175 research outputs found
Pengembangan Metode Penyimpanan, Pengaksesan, dan Visualisasi Pola Temporal untuk Mempermudah Penyeleksian Pola Temporal Terbaik
No full textPada era informasi saat ini salah satu dari masalah yang sering dihadapi oleh perusahaan adalah bagaimana mengolah data mereka yang jumlahnya cukup besar menjadi informasi yang bermanfaat bagi perusahan. Untuk menghadapi tantangan ini banyak teknik-teknik yang sudah dikembangkan untuk menemukan informasi dari data. Informasi ini biasanya disajikan dalam bentuk pola (pattern) atau aturan (rule). Sebagian besar dari teknik ini fokus pada penemuan pola non-temporal dari data non-temporal. Pada kenyataannya banyak data yang dimiliki oleh peruashaan merupakan data temporal. Meskipun teknik untuk menemukan pola dari data temporal juga sudah banyak yang dikemnbangkan akan tetapi teknik-teknik tersebut masih lemah dalam hal post-processing. Untuk menemukan pola yang terbaik, post-processing memegang peran yang sangat penting karena dari ribuan pola yang dihasilkan oleh algoritma penemu pola hanya sebagian kecil saja yang bermanfaat, dan ini hanya daapt diidentifikasi melalui post-processing. Menyeleksi pola dari ribuan pola temporal merupakan pekerjaan yang tidak mudah. Untuk itu perlu dibuat suatu mekanisme yang dapat memudahkan pengguna dalam menyeleksi pola. Dalam penelitian ini dikembangkan suatu metode untuk penyimpanan dan pengaksesan pola yang efisien. Selain itu juga dikembangkan metode visualisasi pola temporal yang memudahkan pengguna dalam menyeleksi pola yang dianggap terbaik
Satellite imagery and machine learning for aridity disaster classification using vegetation indices
No full textCentral Java Province is one of provinces in Indonesia that has a high aridity risk index. Aridity disaster risk monitoring and detection can be done more accurately in larger areas and with lower costs if the vegetation index is extracted from the remote sensing imagery. This study aims to provide accurate aridity risk index information using spectral vegetation index data obtained from LANDSAT 8 OLI satellite. The classification of drought risk areas was carried out using k-nn with the Spatial Autocorrelation method. The spectral vegetation indices used in the study are NDVI, SAVI, VHI, TCI and VCI. The results show a positive correlation and trend between the spectral vegetation index influenced by seasonal dynamics and the characteristics of the High R.A. and Middle R.A. drought risk areas. The highest correlation coefficient is SAVI with a High R.A. amounted to 0.967 and Middle R.A. amounted to 0.951. The results of the Kappa accuracy test comparison show that SVM and k-nn have the same accuracy of 88.30. The result of spatial prediction using the IDW method shows that spectral vegetation index data that initially as an outlier, using the k-nn method, the spectral vegetation index data can be identified as data in the aridity classification. The spatial connectivity test among sub-districts that experience drought was done using Moran’s I Analysis
Pengembangan Metode Penyimpanan, Pengaksesan, dan Visualisasi Pola Temporal untuk Mempermudah Penyeleksian Pola Temporal Terbaik
No full text
Sistem Pencarian Informasi Berbasis Ontologi untuk Jalur Pendakian Gunung Menggunakan Query Bahasa Alami dengan Penyajian Peta Interaktif
No full textMountain climbing path information has been widely available on the internet. However, to get information that suits the needs of climbers take time to browse and compare all the available information. The diversity of the search results content actually confuse the climbers.
This research aims to provide a solution to the problems faced by climbers, by developing an information retrieval system for mountain climbing path using semantic technology (ontology) based approach .
The system is developed by using two knowledge base (ontology), ontology Bahasa represents linguistic knowledge and ontology Mountaineering represents mountaineering knowledge. The system is designed to process and understand natural language input form. The process of understanding the natural language based on syntactic and semantic analysis using the rules of Indonesian grammar.
