Browsing by Author "Tseng, SS"

Now showing 1 - 20 of 76

A content management scheme in a SCORM compliant learning object repository
(2005-09-01)
Su, JM
;
Tseng, SS
;
Wang, CY
;
Lei, YC
;
Sung, YC
With the rapid development of the Internet, e-learning systems have become more and more popular. For sharing and reusing teaching materials in different c-learning system, the Sharable Content Object Reference Model (SCORM) has become the most popular international standard among the existing ones. In an e-learning system, teaching materials are usually stored in a database, called the Learning Object Repository (LOR). In the LOR, a huge amount of SCORM teaching materials, including associated learning objects, will result in management problems in a wired/wireless environment. Recently, the SCORM organization has focused on devising ways to efficiently maintain, search. and retrieve desired learning objects in LORs for users. This effort is referred to as the Content Object Repository Discovery and Resolution Architecture (CORDRA). In this paper, we propose a management approach, called the Level-wise Content Management Scheme (LCMS), that can be used to efficiently maintain, search, and retrieve learning contents from a SCORM compliant LOR. LCMS includes two phases: the Construction phase and Search phase. In the former, the content structure of SCORM teaching materials (Content Package) is first transformed into a tree-like structure. called a Content Tree (CT), to represent each piece of teaching material. Based on Content Trees (CTs), the proposed Level-wise Content Clustering Algorithm (LCCAIg) then creates a multistage graph showing relationships among learning objects (LOs), e.g.. a Directed Acyclic Graph (DAG), called the Level-wise Content Clustering Graph (LCCG). The LCCAIg determines the relationships among LOs in different teaching materials by clustering all of the LOs for each level from bottom to top, according to a similarity measure. Moreover, a maintenance strategy is employed to rebuild the LCCG if necessary by monitoring the condition of each node within the LCCG. The latter employs the LCCG Content Searching Algorithm (LCCG-CSAlg) to traverse the LCCG and retrieve desired learning content with both general and specific LOs, according to queries sent by users in the wire/wireless environment. Some experiments have been done conducted to test the proposed scheme, and the results are reported here.
A coverage-based genetic knowledge-integration strategy
(2000-07-01)
Wang, CH
;
Hong, TP
;
Chang, MB
;
Tseng, SS
;
資訊工程學系
In this paper, we propose a coverage-based genetic knowledge-integration approach to effectively integrate multiple rule sets into a centralized knowledge base. The proposed approach consists of two phases: knowledge encoding and knowledge integration. In the knowledge-encoding phase, each rule in the various rule sets that are derived from different sources (such as expert knowledge or existing knowledge bases) is first translated and encoded as a fixed-length bit string. The bit strings combined together thus form an initial knowledge population. In the knowledge-integration phase, a genetic algorithm applies genetic operations and credit assignment at each rule-string to generate an optimal or nearly optimal rule set. Experiments on diagnosing brain tumors were made to compare the accuracy of a rule set generated by the proposed approach with that of the initial rule sets derived from different groups of experts or induced by various machine learning techniques. Results show that the rule set derived by the proposed approach is more accurate than each initial rule set on its own. (C) 2000 Elsevier Science Ltd. All rights reserved.
A data mining project for solving low-yield situations of semiconductor manufacturing
(2004)
Chen, WC
;
Tseng, SS
;
Hsiao, KR
;
Liu, CC
;
資訊工程學系
With huge amount of semiconductor engineering data stored in database and versatile analytical charting and reporting in production and development, the CIM/MES/EDA systems in the most semiconductor manufacturing companies help users to analyze the collected data in order to achieve the goal of yield enhancement. However, the procedures of semiconductor manufacturing are sophisticated and the collected data among these procedures are thus becoming high-dimensional and huge. Currently, some statistical methods, such like K-W test, covariance analysis, regression analysis, etc., have been used to analyze the information summarized from EDA system, and thus generate too many indexes that can not be easily judged and assimilated by engineers. Besides, too many false alarms may be raised and lots of time is required to check the factuality among them. In order to deal with the large amount and high-dimensional data, the data mining technologies are thus used to solve such problems. In this paper, we would like to propose a data mining solution and describe the experiences applying such solutions for discovering the root causes of low-yield situations in a worldwide semiconductor manufacturing company. Also, the situation of applying such mining solution for manufacturing defects detection in semiconductor manufacturing domain will be reviewed Finally, the architecture of a reasonable, reliable and flexible data mining system will be briefly described.
