Model Pembelajaran Proses Penyeimbang Dinamik Dan Pembuatan Prototipe Mesin Penyeimbang Dinamik Roda Mobil Berbasis Komputer

Winarto, Suhardjono

Abstract: The fast growth of the mobile transportation in Indonesia, especially automobile which in year 2008 increase up to more than 600.000 unit and motorcycle go up to more than 6 million unit, accelerate the supporting technology mainly for its maintenance and repair. The technology of dynamic balancing also must be developed to increase the local content and the domestic market share in industry. For this purpose, it is important to do a experimentally research to learn and to understand the principle of the dynamic balancing process, not only theoretical but also practically experimental. Because of that this research split of by two themes, there are a developing learning model and a prototype of wheel balancing machine for its application. Developing a learning model is used to trace experimentally the fundamental principle of the dynamic balancing process both with phase method and four run method for a single rotating disc according to a miniature of car wheel. The result shows that the four run method has a high balance performance up to 92%, but it needs more time to measure four times the vibration amplitude, after the trial weight on the disc is applied. On the contrary, the phase method can be done fast and easy, however the balance performance is even lower. The reason is that the phase angle depends also on the rotating speed to determine the location of its balance mass. The prototype of the wheel balancing machine uses a swing arm technology, which is developed in the machine tool and vibration laboratory, while the old technology with a sliding mechanism give a vibration signal not a pure sinusoidal, therefore it is difficult to determine the position of the balance mass to be added. This research investigates many tire size with its rim diameter of 12 in., 13 in., 14 in. and 15 in. with three variation of rotating speed of 457 rpm, 510 rpm and 661 rpm for the accommodated balancing process below the natural frequency, on the natural frequency zone and the above natural frequency. The result of the measuring of characteristic of the wheel balancing machine prototype is decreasing amplitude from the initial unbalanced condition to the balance condition or defined as a balance performance of 89.84% for wheel with 13 in. rim, which is performed by rotating speed on the natural frequency zone. By using rotating speed below the natural frequency, the balance performance of 73.9% and 75% is reached by the wheel with 12 in. rim and 14 in. diameter respectively. For the wheel with 15 in. rim diameter gives a balance performance of 89.23%, which is done by the number of rotating above the natural frequency. The balancing process below the natural frequency comes about a phase shift of 180� and the above natural frequency give a phase shift of approximately of 90

Read More …

Klasifikasi Objek Dalam Visi Komputer Dengan Analisis Diskriminan

Amir Hamzahan, Gatot Santosa, Wisnu Widiarto

Abstract: A robotic sensor system is always supported by a computer system called ‘computer vision’. The important concept of computer vision is object classfi fi cation. In this study two algorithms for object classifi cation in this system will be compared. Firstly, A simple method that do not need complex computation and that considered as an informal method is called binary tree decision structure. This method is based on modest caracteristic decriptors of an object such as vertical line, horizontal line or ellipse line. Unfortunately this method has weakness in recognize an image that contaminated by a noise. Secondly, a more formal method with high variability descriptors. In this contect a multivariate statistical approach named discriminant analysis is proposed as an alternative for object classifi cation. This method is operated by computation of a function called Fisher discriminant function that can be used for separating an object. From the data simulation and analysis for calssifi cation of two object i.e. screw and bolt and three objects i.e. alphabet T,O and S it can be shown that discriminant analysis approach can classify an object better than binary decision algorithm. The superority of discriminant method is especially seen when this method is applied for classifi cation of a noisy image of object.

Read More …

Klasifikasi Citra Menggunakan Metode Minor Component Analysis pada Sistem Temu Kembali Citra

Vera Yunita, Yeni Herdiyeni

Abstract: This paper presents a method for image classification. In content-based image retrieval (CBIR) system, retrieving process is done by comparing a query image to all images from image database. This process is not effective because spends much times besides the retrieved images are not always match with the query image particularly for large databases. To solve this problem, image classification is proposed. In this research, minor component analysis (MCA) is used for images classification. For each image, a feature vector describing color, shape and texture. MCA vector will be formed as a representative pattern for each images class. In classification process, image database divided for image training and testing. Train data used to build classification model while test data used for test the accuracy of classification model. From this research it can be concluded that usage of MCA can improve the accuracy about 33.20%. This improvement shows that MCA can be applied in CBIR system, especially for large images database.

Read More …

Pengembangan Sistem Pembentukan Word Graph untuk Teks Berbahasa Indonesia

Deni Romadoni, Sri Nurdiati

Abstract: Knowledge graph is a new method which is used to describe and model natural language or human language. A word is a basic unit of natural language processing. Word graph is a graph which has connected structure of words. Ontology between word and token in word graph is represented by a node, 8 binary relationships, 4 frame relationships, and a focus. In this research, a system to construct a word graph from Indonesian text is developed. The software was initially developed by Mark van Koningsveld, which is called DelftCoStruct. On the previous version, knowledge graph method is not fully implemented. The first step in this research is analyzing the outline description and the requirement system. There are three principal requirements: formation and modification of word graph, analyzing text in word graph, and saving the dictionary capabilities. The second step in this research is designing specification system. The third step is developing and implementing the system designed. And the last step is validating system, to ensuring that all the requirements is fully implemented on this system.