The results of the research that has been conducted shows that the system is able to understand natural language input and is capable of detecting input that is not in accordance with the rules of Indonesian grammar both syntactically and semantically. The system is also able to use a thesaurus of words in the search process. Quantitative test results show that the system is able to understand 69% of inputs are taken at random from the respondents
Analisis Fitur Kalimat untuk Peringkas Teks Otomatis pada Bahasa Indonesia
No full textAbstract— Automatic Text Summarization (ATS) is a technique to create a summary of the document automatically by using computer applications to produce the most important information from the original document. Features are required to perform weighting of sentences, including Log-TFISF (term frequency index sentence frequency), sentence location, sentence overlap, title overlap and sentence relative length. This research conducted an analysis of five features in order to determine the weights of each feature that will get the results of a coherent summary. The five features are implemented in automated text summarization system in Indonesian language that was developed using the method of relative importance of topics. Results from experiments show that sentence location feature has the highest F-Measures namely 0.46 and then consecutive sentence overlap, title overlap, sentence relative length and Log-TFISF, with a value of 0.42, 0.42, 0.35 and 0.32. Relative weights of feature extraction consecutive from the largest are sentence location, sentence overlap, title overlap, sentence relative length and Log-TFISF with a value of 0.25, 0.22, 0.22, 0.19 and 0.12. These relative weights are implemented on ATS, so we get accuracy of 70.62%. It is more accurate 2,86% than without relative weights which accuracy of 67,72%..
.
Keywords— Automatic Text Summarization (ATS), Log-TFISF, sentence location, sentence overlap, title overlap, sentence relative length, bahasa Indonesi
Klasifikasi Data NAP (Nota Analisis Pembiayaan) untuk Prediksi Tingkat Keamanan Pemberian Kredit (Studi Kasus : Bank Syariah Mandiri Cabang Luwuk Sulawesi Tengah)
No full textAbstrak
Setiap bulannya bank syariah mandiri cabang luwuk menerima proposal kredit (NAP) dari nasabah dalam jumlah yang terus meningkat dan perlu respon yang cepat. Dengan demikian, perlu dikembangkan sistem untuk melakukan data mining dari tumpukan data tersebut yang akan digunakan untuk kepentingan tertentu, salah satunya adalah untuk menganalisis resiko pemberian kredit.Teknik data mining digunakan dalam penelitian ini untuk klasifikasi tingkat keamanan pemberian kredit dengan menerapakan algoritma Naïve Bayes Classificatio. Naive bayes classifier merupakan pendekatan yang mengacu pada teorema Bayes yang menkombinasikan pengetahuan sebelumnya dengan pengetahuan baru, sehingga merupakan salah satu algoritma klasifikasi yang sederhana namun memiliki akurasi tinggi. Sebelum dilakukan klasifikasi, data debitur melalui preprocessing. Kemudian dari preprocessing ini dilakukan klasifikasi dengan naive bayes classifier, sehingga menghasilkan model probabilitas klasifikasi untuk prediksi kelas pada debitur selanjutnya. Teknik pengujian akurasi model diukur menggunakan boostrap, dan menunjukkan bahwa nilai akurasi terkecil 80% dihasilkan pada sampel data 100, dan menghasilkan nilai akurasi terbesar 98,66% pada sampel data 463.
Kata kunci— akurasi, naive bayes, data mining, klasifikasi, preprocessing, NAP
Abstract
Every month the Mandiri Syariah Bank Branch Office of Luwuk receives a very large number of proposal credit. Thus, the system should be developed to perform data mining of the heap data to be used for specific purpose, one of which is for the risk analysis of credit allowance. Data mining techniques used in this study for classification level prediction of credit allowance by applying a naïve Bayes Classification algorithm . Naive bayes classifier is an approach that refers to the bayes theorem, is a combination of prior knowledge with new knowledge. So that is one of the classification algorithm is simple but has a high accuracy. Prior to classification, data of debitur has been through a preprocessing. Then the weight is to perform classification with naive bayes classifier. After the data is classified, so produce probabilitas of model classification for prediction class to next debitur. Testing techniques the accuracy of the model was measured by bosstrap, and shows that the smallest value of accuracy is 80% produced in the 100 data sample, and the largest value of accuracy 98,66% on a data sample of 463.
