It has been shown that focusing the training algorithms to the decision boundary vicinity data can improve the accuracy of several classification methods. However, previous approaches for fining decision boundary vicinity data are either computationally tedious or may perform poorly in handling problems with class overlapping. With the application of the nearest neighbor rule, this work proposes a new criterion to characterize the nearness of the training samples to the decision boundary. To demonstrate the effectiveness of the proposed approach, the proposed method is integrated with a nearest neighbor classifier design method and a neural work training approach. Experimental results show that the proposed method can reduce the size and classification error for both of the tested classifiers.

Abstract Council For Economic And Development Executive Yuan established a new land-use scheme in National Comprehensive Development Plan on the purpose of preventing the resource depletion and environmental disasters resulting from improper land-use activities. In the land-use scheme the land resources are identified as ¡¥developable areas¡¦ and ¡¥restricted areas¡¦ according to their importance and fragility. The restricted areas are imposed strict land-use regulations for the purposes of resources conservation and preservation. Thus, this research worked on the bases of the regulated¡¦s opinions. This research starts with discussing the regulation policy theory, the literature on the theory of property rights and a conceptual framework for determining whether the invaded land property rights should be compensated, and for suggesting how to compensate if the compensation is required. Finally, this research applied the framework to the Shoushen National Park to discuss whether the land property rights invaded by the regulations should be compensated and what kind of compensation would be appropriated.

With the rapidly growing volumes of data from various sources, new tools and computational theories are required to extract useful information (knowledge) from large databases. Data mining techniques such as association rules have been proved to be effective in searching hidden knowledge in a large database. However, if we want to extract knowledge from data with temporal components, it becomes necessary to incorporate temporal semantics with the traditional data mining techniques. As mining techniques evolves, mathematical techniques become more involved to help improve the quality and diversity of mining. Fuzzy theory is one that has been adopted for this purpose. Up to now, many approaches have been proposed to discover temporal association rules or fuzzy association rules, respectively. However, no work is contributed on mining fuzzy temporal patterns. We propose in this thesis two data mining systems for discovering fuzzy temporal association rules and fuzzy periodic association rules, respectively. The mined patterns are expressed in fuzzy temporal and periodic association rules which satisfy the temporal requirements specified by the user. Temporal requirements specified by human beings tend to be ill-defined or uncertain. To deal with this kind of uncertainty, a fuzzy calendar algebra is developed to allow users to describe desired temporal requirements in fuzzy calendars easily and naturally. Moreover, the fuzzy calendar algebra helps the construction of desired time intervals in which interesting patterns are discovered and presented in terms of fuzzy temporal and periodic association rules. In our system of mining fuzzy temporal association rules, a border-based mining algorithm is proposed to find association rules incrementally. By keeping useful information of the database in a border, candidate itemsets can be computed in an efficient way. Updating of the discovered knowledge due to addition and deletion of transactions can also be done efficiently. The kept information can be used to help save the work of counting and unnecessary scans over the updated database can be avoided. Simulation results show the effectiveness of the proposed system for mining fuzzy temporal association rules. In our mining system for discovering fuzzy periodic association rules, we develop techniques for discovering patterns with periodicity. Patterns with periodicity are those that occur at regular time intervals, and therefore there are two aspects to the problem: finding the pattern, and determining the periodicity. The difficulty of the task lies in the problem of discovering these regular time intervals, i.e., the periodicity. Periodicites in the database are usually not very precise and have disturbances, and might occur at time intervals in multiple time granularities. To discover the patterns with fuzzy periodicity, we utilize the information of crisp periodic patterns to obtain a lower bound for generating candidate itemsets with fuzzy periodicities. Experimental results have shown that our system is effective in discovering fuzzy periodic association rules.

