Xining Li

Comparison Of Rough-Set And Interval-Set Models For Uncertain Reasoning

In the rough-set model, a set is represented by a pair of ordinary sets called the lower and upper approximations. In the interval-set model, a pair of sets is referred to as the lower and upper bounds which define a family of sets. A significant difference between these models lies in the definition and interpretation of their extended set-theoretic operators. The operators in the rough-set model are not truth-functional, while the operators in the interval-set model are truth-functional. Within the framework of possible-worlds analysis, we show that the rough-set model corresponds to the modal logic system S 5, while the interval-set model corresponds to Kleenes three-valued logic system K 3. It is argued that these two models extend set theory in the same manner as the logic systems S 5 and K 3 extend standard propositional logic. Their relationships to probabilistic reasoning are also examined.

IMAGO: A Prolog-Based System for Intelligent Mobile Agents

This paper presents the preliminary design of the IMAGO project. This project consists of two major parts: the IMAGO Application Programming Interface (API) - an agent development kit based on Prolog, and the MLVM - a multithreading agent server framework. We focus on the IMAGO API and its communication model - a novel mechanism to automatically track down agents and deliver messages in a dynamic, changing world. Examples are given to show the expressive power and simplicity of the programming interface as well as possible applications of the proposed system.

Efficient memory management in a merged heap/stack prolog machine

TraditionalProlog implementations are based on the stack/heap memory architecture: the stack holds local variables and control information, whereas the heap stores data objects which outlive procedure activations. A stack frame can be deallocated when an activation ends while heap space can only be reclaimed on backtracking or by garbage collection. Conventional garbage collection methods may yield poor performance. In this paper, I present a novel memory management approach used in the implementation of Logic Virtual Machine (LVM). The LVM combines the stack and the heap into a single memory block for all dynamical memory requirements, supports coarse-grain two-stream uni cation, and embeds an e cient garbage collection algorithm, the Chronological Garbage Collection (CGC), to reclaim useless memory cells. An experimental LVM emulator has been implemented. Our experimental results show that the proposed approach has low runtime overhead, good virtual memory and cache performance, and very short, evenly distributed pause times during garbage collection. Some benchmarks even revealed that the CGC not only improves the programs cache performance by more than enough to pay its own cost, but also improves the program execution performance which is competitive with the SICStus fastcode.

Link Failure Rate and Speed of Nodes in Wireless Network

In the simulation analysis of wireless network, the speed of nodes is an important property of mobility. The faster the nodes move, the more unstable the network is. Meanwhile, in theoretical analysis, link failure rate is widely used to modeling the stability of a wireless network. This paper presents an analysis of a simplified mobility model and shows that the link failure rate is positively correlated with the average speed of nodes in this model. Though this result is based on a mobility model with many restrictions, a simulation evaluation suggests that the result still holds in the popular random waypoint model and random direction model.

An Integrated-Model QoS-Based Graph for Web Service Recommendation

Web services (WS) are integrated software components that facilitate interoperable machine-to-machine interaction over a network. In the era of Web 2.0, companies worldwide are actively deploying Web services within their business environments. As a result, designing effective Web service recommendation mechanisms based on Quality of Service (QoS) is attracting more attention. However, traditional Neighborhood-based Collaborative Filtering (CF) models fail to capture the actual relationships between users or services due to data sparsity. On the other hand, Random Walk (RW) algorithm, which has been categorized as a sparsity-tolerant recommendation approach, suffers from poor performance in terms of recommendation accuracy. In this paper, we aim at designing a recommendation model that achieves high recommendation accuracy over the transitional RW based model. First, we propose an Integrated-Model QoS-based Graph (IMQG), in which users and services represent the nodes while weighted QoS magnitudes and User/Service similarity measurements serve as the edges. We use Jaccard coefficient in several variants to separately compute similarities of both Users and Services. Then, Top-k Random Walk algorithm is applied to generate final recommendation list to active users. Finally, to demonstrate the effectiveness of our model, comprehensive experiments are conducted on a real-world QoS dataset. Analysis of the results shows high improvement in recommendation accuracy with more tolerance to data sparsity.

