视频广告侦测系统的设计与分析

介绍了一个实用的视频广告自动侦测器系统。给出了视频广告自动侦测器的设计步骤、设计原理和实验结果。探讨了视频的MPEG编码方式和基于DC分量的帧匹配算法,为了提高视频段匹配速度,提出了基于镜头长度序列的镜头相似匹配的视频段匹配算法。最后形成了适用于视频广告侦测的帧匹配与镜头长度序列相结合的视频段匹配算法。实验结果证明了算法的可行性和高效性。     

Revisiting the role of web assurance seals in business-to-consumer electronic commerce

There is conflicting evidence as to the current level of awareness and impact of Web Assurance Seal Services (WASSs). This study examines the effects of an educational intervention designed to increase consumer's knowledge, of security and privacy aspect of business-to-consumer (B2C) e-commerce websites and assurance seal services. The study further explores the relationships among consumers' perceptions about online security, including WASSs awareness, importance of WASS, privacy concerns, security concerns, and information quality, before and after the educational intervention. The study finds that educating consumers about the security and privacy dangers of the web, as well as the role of web assurance seals, does increase their awareness and perceived importance of the seals. However, despite this increased awareness, there is little association between these assurance seals and the two indicators of trustworthiness, concerns about privacy and perceived information quality, of an e-commerce site, even after the intervention. Only security concerns have a statistically significant relationship with WASSs awareness before and after the educational intervention. Implications for theory and practice are discussed.     

商业广告排版问题研究

研究对象是报纸的商业广告排版问题。将部分或全部待排商业广告指派到广告区域中，目标是广告之间不相互叠压地填满整个广告区域。首先根据问题的特点建立数学模型；然后运用两级分支定界算法对模型进行求解。仿真实验结果表明，该算法能快速地找到解，适合于实际应用。     

Factors influencing the perceived credibility of diet-nutrition information web sites

This study investigated the factors that influence the perceived credibility of web sites providing diet and nutrition information. Undergirded by the dual-processing models (i.e., Elaboration Likelihood Model and Heuristic Systematic Model), an online experiment (N = 575) was conducted to examine how perceptions of online diet and nutrition information credibility are influenced by source expertise cues and message accuracy; the effects of prior knowledge and interest in the information also were assessed. Results showed that message accuracy increased perceived credibility of the web site regardless of the level of source expertise. However, source expertise had an important effect on website credibility perceptions among those who exhibited low prior knowledge. Finally, message accuracy had a greater impact on web site credibility among those who were highly involved in the issue compared to those who were less involved. The findings increase our understanding of the factors that impact individuals’ processing of online diet and nutrition information and suggest elements practitioners should consider including to create the most effective online sources for diet and nutrition information.     

The design and evaluation of accessibility on web navigation

Web sites have been deployed to create and sustain business competitiveness in a trend of emerging Web technologies and growing e-commerce. One critical success factor of e-commerce is the ability to allow information to be retrieved from a Web site in an efficient and effective manner. Such ability, being determined by both the Web site structure and the Web page organization, can be measured in terms of accessibility and popularity of Web pages. The relationship between accessibility and popularity of web pages is dynamic in nature and can be analyzed to enhance a Web design. Having observed the lack of means to measure information retrieval of a Web site, this paper purports to introduce a guideline to evaluate Web page accessibility based on several structural-based accessibility models where an innovative accessibility–popularity (A–P) analysis is deployed to measure and, thereby, to modify a Web structure. Both push (i.e. demand driven) strategies and pull (i.e. design driven) strategies are incorporated into such guideline. Further, accessibility models are analyzed and compared in order to identify appropriate applications for each model. The paper is concluded by a summary of future directions of the accessibility models.     

Retrieval of online handwriting by synthesis and matching

Search and retrieval is gaining importance in the ink domain due to the increase in the availability of online handwritten data. However, the problem is challenging due to variations in handwriting between various writers, digitizers and writing conditions. In this paper, we propose a retrieval mechanism for online handwriting, which can handle different writing styles, specifically for Indian languages. The proposed approach provides a keyboard-based search interface that enables to search handwritten data from any platform, in addition to pen-based and example-based queries. One of the major advantages of this framework is that information retrieval techniques such as ranking relevance, detecting stopwords and controlling word forms can be extended to work with search and retrieval in the ink domain. The framework also allows cross-lingual document retrieval across Indian languages.     