Read More …

Koreksi Ejaan Query Bahasa Indonesia Menggunakan Algoritme Damerau Levenshtein

Utis Sutisna, Julio Adisantoso

Abstract: Query spelling on search engine is important to improve the quality information searching result. When user types query for search engine input, sometime spelling mistakes occurred due to position of keyboard and finger movement while typing. As an effect, searching result is incorrect and when user misspells the query, information obtained will not succeed. Therefore, search engine requires an application of spelling corrections. This research proposes correction process for query spelling by giving words suggestions which are obtained by calculating edit distance for each corrected word towards every word in dictionary. The concept for calculating edit distance uses Damerau Levenshtein algorithm which consists of 4 operations: (1) insertion, (2) substitution, (3) deletion, and (4) transposition. This research shows that implementation of Damerau Levenshtein algorithm is able to increase recall-precision value in information retrieval system. It is shown by the increasing of average recall-precision value at 44,82% after correction.

Read More …

Penentuan Rute Terpendek Menggunakan Variasi Fungsi Heuristik Algoritme A* Pada Mobile Devices

Pandu Satria Nur Ananda, Sri Wahjuni, Endang Purnama Giri

Abstract: GPS (Global Positioning System) is a technology that can help people in reaching the destination place in a fastest time and shortest route. Many mobile devices, some of them are cellular phones, have GPS facility. But this type of cellular phone is still expensive. This research intended to develop a shortest path searching application for mobile devices using variation of heuristic function A* algorithm, such as Manhattan, Euclidian, and Square of Euclidian. In addition, we intended to compare the execution time of these heuristic functions in finding shortest route.
The data used in the experiment was Bogor Agricultural University map that obtained from Bogor Agricultural University Facility and Property Directorate. The map ought to modified to grid form to be appeared in cellular phone screen. The matrix that used to grid had a size of 15 x 12 and values of 0, 1, 5, 6, 7, 8, 9, and 10. Value of 0 expressed walkable route, and value of 1 expressed unwalkable route. The rest values expressed the map facilities, like faculties and rectorate buildings.
The experiments indicated that Square of Euclidian had faster execution time and less amount of checked nodes than other heuristic functions. The lack was that this algorithm did not always provide shortest route. Euclidian heuristic function had slower execution time and more amount of checked nodes than others, whereas Manhattan have faster execution time and less amount of checked node than Euclidian. Euclidian and Manhattan heuristic function always provide shortest route in finding destination.

Read More …

Perbandingan Algoritme Pruning pada Decision Tree yang Dikembangkan dengan Algoritme CART

Martin Budi, Rindang Karyadin, Sony Hartono Wijaya

Abstract: Pruning is part of the development of decision tree. As decision tree is developed, some nodes became outliers as the results of noise data. Implementation of the decision tree pruning, can reduce the noise and outliers on the initial decision tree so that it can improve the accuracy of the data classification. Therefore the selection of proper pruning algorithm needs to be done to get the maximum results of the classification.
This experiment uses data from the company's customer credit providers. The data obtained from the data bank at the University of California. Data used in this experiment has twenty variables with two classes and 1000 instances. The data contain thirteen qualitative variables and the rest is a numeric data. The data is a good for use because it does not have a missing value.
The experiment compared three pruning algorithm, Cost Complexity Pruning (CCP), Reduced Error Pruning (REP), Error Based Pruning (EBP). Those pruning algorithms do prune to the decision tree that was developed with the Classification and Regression Tree (CART) algorithm. The comparison of those algorithms is done repeatedly on the data with different conditions both in terms of the instance number and the data variables. Comparison of the algorithm includes a comparison of the accuracy of the decision tree, and the process time of pruning algorithm.
The experiment’s result shows the average error rate of that the REP algorithm will produce the smallest error rate. Although the error rate of REP algorithm is the smallest, the difference value between ERP’s and EBP’s error rate is only 0.5%. Even though they have almost similar error rate, EBP algorithm proposes more simple decision tree than REP algorithm does.

Read More …

Sistem Rekomendasi Penambahan Link pada Website berdasarkan Data Log Website

Kurniawan Aji Saputra, Annisa Annisa

Abstract: Abstract---A good website structure is needed by the website visitors. The website structure which is created by the website developer should be in accordance with user preferences. The easier the user can find the target page, the better the website structure is. However if the user still get some difficulties in finding the location of the target page, website developer must change the website structure. Data logs from the website can help developers to improve the structure of the website. Data log should pass through the process of data preprocessing which are data cleaning, user identification, session identification, path completion and transaction identification. After that, expectation location will be searched in the data log by using Find Expectation Location algorithm. Location expectations are then processed using the optimization algorithm (FirstOnly and OptimizeBenefit) to get the recommendation of adding a page link. Finally the improved structure will be displayed using a scalable vector graphic (SVG).

Read More …

Brain Meets Brawn: Why Grid and Agents Need Each Other

Ian Foster, Argonne National Laboratory and University of Chicago
Nicholas R. Jennings, University of Southampton
Carl Kesselman, University of Southern California

Abstract: The Grid and agent communities both develop concepts and mechanisms for open distributed systems, albeit from different perspectives. The Grid community has historically focused on "brawn": infrastructure, tools, and applications for reliable and secure resource sharing within dynamic and geographically distributed virtual organizations. In contrast, the agents community has focused on "brain": autonomous problem solvers that can act flexibly in uncertain and dynamic environments. Yet as the scale and ambition of both Grid and agent deployments increase, we see a convergence of interests, with agent systems requiring robust infrastructure and Grid systems requiring autonomous, flexible behaviors. Motivated by this convergence of interests, we review the current state of the art in both areas, review the challenges that concern the two communities, and propose research and technology development activities that can allow for mutually supportive efforts.

Read More …