized database of medical record patients. The process consists of two sub-processes, namely sub-process of join tables (Denormalization Process) and sub-process of generating fuzzy rules. In general, the process of generating the fuzzy rules has been discussed in our previous papers [1, 2, 3, 4]. In addition to the process of generating fuzzy rules, this paper proposes a correlation measure of the rules as an additional consideration for evaluating interestingness of provided rules.

APLIKASI QUESTION ANSWERING SYSTEM DENGAN METODE RULE-BASED QUESTION ANSWERING SYSTEM PADA ALKITAB

Jurnal Informatika Vol 11, No 1 (2012): MAY 2012
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Original Source | Check in Google Scholar | Full PDF (228.353 KB)

Abstract

The Bible as the holy book of Christians who are very close to the religious life and as a moral guide for Christians. So the Bible become a necessity when a christians want to search for for specific data or information. But sometimes to find the answer to a question people sometimes having a trouble, because people did not know how to find the answer that they are looking for at the verses in the Bible that’s relatively large of amount. Therefore an application that have an ability to provide answers from the Bible verses that have the possibility of answers to questions raised by the user is needed. Where users can enter questions using keyword when, where, why, whom and what. Question Answering System Application will operate on a digital Bible in Indonesian language by using Rule-Based Question Answering System and created using Visual Basic 6.0 and Microsoft Access 2003 database. Based on application testing that made, the aplication has been able to find answers to the questions that asked according to the keywords. Meanwhile, based on testing with the questionnaire, the application obtained an average percentage of 77.2% from the respondents.

HARD: SUBJECT-BASED SEARCH ENGINE MENGGUNAKAN TF-IDF DAN JACCARD’S COEFFICIENT

Jurnal Teknik Industri Vol 8, No 1 (2006): JUNE 2006
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Original Source | Check in Google Scholar | Full PDF (88.964 KB)

Abstract

This paper proposes a hybridized concept of search engine based on subject parameter of High Accuracy Retrieval from Documents (HARD). Tf-Idf and Jaccard´s Coefficient are modified and extended to providing the concept. Several illustrative examples are given including their steps of calculations in order to clearly understand the proposed concept and formulas. Abstract in Bahasa Indonesia : Paper ini memperkenalkan suatu algorima search engine berdasarkan konsep HARD (High Accuracy Retrieval from Documents) dengan menggabungkan penggunaan metoda TF-IDF (Term Frequency Inverse Document Frequency) dan Jaccard´s Coefficient. Kedua metoda, TF-IDF dan Jaccard´s Coefficient dimodifikasi dan dikembangkan dengan memperkenalkan beberapa rumusan baru. Untuk lebih memudahkan dalam mengerti algoritma dan rumusan baru yang diperkenalkan, beberapa contoh perhitungan diberikan. Kata kunci: HARD, Tf-Idf, koefisien Jaccard, search engine, himpunan fuzzy.

Bayesian Belief Network untuk Menghasilkan Fuzzy Association Rules

Jurnal Teknik Industri Vol 12, No 1 (2010): JUNE 2010
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Original Source | Check in Google Scholar | Full PDF (201.875 KB)

Abstract

Bayesian Belief Network (BBN), one of the data mining classification methods, is used in this research for mining and analyzing medical track record from a relational data table. In this paper, a mutual information concept is extended using fuzzy labels for determining the relation between two fuzzy nodes. The highest fuzzy information gain is used for mining fuzzy association rules in order to extend a BBN. Meaningful fuzzy labels can be defined for each domain data. For example, fuzzy labels of secondary disease and complication disease are defined for a disease classification. The implemented of the extended BBN in a application program gives a contribution for analyzing medical track record based on BBN graph and conditional probability tables.

AN ALTERNATIVE CONCEPT OF FUZZY FUNCTIONAL DEPENDENCY BY USING CONDITIONAL PROBABILITY

Jurnal Informatika Vol 2, No 1 (2001): MAY 2001
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Original Source | Check in Google Scholar

Abstract

This paper proposes an alternative concept of Fuzzy Functional Dependency (FFD) based on the theory of conditional probability. We examine some properties of conditional probability and its relation with fuzzy sets. Moreover, approximate conditional probability of two fuzzy sets can be considered to determine the degree of similarity between two fuzzy sets. Based on this property, we propose an alternative concept of FFD and prove that it satisfies classical/crisp relational database. Abstract in Bahasa Indonesia : Sebuah alternatif konsep dari Fuzzy Functional Dependency (FFD) berdasarkan teori conditional probability diperkenalkan beberapa properties dan conditional probability dalam hubungannya dengan fuzzy sets didiskusikan. Conditional probability dan dua fuzzy sets dipertimbangkan untuk menentukan derajat kesamaan dan dua fuzzy sets. Berdasarkan konsep ini ini, diperkenalkan sebuah alternatif konsep dari FFD. Kata kunci: fuzzy sets, fuzzy functional dependency, fuzzy relational database, conditional probability.