Noninvasive data analysis: a web of information granules

The study is concerned with a development of information granules and their use in data analysis. The design of information granules exploits a concept of overshadowing meaning that, we retain a given level of membership to a given concept unless faced with a contrary evidence. A detailed algorithm is provided and illustrated through a number of numerical studies. The idea of noninvasive data analysis is then introduced and discussed from a standpoint of a minimal level of structural dependencies to be used in the model.     

The effects of attention inertia on advertisements on the WWW

When a viewer browses a web site, one presumably performs the task of seeking information from a sequence of scattered web pages to form a meaningful path. The aim of this study is to explore changes in the distribution of attention to banner advertisements as a viewer advances along a meaningful path and their effects on the advertisements. With aid of an instrument called eye-tracker, a laboratory experiment was conducted to observe directly the attention that subjects allocate along meaningful paths. Our results show that at different levels of depth in a meaningful path, the amount of attention allocated to the content of a web page is not the same, regardless of whether attention indexes were based on dwell time or the number of fixations.     

Effects of ad types, positions, animation lengths, and exposure times on the click-through rate of animated online advertisings

In this study, we focus on the click-through rate for the advertising effectiveness to examine the effects of design factors on animated online advertisings. A factorial experiment with repeated measuring was designed to collect a set of serially correlated click-through data. Ad types, positions, animation lengths, and exposure times were considered as the independent factors in this study. The generalized estimating equations (GEE) approach is introduced to the logistic regression models with correlated binary data. A goodness-of-fit statistic, quasi-likelihood information criterion (QIC) for data correlated models will be used for evaluating GEE-constructed models. The results showed a logistic regression model with order effect, two-factor interaction effect of ad types and ad positions, as well as ad positions and animation lengths are statistically significant. In addition, the GEE model with AR(1) correlation structures was well verified by the data.     

Consumer perceptions on web advertisements and motivation factors to purchase in the online shopping

The principal objective of this study was to investigate the mediating and moderating effects of product involvement and trust toward websites in relation to the effects of the attributes of web advertisements on customers’ purchase intentions. We collected data consisting of a total of 264 responses from individuals with previous experience with purchasing products from online shopping mall sites, and utilized the PLS (partial least squares) method to analyze the collected data.     

An information retrieval process to aid in the analysis of code clones

The advent of new static analysis tools has automated the searching for code clones, which are duplicated or similar code fragments in a program. However, clone detection tools can report many clones if the source code that is being searched is large. Programmers may have difficulty comprehending the extensive results from the detection tool, which may inhibit the ability to maintain the identified clones. Latent Semantic Indexing (LSI) is an information retrieval technique that attempts to find relationships in a corpus based on the analysis of the documents in the corpus and the terms in the documents. In this paper, LSI is used to cluster clone classes that have been identified initially by a clone detection tool. The goal of this paper is to detect trends and associations among the clustered clone classes and determine if they provide further comprehension to assist in the maintenance of clones. Experimental evaluation of the approach is reported from a sequence of tools that are chained together to perform an analysis of clones detected in the Microsoft Windows NT kernel source code.

建立信息关联方法的研究和应用

随着互联网的飞速发展,通过网络提供的信息越来越多,并且更新速度也越来越快,如何为用户提供准确有效的信息成为了各大网站需要解决的主要问题。本文提出了一种建立信息关联的方法,给出了它的设计思想及其算法和实现的关键技术。实验表明,这种方法可以有效的建立起信息之间的关联,在用户获取某条信息的同时,为其提供更多有价值的关联信息,提高了用户获取信息的准确度和有效性,并且有针对性的信息推荐也节省了用户检索信息的时间。     

A novel approach of cluster based optimal ranking of clicked URLs using genetic algorithm for effective personalized web search

In this paper a novel approach is proposed for generating the optimal ranked clicked URLs using genetic algorithm (GA) based on clustered web query sessions for effective personalized web search. Experimental study was conducted on the data set of web query sessions captured in the domains academics, entertainment and sports to test the effectiveness of clusterwise optimal ranked clicked URLs for personalized web search (PWS). The results, which are verified statistically shows an improvement in the average precision of the personalized web search based on optimal ranked clicked URLs over both Classic IR and personalized web search without optimal ranked clicked URLs. Thus the effectiveness of personalized web search using optimal ranked clicked URLs is confirmed for better customizing the web search according to the information need of the user.     

MuseumFinland—Finnish museums on the semantic web

This article presents the semantic portal MuseumFinland for publishing heterogeneous museum collections on the Semantic Web. It is shown how museums with their semantically rich and interrelated collection content can create a large, consolidated semantic collection portal together on the web. By sharing a set of ontologies, it is possible to make collections semantically interoperable, and provide the museum visitors with intelligent content-based search and browsing services to the global collection base. The architecture underlying MuseumFinland separates generic search and browsing services from the underlying application dependent schemas and metadata by a layer of logical rules. As a result, the portal creation framework and software developed has been applied successfully to other domains as well. MuseumFinland got the Semantic Web Challence Award (second prize) in 2004.     

