基于贝叶斯网络模型的用户兴趣联合推送

The association push of interesting information is implemented according to the information that other users with similar interests have read, the refore user can easily find high quality interesting information from the immense Web information. The association push of interesting information based on Bayesian network model ispresented in this paper, it stores the conditional probability and semantic meaning between the terms, so the similarity between the users' interests and the similarity between the user's interest and document can be computed according to the semantic meaning of the terms, therefore the association push of interesting information can beimplemented. The experience result indicates this method is effective.     

自适应网络学习用户界面设计和实现

An adaptive user interface helps to improve the quality of human-computer interaction. Most people at present join to Web-Based Learning by common browser. Due to the one-fits-all user interface, they have to face with the problem of lack of the support on personalized learning. The design and implementation of the adaptive user interface for Web-based learning in this paper is grounded in our work done before, for example interaction model,adaptive user models including domain models. The adaptivity is mainly expressed on learning contents and representation including layout as well as operation.     

Web search engine: Characteristics of user behaviors and their implication

In this paper, first studied are the distribution characteristics of user behaviors based on log data from a massive web search engine. Analysis shows that stochastic distribution of user queries accords with the characteristics of power-law function and exhibits strong similarity, and the user' s queries and clicked URLs present dramatic locality, which implies that query cache and 'hot click' cache can be employed to improve system performance. Then three typical cache replacement policies are compared, including LRU, FIFO, and LFU with attenuation. In addition, the distribution character-istics of web information are also analyzed, which demonstrates that the link popularity and replica pop-ularity of a URL have positive influence on its importance. Finally, variance between the link popularity and user popularity, and variance between replica popularity and user popularity are analyzed, which give us some important insight that helps us improve the ranking algorithms in a search engine.     

Applying RBF network to predict location in mobile network

In mobile network, quality of service （Qos） is difficultly guaranteed for the particularity of mobile network. If the system knows, prior to the mobile subscriber movement, the exact trajectory it will follow, the Qos can be guaranteed. Thus, location prediction is the key issue to provide quality of service to mobile subscriber. In the present paper, RBF Network of Neural Network techniques were used to predict the mobile user＇s next location based on his current location as well as time. The software Matlab 6.5 was used to confirm the parameters of RBF network, and to same training data, makes the detailed contrast with resilient propagation BP and BP in learning time and steps of learning. Experiment results show that predicted locations with RBF are more effective and accurate than resilient BP.     

A Recommendation System Keeping Both Precision and Recall by Extraction of Uninteresting Information

A recommendation system which recommends interesting information to the target user must guarantee high precision and recall. However, there is trade-off between precision and recall. In this paper, we propose a web page recommendation method balancing both of them by take advantage of uninteresting information. The proposed method extracts the interest and uninterest indicators from not only historical interesting web pages but also uninteresting ones in a target genre. The historical interesting and uninteresting information is derived based on the browsing time and bookmarking. The proposed method can keep precision and recall by excluding the uninteresting information from the recommended ones based on the interest and uninterest indicators. The experimental result proved that the proposed method can improve the precision and recall than an existing method.     

An efficient key assignment scheme based on one-way hash function in a user hierarchy

In order to solve the problems resulted from dynamic access control in a user hier-archy,a cryptographic key assignment scheme is proposed by Prof.Lin to promote the performing ability and to simplify the procedure,However,it may cause the security in danger as the user changes his secret key;besides,some secret keys may be disclosed due to the unsuitable selection of the security classes‘‘‘‘‘‘‘‘ identities.Through setting up a one-way hash function onto Lin‘‘‘‘‘‘‘‘s schem,the propsed modification can greatly improve the security of Lin‘‘‘‘‘‘‘‘s scheme.     

用于计算机教育的Android实验软件(英文)

In this paper,we address the need for better hands-on learning materials in computing education and the need for integrating mobile computing education into computing curricular.We propose an innovative learning approach that uses state-of-the-art mobile computing technologies and devices to design a highly accessible and real-world relevant hands-on labware for computing education.The labware employs a modular design such that the learning materials can both be combined together into a single course of mobile program development and be integrated into the related computing courses as building blocks.The labware will help students understand the concepts,improve their problem solving skills,and prepare them for the mobile computing industrial workforce.The pilot modules of the labware have been presented to undergraduate students and have received positive feedbacks.     