Keywords— accuracy, naive bayes, data mining, classification, preprocessing, NA
Analisis Opini Terhadap Fitur Smartphone Pada Ulasan Website Berbahasa Indonesia
No full textThrough online stores, consumers can give an opinion of a product, one of the best-selling products is smartphone. Their opinions become valuable and can be worthwhile to know the advantages or disadvantages of products based on the user’s experience. Therefore, in order to utilize the data of customers' opinions, it is necessary to create a system that automatically performs mining and summarizing opinions on smartphone product. The system consist of five parts: data collection, preprocessing review, feature mining, analysis of opinions and then visualize the results. Data collection is taking data reviews website using web scraping, preprocessing review is for cleaning data reviews. Feature mining stage will find features in the reviews with apriori algorithm to produce frequent item set, then analyze the opinion using lexicon based, language rule and score function. The result will be shown in graphical form. From the testing of feature mining obtained average recall score at 0.63 and precision at 0.72. It depends on good or bad quality of reviews. The results of testing accuracy opinion analysis shows high value with accuracy 81.76 %. The technique showed good results with opinion data which is labeled, using language rule and the implementation of score function
Penerapan Metode Support Vector Machine pada Sistem Deteksi Intrusi secara Real-time
No full textAbstrak
Sistem deteksi intrusi adalah sebuah sistem yang dapat mendeteksi serangan atau intrusi dalam sebuah jaringan atau sistem komputer, umum pendeteksian intrusi dilakukan dengan membandingkan pola lalu lintas jaringan dengan pola serangan yang diketahui atau mencari pola tidak normal dari lalu lintas jaringan. Pertumbuhan aktivitas internet meningkatkan jumlah paket data yang harus dianalisis untuk membangun pola serangan ataupun normal, situasi ini menyebabkan kemungkinan bahwa sistem tidak dapat mendeteksi serangan dengan teknik yang baru, sehingga dibutuhkan sebuah sistem yang dapat membangun pola atau model secara otomatis.
Penelitian ini memiliki tujuan untuk membangun sistem deteksi intrusi dengan kemampuan membuat sebuah model secara otomatis dan dapat mendeteksi intrusi dalam lingkungan real-time, dengan menggunakan metode support vector machine sebagai salah satu metode data mining untuk mengklasifikasikan audit data lalu lintas jaringan dalam 3 kelas, yaitu: normal, probe, dan DoS. Data audit dibuat dari preprocessing rekaman paket data jaringan yang dihasilkan oleh Tshark.
Berdasar hasil pengujian, sistem dapat membantu sistem administrator untuk membangun model atau pola secara otomatis dengan tingkat akurasi dan deteksi serangan yang tinggi serta tingkat false positive yang rendah. Sistem juga dapat berjalan pada lingkungan real-time.
Kata kunci— deteksi intrusi, klasifikasi, preprocessing, support vector machine
Abstract
Intrusion detection system is a system for detecting attacks or intrusions in a network or computer system, generally intrusion detection is done with comparing network traffic pattern with known attack pattern or with finding unnormal pattern of network traffic. The raise of internet activity has increase the number of packet data that must be analyzed for build the attack or normal pattern, this situation led to the possibility that the system can not detect the intrusion with a new technique, so it needs a system that can automaticaly build a pattern or model.
This research have a goal to build an intrusion detection system with ability to create a model automaticaly and can detect the intrusion in real-time environment with using support vector machine method as a one of data mining method for classifying network traffic audit data in 3 classes, namely: normal, probe, and DoS. Audit data was established from preprocessing of network packet capture files that obtained from Tshark.
Based on the test result, the system can help system administrator to build a model or pattern automaticaly with high accuracy, high attack detection rate, and low false positive rate. The system also can run in real-time environment.
Keywords— intrusion detection, classification, preprocessing, support vector machin
- …