The clustering problem has been discussed extensively in the database literature as a tool for many applications, for example, bioinformatics. Traditional clustering algorithms consider all of the dimensions of an input dataset in an attempt to learn as much as possible about each object described. In the high dimensional data, however, many of the dimensions are often irrelevant. Therefore, projected clustering is proposed. A projected cluster is a subset C of data points together with a subset D of dimensions such that the points in C are closely clustered in the subspace of dimensions D. There have been many algorithms proposed to find the projected cluster. Most of them can be divided into three kinds of classification: partitioning, density-based, and hierarchical. The DOC algorithm is one of well-known density-based algorithms for projected clustering. It uses a Monte Carlo algorithm for iteratively computing projected clusters, and proposes a formula to calculate the quality of cluster. The FPC algorithm is an extended version of the DOC algorithm, it uses the mining large itemsets approach to find the dimensions of projected cluster. Finding the large itemsets is the main goal of mining association rules, where a large itemset is a combination of items whose appearing times in the dataset is greater than a given threshold. Although the FPC algorithm has used the technique of mining large itemsets to speed up finding projected clusters, it still needs many user-specified parameters to work. Moreover, in the first step, to choose the medoid, the FPC algorithm applies a random approach for several times to get the medoid, which takes long time and may still find a bad medoid. Furthermore, the way to calculate the quality of a cluster can be considered in more details, if we take the weight of dimensions into consideration. Therefore, in this thesis, we propose an algorithm which improves those disadvantages. First, we observe that the relationship between parameters, and propose a parameter-relationship-based algorithm that needs only two parameters, instead of three parameters in most of projected clustering algorithms. Next, our algorithm chooses the medoid with the median, we choose the medoid only one time and the quality of our cluster is better than that in the FPC algorithm. Finally, our quality measure formula considers the weight of each dimension of the cluster, and gives different values according to the times of occurrences of dimensions. This formula makes the quality of projected clustering based on our algorithm better than that of the FPC algorithm. It avoids the cluster containing too many irrelevant dimensions. From our simulation results, we show that our algorithm is better than the FPC algorithm, in term of the execution time and the quality of clustering.

This research is to discuss the Chinese pre-Qin legalist school, Shang Yang's political thought. First, the purposes of the research are as follows: 1.To understand the background of Shang Yang¡¦s political thought and analyze the foundation of his political thought. 2.To discover the principles of Shang Yang¡¦s political thought. 3.From the principles of Shang Yang¡¦s political thought, analyze his political objectives. 4.To analyze Shang Yang¡¦s actual policies of agriculture and fight. 5.To sum up this research results to comment on Shang Yang's political thought. Therefore, based on the purposes of the research, the principles of Shang Yang¡¦s political thought is rule by law, attempt to achieve the powerful nation. The concrete method is the policies of the agriculture and fight. So, the following are the results of the analysis of this research: 1.Shang Yang advocates ¡¥rule by law.¡¦ 2.Shang Yang promotes anti-wisdom. 3.Shang Yang advocates despotism. 4.Shang Yang contends that clear reward and hard punishment can make the monarch rule smoothly. 5.Shang Yang proposes that the monarch rules by the system, rather than by man.

There should be more interpretations which are derived from data, presented by those professional analysts. The empirical rules and knowledge do help as making statistical inference in Econometrics. The approaches from classical statistical analysis make judges simply resulting from historical data. To be frank, the advantage of this analysis is the objectivity, but there is a fatal drawback. That is, it does not pay attention to some logically extra information. This paper is born for the applications of Bayesian, which has the essential characteristic of accepting subjective outlook, applying empirical rules to study unit root test on exchange rate market. Furthermore, the various distributions of data may have direct effect on the classical statistical inference we use, such as Dickey-Fuller and Phillips-Perron test. To take those defects into consideration, this paper tends not to take the assumption of disturbances in normal distribution as granted. For instance, it is quite common for us to confront the heavy-tailed distribution when studying some data of time series related to stocks and targets of investment. Hence, we will apply more generalized model to do research on Bayesian unit root test. Use the model of Schotman and Van Dijk (1991) and assuming disturbance shaped as independent student-t distribution to revise the unit root test, next, applying to exchange rate market. This is the motif of this paper.

