This paper introduces a non-Horn rule WRM which is a weak form of rational monotony. We explore the effects of adding this non-Horn rule to the rules for the preferential inference. In this paper, a relation is said to be P+WRM iff it is a preferential inference and satisfies the rule WRM. We establish the representation theorem for P+WRM, and compare the strength of WRM with some non-Horn rules appearing in literatures. Moreover, we explore the relation between P+WRM and conditional logic, and demonstrate that P+WRM is equivalent to 'flat' fragment of conditional logic CS4.2. Another contribution of this paper is to explore the relation between two special kinds of preferential models, I.e., PRC model and quasi-linear model. Main result reveals that the latter is a special form of the former.

Fairness, latency and computational complexity are three important factors in evaluating the performance of a scheduling algorithm. Fairness must be satisfied so that service can be distributed according to the reserved rate. Only when latency is irrelevant to the number of connections, is it possible to minimize the end-to-end delay through controlling the reserved rate. Among existing scheduling algorithms, Round Robin is the least complex. However, conventional Round Robin is unable to ensure fairness, and the improved round robin algorithms like Deficit Round Robin, Weighted Round Robin and Virtual Round Robin are unable to ensure that their latencies are irrelevant to the number of connections although they guarantee fairness. Potential Round Robin developed for analysis of fairness and latency reduction is thus proposed. It is based on the introduction of a new concept, Round Potential Function. The function splits service time into a number of service round periods to guarantee fairness regardless of the serving process used in the period. In the analysis of latency, service round periods are re-split into multiple scanning cycles for further service distribution with approximate sorting between scanning cycles. As a result, latency is no longer relevant to the number of connections while the low complexity of round robin is kept.

In this paper, the representability of a family of theories as the set of extensions of a default theory is studied. First, a new necessary condition is given for the representability by means of general default theories; then a sufficient one is presented. The families of theories represented by default theories are also fully characterized. Finally, the paper gives an example of denumerable families of mutually inconsistent theories that are represented by a default theory but not by normal ones.

Generation of photo-realistic images of human hair is a challenging topic in computer graphics. The difficulty in solving the problem in this aspect comes mainly from the extremely large number of hairs and the high complexity of the hair shapes. Regarding to the modeling and rendering of hair-type objects, Kajiya proposed a so-called texel model for producing furry surfaces. However, Kajiya's model could be only used for the generation of short hairs. In this paper, a concise and practical approach is presented to solve the problem of rendering long hairs, and in particular the method of rendering the smooth segmental texels for the generation of long hairs is addressed.

This paper proposes a novel Chinese-English Cross-Lingual Information Retrieval (CECLIR) model PME, in which bilingual dictionary and comparable corpora are used to translate the query terms. The proximity and mutual information of the term-pairs in the Chinese and English comparable corpora are employed not only to resolve the translation ambiguities but also to perform the query expansion so as to deal with the out-of-vocabulary issues in the CECLIR. The evaluation results show that the query precision of PME algorithm is about 84.4% of the monolingual information retrieval.

In dynamic networks, links and nodes will be deleted or added regularly. It is very essential for the routing scheme to have the ability of fault-tolerance. The method to achieve such a goal is to generate more than one path for a given set of source and destination. In this paper, the idea of interval routing is used to construct a new scheme (Multi-Node Label Interval Routing scheme, or MNLIR scheme) to realize fault-tolerance. Interval routing is a space-efficient routing method for networks, but the method is static and determinative, and it cannot realize fault-tolerance. In MNLIR scheme some nodes will have more than one label, thus some pairs of destination and source will have more than one path; the pairs of nodes, which have inheritance relation, will have the shortest path. Using this character, MNLIR scheme has better overall routing performance than the former interval routing scheme, which can be proven by simulations. The common problem concerning the insertion and deletion of nodes and links is considered in this paper. So if the networks have some changes in topology, MNLIR scheme may find alternative path for certain pairs of nodes. In this way, fault-tolerance can be realized with only a little space added to store the multi-node labels.

In this paper, the relationship between argumentation and closed world reasoning for disjunctive information is studied. In particular, the authors propose a simple and intuitive generalization of the closed world assumption (CWA) for general disjunctive deductive databases (with default negation). This semantics, called DCWA, allows a natural argumentation-based interpretation and can be used to represent reasoning for disjunctive information. We compare DCWA with GCWA and prove that DCWA extends Minker's GCWA to the class of disjunctive databases with default negation. Also we compare our semantics with some related approaches. In addition, the computational complexity of DCWA is investigated.

The paper briefly reviews NP-hard optimization problems and their inapproximability. The hardness of solving CLIQUE problem is specifically discussed. A dynamic-programming algorithm and its improved version for CLIQUE are reviewed and some additional analysis is presented. The analysis implies that the improved algorithm, HEWN (hierarchical edge-weighted network), only provides a heuristic or useful method, but cannot be called a polynomial algorithm.

Mining sequential patterns from large databases has been recognized by many researchers as an attractive task of data mining and knowledge discovery. Previous algorithms scan the databases for many times, which is often unendurable due to the very large amount of databases. In this paper, the authors introduce an effective algorithm for mining sequential patterns from large databases. In the algorithm, the original database is not used at all for counting the support of sequences after the first pass. Rather, a tidlist structure generated in the previous pass is employed for the purpose based on set intersection operations, avoiding the multiple scans of the databases.