Execution System for User Programs in Kernel Mode

We developed a system that allows user programs to run in kernel mode. In our proposed system, source codes of the user programs require no modifications to be executed in kernel mode. Furthermore, developers can improve user programs in kernel mode to access kernel resources directly. Developers can therefore focus on such improvements, because other parts of the source code can be left untouched. In this paper, we present a design and implementation of our system and show how the cp command was accelerated with our system. We modified the cp command to call a function that copies buffer caches directly between files. This modification requires only a few lines, and the experimental results show that the new cp reduces the time required to copy files on the memory file system by about 26%.     

A configuration deactivation algorithm for boosting probabilistic roadmap planning of robots

We present a method to improve the execution time used to build the roadmap in probabilistic roadmap planners. Our method intelligently deactivates some of the configurations during the learning phase and allows the planner to concentrate on those configurations that are most likely going to be useful when building the roadmap. The method can be used with many of the existing sampling algorithms. We ran tests with four simulated robot problems typical in robotics literature. The sampling methods applied were purely random, using Halton numbers, Gaussian distribution, and bridge test technique. In our tests, the deactivation method clearly improved the execution times. Compared with pure random selections, the deactivation method also significantly decreased the size of the roadmap, which is a useful property to simplify roadmap planning tasks.     

A Novel Intermediary Framework for Dynamic Edge Service Composition

Multimedia content,user mobility and heterogeneous client devices require novel systems that are able to support ubiquitous access to the Web resources.In this scenario,solutions that combine flexibility,efficiency and scalability in offering edge services for ubiquitous access are needed.We propose an original intermediary framework,namely Scalable Intermediary Software Infrastructure (SISI),which is able to dynamically compose edge services on the basis of user preferences and device characteristics.The SISI framework exploits a per-user profiling mechanism,where each user can initially set his/her personal preferences through a simple Web interface,and the system is then able to compose at run-time the necessary components.The basic framework can be enriched through new edge services that can be easily implemented through a programming model based on APIs and internal functions.Our experiments demonstrate that flexibility and edge service composition do not affect the system performance.We show that this framework is able to chain multiple edge services and to guarantee stable performance.     

敏捷供需链管理系统中的代理(Agent)技术

敏捷供需链是一个受竞争，合作，动态的市场环境影响折不断变化的动态网络，Agent是一个处于特定环境中，能感知并灵活，自主适应环境，可以代理设计者或使用者实现一定任务的封装的计算机实体或程序，Agent技术能较好地解决敏捷供需链环境不可预测，持续变化的问题。     

一个智能用户接口Agent设计与实现

文章主要介绍了DOLTRI-Agent(distance and open learning training resource information retrieval agent)系统中的用户接口Agent(NanDa user interface agent, 简称NDUIA)的设计和实现的关键技术.此系统扩展了memory-based reasoning技术,采用了多个记忆模型和多个分析模型,通过对不同用户使用经验的分析产生该用户专用的用户兴趣模型;同时根据用户兴趣模型和特定场景的使用经验共同作用来提供主动的智能服务,包括信息导引、搜索结果的预处理、智能即时帮助和分类信息的修改等,从而实现软件与人的协作.     

基于移动Agent的网络智能学习系统研究

分析现有网络学习系统的一些不足,以建构主义学习理论为指导,提出把移动Agent技术和网络学习相结合的思想,构建一个基于移动Agent的网络学习系统.系统的设计思想采用"双主模式",为学习者提供了个性化的学习环境,系统的实现采用移动Agent技术,既充分发挥了现有的网络功能,又充分利用现有地理上分布的、功能上独立的各个网络资源,为学习者提供全面而准确的智能化服务,改善了学习效果.     

