A Novel Personalized Web Search Model

A novel personalized Web search model is proposed. The new system, as a middleware between a user and a Web search engine, is set up on the client machine. It can learn a user＇s preference implicitly and then generate the user profile automatically. When the user inputs query keywords, the system can automatically generate a few personalized expansion words by computing the term-term associations according to the current user profile, and then these words together with the query keywords are submitted to a popular search engine such as Yahoo or Google. These expansion words help to express accurately the user＇s search intention. The new Web search model can make a common search engine personalized, that is, the search engine can return different search results to different users who input the same keywords. The experimental results show the feasibility and applicability of the presented work.     

Mining Interesting Knowledge from Web-Log

Web-tog contains a lot of information related with user activities on the Internet. How to mine user browsing interest patterns effectively is an important and chattengeable research topic. On the analysis of the present algorithm‘s advantages and disadvantages, we propose a new concept: support-interest. Its key insight is that visitor will backtrack if they do not find the information where they expect. And the point from where they backtrack is tbe expected location for the page. We present User Access Matrix and the corresponding algorithm for discovering such expected locations that can handle page caching by the browser. Since the URL-URL matrix is a sparse matrix which can be represented by List of 3 tuples, we can mine user preferred sub-paths from the computation of this matrix. Accordinglyo all the sub-paths are merged, and user preferred paths are formed. Experiments showed that it was accurate and scalable. It‘s suitable for website based application, such as to optimize website‘s topologicat structure or to design personalized scrvices.     

A framework for personalized adaptive user interest prediction based on topic model and forgetting mechanism

User interest is not static and changes dynamically. In the scenario of a search engine, this paper presents a personalized adaptive user interest prediction framework. It represents user interest as a topic distribution, captures every change of user interest in the history, and uses the changes to predict future individual user interest dynamically. More specifically, it first uses a personalized user interest representation model to infer user interest from queries in the user’s history data using a topic model; then it presents a personalized user interest prediction model to capture the dynamic changes of user interest and to predict future user interest by leveraging the query submission time in the history data. Compared with the Interest Degree Multi-Stage Quantization Model, experiment results on an AOL Search Query Log query log show that our framework is more stable and effective in user interest prediction.     

基于概念检索的中文搜索引擎研究

在Internet高速发展的信息时代,网络搜索引擎是人们快速获得信息的重要工具之一。然而由于传统的搜索引擎基本上都是采用基于关键词匹配的全文检索技术,导致检索结果不全、无关信息过多。本文给出了一种基于概念检索的中文搜索引擎模型,采用“以网对网”技术来实现概念检索,使搜索引擎从基于关键词的检索提高到基于知识的智能检索。基于概念检索的中文搜索引擎对知识有一定的理解和处理能力,在一定程度上提高了搜索引擎的智能化。     

基于软约束满足理论的用户偏好建模方法

用户建模是电子商务领域的重要研究问题之一,但是到目前为止还没有非常有效的方法能够建模用户的动态偏好等特征。在分析电子商务领域用户偏好特征获取一般特点的基础上,提出基于实例的两阶段递增式用户偏好特征获取方法。并采用基于半环的软约束满足问题表示用户的各类偏好特征,将用户建模看成是潜在方案空间上施加一系列偏好约束的过程。提出不同的偏好约束类型以及利用偏好约束建立用户模型的过程,实现电子商务领域复杂用户偏好的灵活建模。     

Joint Power Control Algorithm for CDMA TDD System

Current uplink power control algorithm of code division multiple access （CDMA） time division duplex（TDD） system contains signal-to-interference-ratio（SIR） and interference measurement, which is based on history information. However, the adjustment of transmission power of one user equipment（UE） will affect SIR of the others, because the power control is done independently of the power adjustment of other UE＇s. So the convergence speed of power control is not fast and the system performance degrades. In the proposed algorithm, all UE＇s assigned in same time slot adjust their transmission power based on current information, which takes into account the adjustments of transmission power of all UE＇s. The simulation results indicated that the proposed algorithm outperforms the old algorithm.     

采用向量空间模型的个性化信息检索方法

为了提升检索结果与用户个性化需求的符合程度,依托向量空间模型提出一种新的检索方法.将用户查询关键词和语料库内的文本信息都映射为向量,从而把检索过程转化为向量相似性的比对.在比对过程中,通过关键词权重突出用户个性化需求,通过余弦相似度判断符合程度.实验结果表明:文中方法的检索结果与用户需求的符合程度明显提高.     

Authenticated Diffie-Hellman key agreement protocol with forward secrecy

Forward secrecy is an important security property in key agreement protocol. Based on Ham＇s protocol, in this paper a new authenticated Diffie-Hellman key agreement protocol with half forward secrecy is proposed. This protocol is also based on a single cryptographic assumption, and is user authentication and shared key authentication. More importantly, our protocol provides forward secrecy with respect to one of the parties. For this reason, besides the advantages of Ham＇s protocol, in practice, our protocol can reduce the damages resulted from the disclosure of the user＇s secret key and it is very beneficial to today＇s communication with portable devices.     

A new efficient blind signcryption

In a blind signcryption, besides the functions of digital signature and encryption algorithm for authentication and confidentiality, a user can delegates another user＇s capability with the anonymity of the participants guaranteed. Some blind signcryptions were proposed but without a blind signcryption with public public verifiability. In this paper, verifiability that is proved to be efficient and secure is proposed. Through the security analysis, we proved that the scheme can offer confidentiality, integrity, unforgeability, non-repudiation and public verifiability. The coming research direction is also summarized.     