The development of Taiwanese folk songs comprises not only cultural implication of Taiwan, but also the reflection of social phenomenon. The research of Taiwanese folk songs is the same as Taiwan history research. Through the meaning of those songs, we can comprehend the different ideologies and policies between each authority. In the research, we also can understand there are a lot of difficulties for Taiwanese in pursuing the democracy and seeking the belonging and identification. This research analyzed the Taiwanese folk songs with the political history in the background. There are various types of Taiwanese folk song in the history. For the research, I separate it into 3 periods: the Japanese colonization period (1895-1945), the Martial law period (1946-1986), and the rescind martial law up to the party turn-over period (1987-2000). This essay aims to understand the relevance between the Taiwanese folk songs and the state ruling, including constitutional order and policies. We also try to account for the effect of the state ruling as well as the surveillance and pacification on the society. Following this proposition, this studying can help us to understand the difference between the various regimes¡¦ organization and the cultural policies. What is the figuration of the development of Taiwan society? How the state affects the lives of people? How people question the legitimacy of state and expression their challenge in folk songs?

Rule induction is a popular technique for knowledge acquisition and data mining. Many techniques, such as ID3, C4.5, CART (tree induction tecniques) and Artificial Neural Networks have been developed and widely used. However, most techniques are either based on categorical or numerical mechanisms to assess the importance of different input variables, which may not produce the optimal rule when a mixture of variables exists. In 1992, Liang proposed a composite approach called CRIS that use different method to analyze different types of data in inducing rules for binary classification. Yang conducted a follow-up research to extend the original algorithm to multiple categories. However, both methods do not take variable interaction into consideration. The purpose of this research is to extend previous approach and extend by including second-order interaction. We also take into consideration the kurtosis and skewness of data for numerical variables. For categorical data, we also adopt ID3 algorithm to handle classes with low representation in the sample. In order to evaluate this technique, we develop a prototype CRIS 3.0 and compare with existing techniques, including multi-category-CRIS, CART and C4.5 as benchmark. The results show that CRIS 3.0 has the highest probability of producing the highest prediction accuracy.

none

Abstract After the martial law was repealed, Taiwan has been embarked on the transformative process of political democratization and economic liberalization. As the first party turn-over in 2000 election changed radically the strategic environment of social welfare resource distribution, especially impacted the welfare for civil service and teachers, it is worthy of our reexploration furtherly on the status quo of Taiwan¡¦s social welfare resource distribution after democratization. Taking civil service and teachers for example, the author comprehensively worked up, compared and analyzed the allocational transition of social welfare resources respectively in the period of authoritarian rule and the transformative period of democratization, then presented policy and suggestion of social welfare resource distribution for civil service and teachers after democratization. The study surveyed the approaches of allocating social welfare for civil service and teachers by the research methods of analyzing the ideology of each party and the aspect of benefit transference by welfare policy, as well as expounding the supply side of government¡¦s welfare expenditure, then putting all the data above analyzed under the environmental veins of politics, economics and society so as to present an integrated skeleton of research. Moreover, the author chose 8 years before and after democratic transformation in Taiwan as research cases to engage in comparing and analyzing the social welfare resource distribution for civil service and teachers, i.e. the periods of DPP ruling (2000~2008) and KMT ruling (1994~1999). The research found that the social welfare resources laid particular stress on civil service and teachers in the period of authoritarian rule which possessed a savor of favor and grace. Secondly, there were full of variables and uncertainty in reformative policy of social welfare resource distribution for civil service and teachers during transformational period of democratization, but it would only turn the issue of welfare-alloctation to become an instrument among parties for exchanging political loyalty and consolidating their regime. Futhermore, while observing the situation before and after the transformative period of democratization, there was no significant difference in the expenditure items of welfares for civil service and teachers, but the allocation and beneficiary of social welfare tended to be generalized. The thesis suggested that the social welfare resource distribution should follow the principles of democracy and the rule of law, and build up the concepts of legalized welfare as swift as possible, i.e. such issue should be regulated, supervised and controlled by public opinions, then would the reasonable reform of welfares for civil service and teachers would be fulfilled gradually. Key words: Social Welfare Resource Distribution, Civil Service and teachers, Authoritarian Rule, Democratization