A distributed intrusion detection model for the domain name system
(2002-11-01)
Chen, CS
;
Tseng, SS
;
Liu, CL
;
資訊工程學系
;
Department of Computer Science
We have investigated the problem of detecting DoS-like DNS anomalies in DNS system. In this paper, we propose a distributed Two-phase DNS anomaly detection model for solving the problem. Three sets of algorithms corresponding to different configurations are proposed, including one sequential algorithm and two distributed algorithms, each with an increasing level of parallelism. The complexity of these algorithms have been found to be O (n l(og)n). The distributed algorithms show at least a constant (1-1/C-k), C-k > 1, improvement over the sequential one. To evaluate the performance, we have implemented the algorithms and applied them to a number of examples. The experimental result shows a speed up of about 1.68 on the test example for running on an enhanced distributed architecture with C-IDS over the sequential one. A higher speedup might be common because DNS anomalies will make the traffic distribution more concentrated on the outliers, and the computation will usually converge much more quickly.
A framework of features selection for the case-based reasoning
(2000)
Chen, WC
;
Tseng, SS
;
Chen, JH
;
Jiang, MF
;
資訊工程學系
CBR is a problem solving technique that reuses past cases and experiences to find a solution to the problems. A critical issue in case-based reasoning is to select the correct and enough features to represent a case. For this reason, the analysis of cases and extraction the necessary features to represent a case are highly recommended in building a CBR system. However, this task is difficult to carry out since such knowledge often cannot be successfully and exhaustively captured and represented. In this paper, a framework of features mining system for the case-based reasoning including two phases has been proposed. The techniques of features selection, data analysis and machine learning can thus be effectively integrated. This will promote flexibility and expandability of case-based reasoning system.
A fuzzy inductive learning algorithm for parallel loop scheduling
(1997)
Tsai, CJ
;
Tseng, SS
;
Wang, CH
;
Yang, CT
;
Jiang, MF
The conventional symbolic learning algorithm can not infer data that contains fuzzy information. In the past few years, we have designed a parallel loop scheduling based upon knowledge based approach that is called KPLS to choose an appropriate schedule for different loop to assign loop iterations to a multiprocessor system for achieving high speedup rates. Unfortunately, we found that these attributes that were applied in KPLS contain some fuzzy information, which are inapplicable to the traditional symbolic learning strategy for inferring some concept descriptions. In this paper, we apply fuzzy set concept to AQR learning algorithm that is called FAQR. FAQR which can induce fuzzy linguistic rules from fuzzy instances is then proposed to solve the above parallel loop scheduling problem. Some promising inference rules have been found and applied to infer the choice of parallel loop scheduling. Besides, we apply FAQR in IRIS Flower Classification Problem. Experimental results show that our method yields high accuracy in both different domains.
A fuzzy inductive learning strategy for modular rules
(1999-04-01)
Wang, CH
;
Liu, JF
;
Hong, TP
;
Tseng, SS
;
資訊工程學系
In real applications, data provided to a learning system usually contain linguistic information which greatly influences concept descriptions derived by conventional inductive learning methods. The design of learning methods to learn concept descriptions in working with vague data is thus very important. In this paper, we apply fuzzy set concepts to machine learning to solve this problem. A fuzzy learning algorithm based on the maximum information gain is proposed to manage linguistic information. The proposed learning algorithm generates fuzzy rules from "soft" instances, which differ from conventional instances in that they have class membership values. Experiments on the Sports and the Iris Flower classification problems are presented to compare the accuracy of the proposed algorithm with those of some other learning algorithms. Experimental results show that the rules derived from our approach are simpler and yield higher accuracy than those from some other learning algorithms. (C) 1999 Elsevier Science B.V. All rights reserved.