APPLICATION OF CONDITIONAL PROBABILITY IN PREDICTING INTERVAL PROBABILITY OF DATA QUERYING

Jurnal Informatika Vol 5, No 2 (2004): NOVEMBER 2004
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Original Source | Check in Google Scholar

Abstract

This paper discusses fuzzification of crisp domain into fuzzy classes providing fuzzy domain. Relationship between two fuzzy domains, Xi and Xj, can be represented by using a matrix, wij. If Xi has n elements of fuzzy data and Xj has m elements of fuzzy data then wij is n x m matrix. Our primary goal in this paper is to generate some formulas for predicting interval probability in the relation to data querying, i.e., given John is 30 years old and he has MS degree, how about his probability to get high salary.

PENGGUNAAN HIBRIDISASI GENETICS ALGORITHMS DAN FUZZY SETS UNTUK MEMPRODUKSI PAKET SOAL

Jurnal Informatika Vol 6, No 1 (2005): MAY 2005
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Original Source | Check in Google Scholar

Abstract

At least, two important factors, discrimination and difficulty, should be considered in determing whether a problem should be in a packet of problems produced for students entrance examination at a university. The higher the discrimination degree of a problem, the better the problem is used to make a selection of participants based on their intellectual capability. How to provide a packet of entrance examination problems satisfying a determined pattern of discrimination and difficulty is a major problem in this paper for which an algorithm, it can be proved that the beneficiary of applying fuzzy sets and fuzzy relation in determing the first chromosome in the process of GA is that the process can reach tolerable solutions faster. Maximum number of generation is still needed as a threshold to overcome the problem of run time system overflow. Generally, the problems in the form of passages tend to have lower fitnest cost. Abstract in Bahasa Indonesia : Proses penyusunan paket soal (misalnya soal untuk test seleksi masuk universitas) yang diambil dari suatu bank soal, minimal harus memperhatikan dua aspek penting yaitu: tingkat kesulitan dan tingkat diskriminan soal. Semakin tinggi tingkat diskriminan suatu soal, semakin baik soal tersebut dipakai untuk menyeleksi kemampuan peserta test. Permasalahan yang dihadapi adalah bagaimana agar pembuat soal dapat memilih dan menentukan kombinasi soal-soal yang tepat (optimum) sehingga dapat memenuhi tingkat kesulitan dan diskriminan yang dikehendaki. Untuk menyelesaikan masalah ini, diperkenalkan suatu algoritma yang disusun dengan menggunakan hibridisasi metode Genetics Algorithm dan fuzzy sets. Dari hasil pengujian, didapatkan bahwa penggunaan fuzzy sets dan fuzzy relations dalam pemilihan kromoson awal akan lebih mempercepat pencapaian tolerable solutions. Tetap dibutuhkan treshould maksimum jumlah generasi yang dilakukan untuk mencegah run time sistem overflow. Soal bacaan cenderung memiliki nilai Fitnest Cost yang lebih rendah daripada nilai Fitnest Cost untuk paket soal bukan soal bacaan. Kata kunci: genetics algorithm, fuzzy sets, hibridization method.

APLIKASI PEMINJAMAN RUANGAN, KENDARAAN, PERALATAN DAN PENYAMPAIAN KELUHAN SERTA ANALISIS KETERSEDIAAN BARANG DAN KINERJA PELAYANAN PADA UPPK UNIVERSITAS KRISTEN PETRA

Jurnal Informatika Vol 6, No 2 (2005): NOVEMBER 2005
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Original Source | Check in Google Scholar

