<link rel="stylesheet" href="styles.cf390c6bfbe7555f.css">

學術出版;;Publications

Permanent URI for this communityhdl:11536/1

收集所有交通大學研究人員與學生個人或團體所發表之各項學術著作，涵蓋期刊論文、教師專書、學位論文等項。;;Collections of Publications of NCTU Researchers and Students Like Journal Papers and Thesis.

Browse

Filters

1996 - 1999122000 - 200632

Settings

🏛️國立陽明交通大學機構典藏

Access type: Metadata onlyAuthor: search.filters.author.Tseng, SS

Search Results

Now showing 1 - 10 of 44
  • Thumbnail Image
    Item type:Item,
    PPD: A practical parallel loop detector for parallelizing compilers on multiprocessor systems
    (1996-11-01)
    Yang, CT
    ;
    Wu, CT
    ;
    Tseng, SS
    ;
    資訊工程學系
    ;
    Department of Computer Science
    It is well known that extracting parallel loops plays a significant role in designing parallelizing compilers. The execution efficiency of a loop is enhanced when the loop can be executed in parallel or partial parallel, like a DOALL or DOACROSS loop. This paper reports on the practical parallelism detector (PPD) that is implemented in PFPC (a portable FORTRAN parallelizing compiler running on OSF/1) at NCTU to concentrate on finding the parallelism available in loops. The PPD can extract the potential DOALL and DOACROSS loops in a program by invoking a combination of the ZIV test and the I test for verifying array subscripts. Furthermore, if DOACROSS loops are available, an optimization of synchronization statement is made. Experimental results show that PPD is more reliable and accurate than previous approaches.
  • Thumbnail Image
    Item type:Item,
    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.
  • Thumbnail Image
    Item type:Item,
    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.
  • Thumbnail Image
    Item type:Item,
    Primal-dual version spaces for acquisition of disjunctive concepts
    (1998-06-01)
    Hong, TP
    ;
    Tseng, SS
    ;
    資訊工程學系
    ;
    Department of Computer Science
    Among incremental learning strategies, the "version space" learning strategy is one of the most well known. This learning strategy is, however, applicable only to learning conjunctive concepts. When the concepts to be learned are in disjunctive form, the version space learning strategy returns a null Version space that cannot correctly represent the desired concepts. In this paper, we present a modification of the original version space strategy that enables learning of disjunctive concepts. The new proposed version-space-based learning strategy, called the "primal-dual version-spaces" learning strategy, learns disjunctive concepts incrementally and without saving past training instances. The correctness of its underlying algorithm is analyzed and proven.
  • Thumbnail Image
    Item type:Item,
    Genetic algorithm for extended cell assignment problem in wireless ATM network
    (2000)
    Din, DR
    ;
    Tseng, SS
    ;
    Jiang, MF
    ;
    資訊工程學系
    ;
    Department of Computer Science
    In this paper, we investigate the extended cell assignment problem which optimally assigns new and split cells in PCS (Personal Communication Service) to switches in a wireless ATM network. Given cells and switches in an ATM network (whose locations are fixed and known), the problem is assigning cells to switches in an optimum manner. We would like to do the assignment in as attempt to minimize a cost criterion. 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 modeled as a complex integer programming problem and finding an optimal solution to this problem is NP-complete. A stochastic search method, based on a genetic approach is proposed to solve this problem. Simulation results show that genetic algorithm is robust for this problem.
  • Thumbnail Image
    Item type:Item,
    Personal rules generation for intelligent e-mail service system
    (2000-06-01)
    Tsai, CJ
    ;
    Tseng, SS
    ;
    Cheng, HT
    ;
    資訊工程學系
    ;
    Department of Computer Science
    As E-mail becomes more popular over network, the problem of receiving a lot of undesired E-mails arise. All of the previous systems, which have been developed to solve this problem, may be divided into two different categories, client-side, and server-side, from the view of location of filter. The former case always causes waste of network traffic and the latter case results that ail users may only share a small predefined set of filtering rules. In this paper, we propose a new architecture of E-mail service system, called Intelligent E-mail Service System (IESS), integrating client-side and server-side to help users manage their E-mails. A data mining approach is used in client to find managing rules from user's behavior of reading E-mail automatically. Then these rules would be stored in rule base for predicating E-mail at server. The prediction of the E-mail is made before transferring it to client; therefore, the undesired E-mails can be filtered at server to save network traffic. In addition, the feedback mechanism is provided to refine the prediction accuracy based on each user's decisions for reading E-mails. Experimental results show that IESS can infer E-mail reading preference of individual users correctly up to the accuracy about 80%. We may conclude that IESS can help users manage their E-mails effectively.