A generalized version space learning algorithm for noisy and uncertain data
(1997-03-01)
Hong, TP
;
Tseng, SS
;
資訊工程學系
;
Department of Computer Science
This paper generalizes the learning strategy of version space to manage noisy and uncertain training data. A new learning algorithm is proposed that consists of two main phases: searching and pruning. The searching phase generates and collects possible candidates into a large set; the pruning phase then prunes this set according to various criteria to find a maximally consistent version space. When the training instances cannot completely be classified, the proposed learning algorithm can make a trade-off between including positive training instances and excluding negative ones according to the requirements of different application domains. Furthermore, suitable pruning parameters are chosen according to a given time limit, so the algorithm can also make a trade-off between time complexity and accuracy. The proposed learning algorithm is then a flexible and efficient induction method that makes the version space learning strategy more practical.
A genetic algorithm for Multicast Routing under Delay Constraint in WDM network with different light splitting
(2005-01-01)
Chen, MT
;
Tseng, SS
;
資訊工程學系
;
Department of Computer Science
Because optical WDM networks will become a realistic choice for buildings backbones, multicasting in the WDM network should be supported for various network applications. In this paper, a new multicast problem, Multicast Routing under Delay Constraint Problem (MRDCP), routing a request with delay bound to all destinations in a WDM network with different light splitting is solved by genetic algorithms (GAs), where different light splitting means that nodes in the network can transmit one copy or multiple copies to other nodes by using the same wavelength. The MRDCP can be reduced to the Minimal Steiner Tree Problem (MSTP) which has been shown to be NP-Complete. We propose a destination-oriented representation to represent chromosomes, three general genetic operators (selection, crossover, and mutation), four types of operators (Chromosome Crossover, Individual Crossover, Chromosome Mutation, and Individual Mutation). Four mutation heuristics (Random Mutation (RM), Cost First Mutation (CFM), Delay First Mutation (DFM), and Hybrid Mutation (HM)) are employed in the GA method. Finally, experimental results show that our solution model call obtain a near optimal solution.
A genetic algorithm for solving dual-homing cell assignment problem of the two-level wireless ATM network
(2002-11-01)
Din, DR
;
Tseng, SS
;
資訊工程學系
;
Department of Computer Science
In this paper, we investigate the optimal assignment problem, which assigns cells in Personal Communication Service to switches on Asynchronous Transfer Mode network in an optimum manner. The cost has two components: one is the cost of handoffs that involve two switches, and the other is the cost of cabling. This problem is model as dual-homing cell assignment problem, which is a complex integer programming problem. Since finding an optimal solution of this problem is NP-hard, a stochastic search method, based on a genetic approach, is proposed to solve this problem. In this paper, domain-dependent heuristics are encoded into crossover operations, mutations of genetic algorithm (GA) to solve this problem. Simulation results show that GA is robust for this problem. (C) 2002 Elsevier Science B.V. All rights reserved.
A genetics-based approach to knowledge integration and refinement
(2001-01-01)
Wang, CH
;
Hong, TP
;
Tseng, SS
;
資訊工程學系
;
Department of Computer Science
In this paper, we propose a genetics-based knowledge integration approach to integrate multiple rule sets into a central rule set. The proposed approach consists of two phases: knowledge encoding and knowledge integrating. In the encoding phase, each knowledge input is translated and expressed as a rule set, and then encoded as a bit string. The combined bit strings form an initial knowledge population, which is then ready for integrating. In the knowledge integration phase, a genetic algorithm generates an optimal or nearly optimal rule set from these initial knowledge inputs. Furthermore, a rule-refinement scheme is proposed to refine inference rules via interaction with the environment. Experiments on diagnosing brain tumors were carried out to compare the accuracy of a rule set generated by the proposed approach with that of initial rule sets derived from different groups of experts or induced by means of various machine learning techniques. Results show that the rule set derived using the proposed approach is much more accurate than each initial rule set on its own.