基于Agent的用户偏好决策模型及其应用

智能体及其组成的智能体系统是人们解决用传统方法难以解决的复杂问题的一个新途径,以复杂适应系统、智能体系统等理论为指导,吸取了智能体强化学习的基本原理,提出了一种基于智能体的用户偏好决策模型,并将其成功应用在互联网搜索中,不仅提高了搜索的质量,同时具有一定的创新性。     

软件Agent的解释功能

在一些应用领域中,Agent对其使用者的透明性是相当重要的,因此有必要在Agent结构中加入解释机构增强它的透明性。虽然Agent中加入解释机构将增加Agent的复杂性,但是由于具有解释功能使得Agent的可信赖性增强,并且更能体现Agent的友善性,因此也就更容易被Agent的使用者接受,更有利于Agent的应用与推广。     

智能Agent在比较购物中的应用

用智能Agent实现比较购物要解决海量信息搜索、根据用户偏好进行个性化筛选、用户偏好学习等问题。通过事先把主要电子商务网站的内容下载保存到本地数据库来实现快速搜索，通过关键向量信息过滤法可以实现筛选，通过样本学习和反馈学习可以获取用户偏好。以此实现用智能Agent来替用户比较、搜索合适的商品，从而大大提高在线购物的范围和效率，实现个性化的导购。     

基于Agent的DSS模型库运行机制研究

研究模型决策机制问题,模型库是决策支持系统的核心部分,为了提高决策支持系统模型库的通用性和可扩展性,提出了一种基于Agent的DSS模型库通用架构方法和标准化模型策略.主要研究基于Agent的模型表示策略、模型库系统架构及各功能Agent模型,详细阐述模型库中各功能构件的工作原理、模型Agent的自适应演化行为,并在JADE平台上完成了系统原型.实验证明,DSS方法可以带来模型与模型库的标准化,模型的复合也被转换成一个多Agent之间的协作问题,为当前模型库系统的建设提供了一个新的设计策略.     

An ontological Proxy Agent with prediction,CBR, and RBR techniques for fast query processing

This paper proposed a three-tier Proxy Agent working as a mediator between the users and the backend process of a FAQ system. The first tier makes use of an improved sequential patterns mining technique to propose effective query prediction and cache services. The second tier employs an ontology-supported case-based reasoning technique to propose adapted query solutions and tune itself by retaining and updating highly-satisfied cases identified by the user. Finally, the third tier utilizes an ontology-supported rule-based reasoning to generate possible solutions for the user. Our experiments show around 79.1% of the user queries can be answered by Proxy Agent, leaving about 20.9% of the queries for the backend Answerer Agent to take care, which can effectively alleviate the overloading problem usually associated with a backend server.     

基于多Agent远程教学系统的探讨

Agent是一种新兴技术，应用Agent技术建立的网上教学系统，。可以较好地解决现阶段网上教学的不足，从而提高网上教育的教学质量和学习效果。     

Improving Friend Recommendation for Online Learning with Fine-Grained Evolving Interest

Friend recommendation plays a key role in promoting user experience in online social networks (OSNs). However, existing studies usually neglect users’ fine-grained interest as well as the evolving feature of interest, which may cause unsuitable recommendation. In particular, some OSNs, such as the online learning community, even have little work on friend recommendation. To this end, we strive to improve friend recommendation with fine-grained evolving interest in this paper. We take the online learning community as an application scenario, which is a special type of OSNs for people to learn courses online. Learning partners can help improve learners’ learning effect and improve the attractiveness of platforms. We propose a learning partner recommendation framework based on the evolution of fine-grained learning interest (LPRF-E for short). We extract a sequence of learning interest tags that changes over time. Then, we explore the time feature to predict evolving learning interest. Next, we recommend learning partners by fine-grained interest similarity. We also refine the learning partner recommendation framework with users’ social influence (denoted as LPRF-F for differentiation). Extensive experiments on two real datasets crawled from Chinese University MOOC and Douban Book validate that the proposed LPRF-E and LPRF-F models achieve a high accuracy (i.e., approximate 50% improvements on the precision and the recall) and can recommend learning partners with high quality (e.g., more experienced and helpful).