Abstract:Uncertainty exists widely in the subjective and objective world. In all kinds of uncertainty, randomness and fuzziness are the most important and fundamental. In this paper, the relationship between randomness and fuzziness is discussed. Uncertain states and their changes can be measured by entropy and hyper-entropy respectively. Taken advantage of entropy and hyper-entropy, the uncertainty of chaos, fractal and complex networks by their various evolution and differentiation are further studied. A simple and effective way is proposed to simulate the uncertainty by means of knowledge representation which provides a basis for the automation of both logic and image thinking with uncertainty. The AI (artificial intelligence) with uncertainty is a new cross-discipline, which covers computer science, physics, mathematics, brain science, psychology, cognitive science, biology and philosophy, and results in the automation of representation, process and thinking for uncertain information and knowledge.

Abstract:Aiming at the requirements specification and related checking of embedded real-time software, a visual modeling language, RTRSM* (real-time requirements specification model*), which is compositional and based on hierarchical and concurrent finite state machine, is proposed. It uses state transitions with duration and scheduled events to describe timing constraints, and can support the description of interactivity and timing constraints effectively. Additionally, RITL (real-time interval temporal logic), a kind of prepositional temporal logic, is presented to make up for RTRSM*’s defect description of global system properties, which is the drawback of operational specification languages. Interpreted over timed state sequences, RITL is able to deal with the description of both point-based and interval-based metric temporal properties, and supports the property description of RTRSM* models naturally and comprehensively. The verification and validation of the resulted requirements specification, especially issues with respect to the reachability graph of RTRSM* models with finite system states and the simulation execution of the specification, are also explored.

Abstract:In this paper, a facial feature point tracking scheme is proposed by integrating Lucas-Kanade optical flow tracking algorithm and the face alignment statistical model, DAM (direct appearance model), together in a Bayesian framework. The prediction of feature positions from Lucas-Kanade algorithm exploits the inter-frame correlations and accelerates the tracking speed. The texture-shape constraint under DAM improves the localization accuracy and robustness. Experiments show that this method adapts well to the various face movements. It can be used in face recognition or 3D face modeling.

Abstract:One of the interferences between inheritance and concurrency is inheritance anomaly. From the view of cognitive computational neuroscience, a direct information representation method is presented based on neural system dynamics and graphic theory. A group of neurons and their connections representing perceptual information directly and the dynamical behaviors of neurons are defined firstly, and then a two-layer neural network is designed to record characteristics of stimulus and connect a specialized neural circuit that responding to the perception of that stimulus respectively. This could be achieved by the structure learning algorithm. The circuit constituted by neurons in two layers is also served as an associative memory of stimulus whose credibility is decided by the degree of connection of the circuit. The direct representation method is of very significance to the research of semantic representation and inference driven by semantics in artificial intelligence.

Abstract:The MIPS (model checking integrated planning system) has shown distinguished performance in the second and the third international planning competitions. In this paper, a declarative approach to adding domain knowledge in MIPS is presented. And DCIPS (domain constraints integrated planning system) has been developed according to this method. DCIPS allows different types of domain control knowledge such as objective, procedural or temporal knowledge to be represented and exploited in parallel, thus combining the ideas of ‘lanning = actions + states’into domain control knowledge. An advantage of this approach is that the domain control knowledge can be modularly formalized and added to the planning problem as desired. DCIPS is experimentally verified on the three examples in the transportation domain from AIPS 2002 planning competition where it leads to significant speed-ups.

Abstract:This paper addresses the semi-structured query rewriting problem for TSL (tree specification language), a language for querying semi-structured data. An algorithm that can find the maximally-contained rewriting query is presented, when a semi-structured query and a set of semi-structured views are given. The idea is borrowed from MiniCon, a scalable relational query rewriting algorithm, and some new problems for semi-structured query rewriting, e.g., object-id dependency and set value variable mapping, are solved. It is shown that the algorithm is correct.

Abstract:XQuery is the recommended standard for XML Query. XQuery processing strategies can be classified into two categories: core syntax based strategy (node-oriented) and algebra based strategy (set-oriented). Neither of them can handle XQuery well. The syntax based strategy is inefficient and hard to optimize, while the current algebra based strategies can not satisfy the flexible programming characteristics of XQuery. After summarizing the current stage and unsolved problems of former algebra based works, an effective XQuery algebra system, OrientXA, is proposed in this paper, ideas from both strategies are embodied in it. OrientXA introduces the notion of Construct Pattern Tree for the first time. The Construct operator in it materializes the flexible characteristics of XQuery. Corresponding to its expressive operators, it is able to express all the queries in W3C use cases and XMark benchmark.

Abstract:Set type is an important data type in object-oriented database system and object-relational database system. An index structure of set type Set_struc is presented in this paper. In Set_struc all sets are organized as a tree, and the sets with common prefix are merged. So the size of the index will be decreased for the data set with a large number of repeated data and frequent patterns. Based on Set_struc, an algorithm of join operation with Set_struc is presented. Its performance is better than other methods such as PSJ (partition based join).