An empirical study of sentiment analysis for chinese documents

Up to now, there are very few researches conducted on sentiment classification for Chinese documents. In order to remedy this deficiency, this paper presents an empirical study of sentiment categorization on Chinese documents. Four feature selection methods (MI, IG, CHI and DF) and five learning methods (centroid classifier, K-nearest neighbor, winnow classifier, Naïve Bayes and SVM) are investigated on a Chinese sentiment corpus with a size of 1021 documents. The experimental results indicate that IG performs the best for sentimental terms selection and SVM exhibits the best performance for sentiment classification. Furthermore, we found that sentiment classifiers are severely dependent on domains or topics.     

关于信息管理与信息系统专业《信息存储与检索》课程的研究

本文从信息管理与信息系统的专业目标开始分析,确立《信息存储与检索》课程在该专业中的地位,然后阐述《信息存储与检索》课程的跨多学科的特点,分析当前大学的主要教材,选择最适合信息管理与信息系统专业的教材,针对所选教材阐述了该课程的教学内容及教学方式、方法。     

An analysis of multiple factors relating to teachers' problematic information security behavior

Given the high Internet penetration rate and the huge repository of data stored online, there is a growing trend urging people to utilize data. However, the potential for the malicious use of data disclosed online necessitates attention. Risky information security behavior often leads to damage. Previous research has focused on information security behavior in the workplace; however, there has been little research on teachers' perceptions of their own information security behavior, in particular for teachers in primary and secondary education. For students at this age, their teachers can serve as models. Through understanding teachers' information security behavioral intentions and related protection motivation, we can design training programs for teachers and hence increase teachers' as well as students' normative judgment with regard to information security behavior. The purpose of this research is to explore those factors that relate to teachers' information security behavior as grounded in Protection Motivation Theory. Additionally, the construct of social norms was incorporated based on several studies. Overall, we wish to examine how perceived severity, vulnerability, response-efficacy, self-efficacy, response costs and social norms related to teachers' problematic information security behavior. Structural equation modeling was implemented to analyze the relationships. The results and implications are presented.     

An empirical analysis of information retrieval based concept location techniques in software comprehension

Concept location, the problem of associating human oriented concepts with their counterpart solution domain concepts, is a fundamental problem that lies at the heart of software comprehension. Recent research has attempted to alleviate the impact of the concept location problem through the application of methods drawn from the information retrieval (IR) community. Here we present a new approach based on a complimentary IR method which also has a sound basis in cognitive theory. We compare our approach to related work through an experiment and present our conclusions. This research adapts and expands upon existing language modelling frameworks in IR for use in concept location, in software systems. In doing so it is novel in that it leverages implicit information available in system documentation. Surprisingly, empirical evaluation of this approach showed little performance benefit overall and several possible explanations are forwarded for this finding.

Enabling the creation of domain-specific reference collections to support text-based information retrieval experiments in the architecture, engineering and construction industries

The increasing importance of text-based information retrieval (IR) developments in the architecture, engineering, and construction industries (AEC) and the lack of sharable testing resources to support these developments call for an approach that can be used to generate domain-specific reference collections. To address this need, the authors investigated the characteristics of the testing environment in AEC and ways to adapt dominant collection preparation methods for the domain. This paper presents the authors’ collection generation approach through the preparation process of the Taiwanese National Center for Research on Earthquake Engineering (NCREE) collection. The collection’s Chinese-to-English translation instruments are also discussed as matching semantic/linguistic resources are highly valued in AEC’s text-based IR developments. The paper also includes a use case for the NCREE collection to show how a collection generated by the proposed approach could be applied to support research experiment and validation. The direct outputs, the NCREE collection and its translation instruments, are sharable and reusable testing resources, while mechanisms for seeking collections from other researchers are part of the extended research endeavors.     

模糊聚类在Web信息检索中的应用研究

如何从大量信息中快速、有效地进行Web信息检索已经成为一项重要的研究课题，但是传统的搜索引擎所提供的搜索结果仅仅按照与查询的相关性从高到低排成一个有序列表，不具备层次性，用户使用起来并不方便，该文基于Web资源中词语的不分明性即模糊性，提出采用模糊聚类的方法自动组织搜索引擎的结果来解决这个问题。