  • Thumbnail Image
    Item type:Item,
    Data types generalization and transformation for data mining algorithms
    (2000-06-01)
    Jiang, MF
    ;
    Tseng, SS
    ;
    Liao, SY
    ;
    資訊工程學系
    ;
    Department of Computer Science
    In most applications, knowing the characteristics of the data mining methods is usually the key factor to determine the kind of data mining methods should be applied. In this work, we first survey and analyze six kinds of data mining algorithms and four kinds of data forms, and then find out the relations between the mining algorithms and the data types. According to the relations, the suitable generalized data types will be created and is then used to transform the generalized data types of data sources to the suitable one for the selected mining algorithm. With the proposed preprocessing work for data mining algorithms, users can select appropriate mining algorithm just for the goal of application without considering the data types. Finally. two applications for e-mail management and e-mail log analysis are proposed to show the practicality of our method.
  • Thumbnail Image
    Item type:Item,
    A simulated annealing algorithm for extended cell assignment problem in a wireless ATM network
    (2001)
    Din, DR
    ;
    Tseng, SS
    ;
    資訊工程學系
    ;
    Department of Computer Science
    In this paper, we investigate the extended cell assignment problem which optimally assigns new adding and splitting cells in PCS (Personal Communication Service) to switches in a wireless ATM (Asynchronous Transfer Mode) network. Given cells in a PCS and switches on an ATM network (whose locations are fixed and known), we would like to do the assignment in an attempt to minimize a cost criterion. 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 modeled as a complex integer programming problem, and finding an optimal solution to this problem is NP-hard. A simulated annealing algorithm are proposed to solve this problem. The simulated annealing algorithm, ESA (enhanced simulated annealing), generates constraint-satisfy configurations, and uses three configuration perturbation schemes to change current configuration to a new one. Experimental results indicate that ESA algorithm has good performances.
  • Thumbnail Image
    Item type:Item,
    A two-phase fuzzy mining and learning algorithm for adaptive learning environment
    (2001)
    Tsai, CJ
    ;
    Tseng, SS
    ;
    Lin, CY
    ;
    資訊工程學系
    ;
    Department of Computer Science
    As computer-assisted instruction environment becomes more popular over the world, the analysis of historical learning records of students becomes more important. In this work, we propose a Two-Phase Fuzzy Mining and Learning Algorithm, integrating data mining algorithm, fuzzy set theory, and look ahead mechanism, to find the embedded information, which can be provided to teachers for further analyzing, refining or reorganizing the teaching materials and tests, from historical learning records.
  • Thumbnail Image
    Item type:Item,
    A solution model of integrating cells of PCS to switches in wireless ATM network
    (2002-08-01)
    Din, DR
    ;
    Tseng, SS
    ;
    資訊工程學系
    ;
    Department of Computer Science
    In this paper, we investigate the optimal assignment problem of cells in PCS (Personal Communication Service) to switches on a ATM (Asynchronous Transfer Mode) network. Given cells and switches on an ATM network (whose locations are fixed and known), the problem is to group cells into clusters and assign these clusters to switches in an optimum manner. This problem is modeled as a complex integer programming problem. Since finding an optimal solution of this problem is NP-hard, a heuristic solution model consists of three phases (Cell Pre-Partitioning Phase, Cell Exchanging Phase, and Cell Migrating Phase) is proposed. Experimental results show that Cell Exchanging and Cell Migrating Phases can really reduce total cost near 44% on average.