Abstract:Segmentation of time series is one of the important tasks in time series data mining. Segmentation has two major uses: It may be performed either to detect when the system that creates the time series has changed or to create a high level representation of the time series for indexing, clustering, and classification. Approaches to on-line segmentation of time series are necessary when identifying and predicting temporal patterns in real-time time series databases are needed, and this is the focus of this paper. A formal description of segmenting time series problem and a criterion for the evolution of segmentation algorithms are presented. An on-line iterative algorithm of segmenting time series, called OLS (on-line segmentation), is then proposed. OLS is independent of a priori knowledge about the segmented time series. Experimental results demonstrate that OLS can on-line detect the critical change points of time series with less 憃ver fit?than that of competitive algorithms.

Abstract:Secret protection is studied after introducing the quantifying permission idea in the view of access control. Meta-Permission is derived from the view of ‘uality’and‘uantity’in philosophy. Compared with traditional cognition and permission used in access control, it comprehensively and deeply reflects the essence of permission. Combining with the threshold idea and RBAC, a threshold access control scheme based on quantifying permission is proposed. This scheme is applicable as widely as secret-sharing scheme, and shares the same function in protecting secret. Furthermore, it has some distinct advantages such as no relation in knowledge between the secret pieces hold by participants and the protective secret target, the ability for expressing the difference among participants?trustworthiness, and the low complexity in computing.

Abstract:With the rapid increment of network bandwidth, the overheads of protocol on host have become the bottleneck of system performance. A lot of work including optimization for protocol mechanism, implementation and architecture have been developed in the fields of protocol acceleration. After analyzing overhead and performance of the protocols, it is attempted to survey the developments of TCP (transfer control protocol) accelerating on mechanism, implementation and architecture. Finally, some future directions are discussed.

Abstract:MLS (multilevel security) is being widely applied in many security critical systems, but it can’t implement many important security policies such as ‘channel-control’. In this paper, the concept of trust degree is introduced into the MLS to implement policies like ‘channel-control’ conveniently. An access control state machine model which enforces the trust degree based multilevel security policy is established, and is proved to be secure for this policy. It is also proved that this model can enforce all static information flow policies. An extension of the model is also offered to support the dynamic change of storage objects’ security labels. The model avoids the disadvantage of MLS’ not being able to resolve the problem of secure downgrading and not taking integrity into consideration, and at the same time it retains the advantage of easy understanding and use enjoyed by the traditional classified policy models.

Abstract:With the development of molecular marker technique and optical imaging technique, In Vivo bioluminescent imaging attracts more and more attention and is extensively used to non-invasively visualize the physiological and pathological process of biological tissues in real time. The photon propagation model and photon propagation rule are two key components of bioluminescent imaging. In this paper, a photon propagation model of bioluminescent imaging based on Monte Carlo method is proposed. With known quantities (e.g. parameters of bioluminescent source, biological tissue, and CCD detectors) and Monte Carlo technique, two mathematic models including photon generation from bioluminescent sources and photon propagation in biological tissue are realized. Then, several comparative experiments are performed and experimental results are given, which illustrates the correctness and efficiency of this algorithm.

Abstract:In a distributed cooperative learning setting, learners and tutors need a shared information space to interact with learning materials and to acquire awareness of their cooperators’ presence and activities. Learners and tutors also require some coordination and communication means to harmonize their actions. For this purpose, a cooperative visual hypermedia technology is developed. It can provide a shared information space in which editable visual hypermedia objects can represent not only learning materials (i.e. the content) but also learning processes (i.e. the coordination and learning strategies). The cooperative hypermedia is integrated with process support, communication channels, and document handling. Therefore, it can coordinate learners’ activities through specific learning processes and textual or audio communication tools. For wide availability, the cooperative hypermedia and its associated documents are made accessible from the Web. To show the strength and the weakness of the approach, a use case together with some user experience is also presented.

Abstract:The Audio/Video Stream Specification was brought forward by OMG (Object Management Group) based on the extension to traditional CORBA (common object request broker architecture) architecture. Multimedia applications could get more benefits, such as open and standard, from the distributed object computing brought by this Specification. However, there is not a whole scheme in the specification to solve the security problem of distributed multimedia applications. Because of this immature security, a security stream model―SecStream is presented, which brings the CORBA security service and the data security service into the old stream architecture. Subsequently, design and implementation of the above model are discussed. Finally, the SecStream model is successfully applied to an application, where the flexibility and extensibility of this model are verified.

Abstract:In this paper, the 2004 year’s application for funding and supporting General Program by Computer Science Division of Information Science Department of National Natural Science Foundation of China (NSFC) are summarized. Change in number of proposals and fund are given. Reviewing principles are provided.