A hybrid parallel loop scheduling scheme on grid environments
(2005)
Shih, WC
;
Yang, CT
;
Tseng, SS
;
資訊工程學系
;
Department of Computer Science
Effective loop-scheduling can significantly reduce the total execution time of a program on grid environments, especially for loop-intensive applications. This paper describes a two-phased method, named HPLS (Hybrid Parallel Loop Scheduling), to dynamically schedule loop iterations of a program on grid environments. In the first phase, most of the workload is dispatched to each node for execution according to its performance. Then, some well-known self-scheduling scheme is utilized to schedule the remaining workload. Experimental results showed that in most cases our approach could produce more efficient scheduling than previous schemes on our testbed grid. In addition, the results suggest that our approach is suitable for loop scheduling on grid environments.
A hybrid parallel loop scheduling scheme on heterogeneous PC clusters
(2005)
Shih, WC
;
Yang, CT
;
Chen, PI
;
Tseng, SS
;
資訊工程學系
Traditional loop-scheduling schemes are designed for homogeneous computing environments, and are probably not suitable for emerging heterogeneous PC clusters. This paper applies a two-phased method, named HPLS (Hybrid Parallel Loop Scheduling), to heterogeneous PC-cluster environments. The key idea is to distribute most of the workload to each node according to its performance, which is modeled and estimated in advance. Experiments on our testbed PC-cluster showed that, in most cases, our method could reduce the execution time of application programs more than previous schemes could.
A new approach for constructing the concept map
(2004)
Sue, PC
;
Weng, JF
;
Su, JM
;
Tseng, SS
;
資訊工程學系
For achieving the adaptive learning, a predefined concept map of a course is often used to provide adaptive learning guidance for learners. However, it is difficult and time consuming to create the concept map of a course. Thus, how to automatically create a concept map of a course becomes an interesting issue. In this paper, we propose a Two-Phase Concept Map Construction (TP-CMC) approach to automatically construct the concept map by learners' historical testing records. Phase I is used to preprocess the testing records. We apply Fuzzy Set Theory to transform the numeric testing records of learners into symbolic, apply Education Theory (Item Analysis for Norm-Referencing) to further refine it, and apply Data Mining approach to find its grade fuzzy association rules. Then, in Phase 2, based upon our observation in real learning situation, we use multiple rule types to further analyze the mined rules and then propose a heuristic algorithm to automatically construct the concept map. Finally, the Redundancy and Circularity of the concept map constructed are also evaluated.
A new hybrid learning algorithm for non-linear boundaries
(1998-06-01)
Wang, CH
;
Hong, TP
;
Tseng, SS
;
資訊工程學系
;
Department of Computer Science
In this paper, we propose a new hybrid learning algorithm, ETNC, which incorporates the popular decision-tree learning algorithm ASSISTANT into a modified three-layer back-propagation learning method to construct an entropy-tree net classifier. The new Iearning algorithm also adopts a tree-pruning mechanism to avoid overfitting problems. The new algorithm decreases both the tree size and error rate, especially for complex classification problems. Furthermore, it is not necessary for users to lay out the structure of a tree net in advance; instead, the structure is automatically constructed in the tree-growing process. Finally, the results of experiments in diagnosing brain tumors and classifying sugar canes are described to compare the proposed algorithm with two other learning methods, the back-propagation learning algorithm and ASSISTANT, in terms of accuracy, knowledge complexity and learning speed. Experimental results show that the proposed learning algorithm can provide a good trade-off between accuracy, knowledge complexity and learning speed.