Abstract

Unit Pelayanan dan Pemeliharaan Kampus (UPPK) Universitas Kristen Petra has a difficulty in facing the customers. Sometimes, a customer has to contact or meet UPPK staff just for asking schedule or give complain. Besides, UPPK needs a system that has ability to analyze equipments needed, base on customers request and to analyze serving evaluation. Because of these problems, a web based system of distributing information, equipments and tools ordering, transportation ordering, room ordering, sport-field ordering, giving complain, equipments needed analysis and serving evaluation analysis is made. With this system, customers can obtain schedule information and then make an order, or giving complain, without have to meet the UPPK staff. This system is developed using PHP and MySQL as database server. The system is validated by distributing some questionnaires for UPPK staff and customers that give an average value of 3.63 (scale 1-4). Therefore, it can be concluded that this system can help to solve the problems. Besides, the systems also give a facility to analyze the serving evaluation and the need of equipments and tools. Abstract in Bahasa Indonesia : Unit Pelayanan dan Pemeliharaan Kampus (UPPK) Universitas Kristen Petra mengalami kesulitan dalam berhadapan dengan customer. Terkadang customer harus menghubungi atau bertatap muka dengan pihak UPPK hanya untuk menanyakan jadwal pemakaian atau menyampaikan keluhan. Selain hal itu UPPK juga membutuhkan suatu sistem yang mampu menganalisa kebutuhan peralatan/barang berdasar permintaan customer serta analisa kinerja yang telah dilakukan. Melihat permasalahan yang ada, maka dibuat suatu sistem penyampaian informasi, pemesanan peralatan/barang, pemesanan ruang khusus, pemesanan lapangan, pemesanan kendaraan, penyampaian keluhan, analisa kebutuhan barang dan kinerja berbasis web. Dengan sistem ini customer dapat memperoleh informasi jadwal dan melakukan pemesanan serta penyampaian keluhan tanpa harus bertatap muka dengan pihak UPPK. Sistem ini dikembangkan dengan bahasa PHP dan database server MySQL. Dari hasil pengujian dengan kuisioner kepada UPPK dan customer pengguna fasilitas didapat nilai rata-rata 3,63 (dari skala 1-4), sehingga dapat disimpulkan bahwa sistem dapat membantu memecahkan permasalahan yang ada. Selain itu sistem juga memberikan fasilitas untuk menganalisa evaluasi layanan dan kebutuhan barang. Kata kunci: web, pemesanan, analisa kebutuhan.

APLIKASI SEARCH ENGINE PAPER KARYA ILMIAH BERBASIS WEB DENGAN METODE FUZZY RELATION

Jurnal Informatika Vol 6, No 2 (2005): NOVEMBER 2005
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Original Source | Check in Google Scholar

Abstract

The number of paper collected by an educational institution is incresing each year. The increasing number of paper collected demand a method in order to find the right paper everytime there is someone who needs a reference. By far, most search engine still depend on keyword matching / string maching to find the apropriate result. This method will only find the apropriate paper based on the occurance of the inserted keyword on the paper. This research will discuss a searching system using fuzzy relation, by using fuzzy relation the relation between keyword and paper is found and determined. Searching system using fuzzy relation allows the search result include paper that do not have the keyword to be shown as a result. This result is made posssible because the word which occur in the paper is related to keyword inserted. Abstract in Bahasa Indonesia : Banyaknya jumlah paper yang dikoleksi sebuah lembaga pendidikan setiap tahun akan bertambah. Seiring dengan pertambahan jumlah paper tersebut maka diperlukan sebuah metode untuk mencari paper agar bila membutuhkan referensi maka paper/dokumen yang diperlukan dapat dengan mudah dapat ditemukan. Sejauh yang ada saat ini, kebanyakan mesin pencari masih mengandalkan pencarian dengan menggunakan keyword matching/string matching sehingga mengakibatkan hasil pencarian hanya akan menampilkan paper-paper yang mempunyai keyword/kata kunci yang dicari. Penelitan ini membahas sebuah sistem pencarian dengan menggunakan metode fuzzy relation, dimana dengan fuzzy relation didapatkan hubungan antara keyword dan paper. Dengan metode fuzzy relation maka sebuah pencarian mempunyai kemungkinan menampilkan hasil berupa paper yang tidak mengandung keyword yang dicari. Karena kata yang mengakibatkan paper (yang tidak mengandung keyword muncul) mempunyai hubungan dengan keyword yang dimasukkan. Kata kunci: mesin pencari, relasi fuzzy, sistem cerdas.

SOLVING QUICKEST PATH PROBLEM USING AN EXTENDED CONCEPT OF TRANSITIVE CLOSURE METHOD

Jurnal Informatika Vol 6, No 2 (2005): NOVEMBER 2005
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Original Source | Check in Google Scholar

Abstract

A travel company offers a transportation service by picking up customers from certain places and sending them to the other places. Concerning traffic in time domain function for every path, the problem is how to find the quickest paths in picking up the customers. The problem may be considered as the quickest path problem. We may consider the quickest path problem as an extension of the shortest path problem. The quickest path problem optimizes length of time in visiting some places. The quickest path is not always the shortest path. In the relation to the problem, this paper proposes an algorithm for solving the quickest path problem. The algorithm is constructed by modifying a concept of transitive closure method.