Inter-agent communication in IMAGO prolog

A mobile agent application often involves a collection of agents working together for a common task. For cooperation among agents to succeed, an effective inter-agent communication framework is required. This paper describes the design of the communication mechanism in IMAGO Prolog. IMAGO Prolog is a variant of Prolog with an extended API for intelligent mobile agent applications. It deploys mobile messengers for inter-agent communication. Messengers are anonymous, thin agents dedicated to deliver messages. A messenger can move, clone, and make decisions for its assigned task: track down the receiving agent and reliably deliver messages in a dynamic, changing world. Moreover, agent communication language is purely declarative and consistent with the syntax, semantics and pragmatics of Prolog. As a result, mobile agents exchange information and achieve synchronization through first order logic terms and unification.

On the design of a mobile agent environment for context-aware M-commerce

Due to technological advances of handheld computing and communication devices, Mobile commerce (M-commerce) has emerged and attracted a growing number of research efforts. M-commerce not only extends Internet-based Electronic commerce (E-commerce), but also offers a unique business opportunity with its own features, such as ubiquity, accessibility, portability, etc. In this paper, we discuss some important issues in the design of an M-commerce platform based on agent technology and context-aware workflow analysis. Deploying mobile agents and context-awareness in M-commerce can reduce unnecessary network traffic, tolerate poor network connectivity, provide more advanced services, support automation of decision-making, reduce participation costs and improve trading efficiencies.

Implementing an Mobile Agent Platform for M-Commerce

Mobile commerce (M-commerce), the traditional E-commerce combined with mobile devices and wireless networks, is likely to become a major business model in the near future. M-commerce not only extends Internet-based E-commerce, but also offers a unique business opportunity with its own features, such as ubiquity, accessibility, portability, etc. Most E-commerce applications adopt client/server model in which a commercial transaction generally requires a stable communication connection being established between the client and the server. Obviously, this approach poses a barrier to the development of M-commerce applications. In this paper, we propose a mobile agent oriented M-commerce platform. Due to the flexibility and mobility, mobile agent based M-commerce can complement the existing client/server based E-commerce model to enable consumers to conduct business without time and space restrictions. This paper will discuss the design and implementation of an experimental mobile agent platform for M-commerce. Issues to be investigated involve security, service discovery and database access functionalities. The goal of this research is to deploy the mobile agent paradigm for M-commerce applications.

Large Scale Deployment a Mobile Agent Approach to Network Management

In this paper, we describe a possible approach for using mobile agents to manage a group of networked devices. Our system goes one step further from the traditional one-to- one device management setting by providing a one-to-many architecture able to perform simple administrative tasks on behalf of the user. We show how the use of mobile agents eases the task of maintaining, monitoring and managing a very large number of devices and consequently, reduces the work load of the people in charge of these devices.

A new term representation method for prolog

Abstract Various Prolog systems can be classified into two categories: Structure Sharing (SS) and Structure Copying (SC). The fundamental distinction between SS and SC is the way in which they represent structures. SS represents a structure instance by a two-pointer molecule with one end toward the structure skeleton and the other toward a binding environment. On the other hand, SC makes a concrete copy of a structure whenever the structure is matched against a free variable. SS was used in earlier Prolog implementations, whereas SC has been accepted as the de facto standard in modern Prolog implementations. However, analysis and practical comparison of SS and SC claim that programs can be written that make any one method almost arbitrarily worse than the other. In this paper, I propose a new Prolog term representation approach: Program Sharing (PS). The major contribution of this work is that PS has the advantages of both SC (representing terms of different types to fit in the size of a machine word) and SS (low overhead in constructing a dynamic structure instance), and the concept of program sharing could be used to realize all-special-case instruction-driven unification. PS has been adopted in the design of a new Prolog abstract machine, the LAM 1 2 . I have implemented an experimental LAM 1 2 emulator in C. Benchmarks show that this new approach is very promising in memory utilization and reasonably close to a very good SC-based system in performance.