高性能FTP搜索引擎的设计

FTP(file transfer protcol)搜索引擎的重点在于中文分词技术和检索技术.使用了一种面向FTP搜索优化的最大前向匹配分词算法,并将用户查询作为反馈来更新分词算法中所使用的字典,结合倒排索引技术实现了一个高性能的FTP搜索引擎的原型系统.压力测试结果表明此FTP搜索引擎具有很高的性能.     

Research on Bayesian Network Based User＇s Interest Model

It has very realistic significance for improving the quality of users＇ accessing information to filter and selectively retrieve the large number of information on the Internet. On the basis of analyzing the existing users＇ interest models and some basic questions of users＇ interest （representation, derivation and identification of users＇ interest）, a Bayesian network based users＇ interest model is given. In this model, the users＇ interest reduction algorithm based on Markov Blanket model is used to reduce the interest noise, and then users＇ interested and not interested documents are used to train the Bayesian network. Compared to the simple model, this model has the following advantages like small space requirements, simple reasoning method and high recognition rate. The experiment result shows this model can more appropriately reflect the user＇s interest, and has higher performance and good usability.     

网络级科研发现系统个性化服务研究

阐述了网络级科研发现系统及个性化服务的概念,针对用户的检索行为,分别从检索界面、检索过程、检索结果 3个方面具体论述了网络级科研发现系统的个性化服务,探讨了该系统个性化服务的优势和深入挖掘个性化服务的重要性。     

Automatic user goals identification based on anchor text and click-through data

Understanding the underlying goal behind a user＇s Web query has been proved to be helpful to improve the quality of search. This paper focuses on the problem of automatic identification of query types according to the goals. Four novel entropy-based features extracted from anchor data and click-through data are proposed, and a support vector machines （SVM） classifier is used to identify the user＇s goal based on these features. Experi- mental results show that the proposed entropy-based features are more effective than those reported in previous work. By combin- ing multiple features the goals for more than 97% of the queries studied can be correctly identified. Besides these, this paper reaches the following important conclusions： First, anchor-based features are more effective than click-through-based features; Second, the number of sites is more reliable than the number of links; Third, click-distribution- based features are more effective than session-based ones.     

A Deep Web Query Interfaces Classification Method Based on RBF Neural Network

This paper proposes a new approach for classification for query interfaces of Deep Web, which extracts features from the form＇s text data on the query interfaces, assisted with the synonym library, and uses radial basic function neural network （RBFNN） algorithm to classify the query interfaces. The applied RBFNN is a kind of effective feed-forward artificial neural network, which has a simple networking structure but features with strength of excellent nonlinear approximation, fast convergence and global convergence. A TEL_8 query interfaces＇ data set from UIUC on-line database is used in our experiments, which consists of 477 query interfaces in 8 typical domains. Experimental results proved that the proposed approach can efficiently classify the query interfaces with an accuracy of 95.67%.     

智能元搜索引擎的研究

介绍了一种基于Internet的信息检索系统：智能元搜索引擎。该系统采用分布式Agent技术,以词典作为信息匹配和交互的媒体,能够很好地对用户的兴趣进行自适应学习,对用户提供个性化的信息检索服务。     

搜索引擎作弊及反作弊技术探究

一些网页为了更容易被用户检索访问,采用作弊手段欺骗搜索引擎来提高排名。搜索引擎作弊具备多样性和复杂性,而且往往是多种作弊方式相互结合共同作弊,给搜索引擎的反作弊检测带来了不小困难。文章从分析搜索引擎作弊流行原因出发,对当前比较流行的搜索引擎作弊方式进行了初步研究及分类,对作弊的工作过程及方式进行了综述,针对目前每种作弊技术对应的反作弊技术进行了比较分析,并对未来反作弊技术的发展方向进行了展望。     

An improved sanitizable signature scheme

A new sanitizable signature scheme is proposed, in which the security flaw of Miyazaki＇s sanitizable signature scheme SUMI-4 is improved. The new scheme overcomes the shortcomings of the original scheme SUM1-4 by using sanitizable authorization certificates. The new scheme enables the primitive signer to limit the sanitizer＇s power and still satisfies the security request of sanitizable signature.     

基于知识图谱的文本观点检索方法

文本观点检索旨在检索出与查询主题相关并且表达用户对主题观点的文档。由于用户查询时输入通常很短,难以准确表示查询的信息需求。知识图谱是结构化的语义知识库,通过知识图谱中的知识有助于理解用户的信息需求。因此,提出了一种基于知识图谱的文本观点检索方法。首先由知识图谱获取候选查询扩展词,并计算每个候选词扩展词分布、共现频率、邻近关系、文档集频率,然后利用4类特征通过SVM分类得到扩展词,最后利用扩展词对产生式观点检索模型进行扩展,实现对查询的观点检索。实验表明,在微博和推特两个数据集上,与基准工作对比,所提出的方法在MAP、NDCG等评价指标上均有显著的提升。     

基于深度神经网络和概率矩阵分解的混合推荐算法

针对个性化推荐中用户和项目描述信息未充分利用,用户评分矩阵数据集极端稀疏的情况,提出了基于深度神经网络和概率矩阵分解(PMF)的混合推荐算法.首先,对用户和项目描述信息进行预处理,形成包含用户偏好特征的用户和项目特征集,再将各特征输入深度神经网络模型中进行训练.同时,利用概率矩阵分解模型,根据用户评分矩阵通过最大后验估计优化得到潜在特征向量;然后,通过对概率矩阵分解模型的用户和项目潜在特征向量以及深度神经网络模型的真实特征向量进行迭代更新,收敛得到融合用户和项目真实信息的潜在特征向量;最后,利用该特征向量对用户进行个性化推荐.实验证明,本文算法较经典推荐算法以及前人算法在均方误差与平均绝对误差指标上均有改善,说明本文算法的有效性.