A new probabilistic induction method
(1997-02-01)
Hou, RH
;
Hong, TP
;
Tseng, SS
;
Kuo, SY
;
資訊工程學系
Knowledge acquisition by interviewing a domain expert is one of the most problematic aspects of the development of expert systems. As an alternative, methods for inducing concept descriptions from examples have proven useful in eliminating this bottleneck. In this paper, we propose a probabilistic induction method (PIM), which is an improvement of the Chan and Wong method, for detecting relevant patterns implicit in a given data set. PIM uses the technique of residual analysis and several heuristics to effectively detect complex relevant patterns and to avoid the problem of combinatorial explosion. A reasonable trade-off between the induction time and the classification ratio is achieved. Moreover, PIM quickly classifies unknown objects using classification rules converted from the positively relevant patterns detected. Three experiments are conducted to confirm the validity of PIM.
A new probabilistic induction method
(1997-02-01)
Hou, RH
;
Hong, TP
;
Tseng, SS
;
Kuo, SY
;
資訊工程學系
Knowledge acquisition by interviewing a domain expert is one of the most problematic aspects of the development of expert systems. As an alternative, methods for inducing concept descriptions from examples have proven useful in eliminating this bottleneck. In this paper, we propose a probabilistic induction method (PIM), which is an improvement of the Chan and Wong method, for detecting relevant patterns implicit in a given data set. PIM uses the technique of residual analysis and several heuristics to effectively detect complex relevant patterns and to avoid the problem of combinatorial explosion. A reasonable trade-off between the induction time and the classification ratio is achieved. Moreover, PIM quickly classifies unknown objects using classification rules converted from the positively relevant patterns detected. Three experiments are conducted to confirm the validity of PIM.
A novel manufacturing defect detection method using association rule mining techniques
(2005-11-01)
Chen, WC
;
Tseng, SS
;
Wang, CY
;
資訊工程學系
;
Department of Computer Science
In recent years, manufacturing processes have become more and more complex, and meeting high-yield target expectations and quickly identifying root-cause machinesets, the most likely sources of defective products, also become essential issues. In this paper, we first define the root-cause machineset identification problem of analyzing correlations between combinations of machines and the defective products. We then propose the Root-cause Machine Identifier (RMI) method using the technique of association rule mining to solve the problem efficiently and effectively. The experimental results of real datasets show that the actual root-cause machinesets are almost ranked in the top 10 by the proposed RMI method. (c) 2005 Elsevier Ltd. All rights reserved.
A novel manufacturing defect detection method using data mining approach
(2004)
Chen, WC
;
Tseng, SS
;
Wang, CY
;
資訊工程學系
;
Department of Computer Science
In recent years, the procedure of manufacturing has become more and more complex. In order to meet high expectation on quality target, quick identification of root cause that makes defects is an essential issue. In this paper, we will refer to a typical algorithm of mining association rules and propose a novel interestingness measurement to provide an effective and accurate solution. First, the manufacturing defect detection problem of analyzing the correlation between combinations of machines and the result of defect is defined. Then, we propose an integrated processing procedure RMI (Root cause Machine Identifier) to discover the root cause in this problem. Finally, the results of experiments show the accuracy and efficiency of RMI are both well with real manufacturing cases.
A parallelized indexing method for large-scale case-based reasoning
(2002-08-01)
Chen, WC
;
Tseng, SS
;
Chang, LP
;
Hong, TP
;
Jiang, MF
Case-based reasoning (CBR) is a problem-solving methodology commonly seen in artificial intelligence. It can correctly take advantage of the situations and methods in former cases to find out suitable solutions for new problems. CBR must accurately retrieve similar prior cases for getting a good performance. In the past, many researchers proposed useful technologies to handle this problem. However, the performance of retrieving similar cases may be greatly influenced by the number of cases. In this paper, the performance issue of large-scale CBR is discussed and a parallelized indexing architecture is then proposed for efficiently retrieving similar cases in large-scale CBR. Several algorithms for implementing the proposed architecture are also described. Some experiments are made and the results show the efficiency of the proposed method. (C) 2002 Elsevier Science Ltd. All rights reserved.