基于重排序融合的社会图书检索系统

用图书的出版信息和用户生成的社会信息从社会媒体中搜索出相关的图书已成为信息检索系统的一个研究热点。然而大部分的信息检索系统都是由单一的检索方法构成,随着用户需求的不断增加,这些系统难以满足用户需求。针对上述问题,提出了一种基于重排序融合的图书检索系统。首先,使用伪相关反馈技术对用户查询内容进行扩展,并将检索结果作为初排序结果;其次,使用用户生成的社会信息特征对初排序结果进行重排序;最后,采用排序学习模型对多种重排序策略得到的结果进行融合。在INEX 2012-2014 Social Book Search公开数据集上针对其它先进检索系统进行了对比实验,实验结果表明,系统的性能（NDCG@10）优于其它方法构成的图书检索系统。     

Visual exploration and retrieval of XML document collections with the generic system X2

This article reports on the XML retrieval system x² that has been developed at the University of Munich over the last 5 years. In a typical session with x², the user first browses a structural summary of the XML database in order to select interesting elements and keywords occurring in documents. Using this intermediate result, queries combining structure and textual references are composed semiautomatically. After query evaluation, the full set of answers is presented in a visual and structured way. x² largely exploits the structure found in documents, queries and answers to enable new interactive visualization and exploration techniques that support mixed IR and database-oriented querying, thus bridging the gap between these three views on the data to be retrieved. Another salient characteristic of x² that distinguishes it from other visual query systems for XML is that it supports various degrees of detailedness in the presentation of answers, as well as techniques for dynamically reordering, grouping and ranking retrieved elements once the complete answer set has been computed.     

大规模社交网络中高效的关键用户选取方法

针对大规模社交网络及其用户发布消息的历史数据,如何快速有效地选取具有较强信息传播能力的关键用户,提出了一种关键用户选取方法。首先,利用社交网络的结构信息,构建以用户为节点的有向图,利用用户发布消息的历史数据,基于Spark计算框架,定量计算由用户活跃度、转发交互度和信息量占比刻画的权重,从而构建社交网络的有向带权图模型;然后,借鉴PageRank算法,建立用户信息传播能力的度量机制,给出基于Spark的大规模社交网络中用户信息传播能力的计算方法;进而,给出基于Spark的d-距选取算法,通过多次迭代,使得所选取的不同关键用户的信息传播范围尽量少地重叠。建立在新浪微博数据上的实验结果表明,所提方法具有高效性、可行性和可扩展性,对于控制不良突发信息传播、社交网络舆情监控具有一定的支撑作用。     

An evaluation of semantic fisheye views for opportunistic search in an annotated image collection

Visual interfaces are potentially powerful tools for users to explore a representation of a collection and opportunistically discover information that will guide them toward relevant documents. Semantic fisheye views (SFEVs) are focus + context visualization techniques that manage visual complexity by selectively emphasizing and increasing the detail of information related to the users focus and deemphasizing or filtering less important information.In this paper we describe a prototype for visualizing an annotated image collection and an experiment to compare the effectiveness of two distinctly different SFEVs for a complex opportunistic search task. The first SFEV calculates relevance based on keyword-content similarity and the second based on conceptual relationships between images derived using WordNet. The results of the experiment suggest that semantic-guided search is significantly more effective than similarity-guided search for discovering and using domain knowledge in a collection.     

Task and user effects on reading patterns in information search

We report on an investigation into people’s behaviors on information search tasks, specifically the relation between eye movement patterns and task characteristics. We conducted two independent user studies (n = 32 and n = 40), one with journalism tasks and the other with genomics tasks. The tasks were constructed to represent information needs of these two different users groups and to vary in several dimensions according to a task classification scheme. For each participant we classified eye gaze data to construct models of their reading patterns. The reading models were analyzed with respect to the effect of task types and Web page types on reading eye movement patterns. We report on relationships between tasks and individual reading behaviors at the task and page level. Specifically we show that transitions between scanning and reading behavior in eye movement patterns and the amount of text processed may be an implicit indicator of the current task type facets. This may be useful in building user and task models that can be useful in personalization of information systems and so address design demands driven by increasingly complex user actions with information systems. One of the contributions of this research is a new methodology to model information search behavior and investigate information acquisition and cognitive processing in interactive information tasks.     

Effectiveness of keyword-based display and selection of retrieval results for interactive searches

We present an approach to increasing the effectiveness of ranked-output retrieval systems that relies on graphical display and user manipulation of “views” of retrieval results, where a view is the subset of retrieved documents that contain a specified subset of query terms. This approach has been implemented in a system named VIEWER (VIEwing WEb Results), acting as an interface to available search engines. An experimental evaluation of the performance of VIEWER in contrast to AltaVista is the major focus of the paper. We first report the results of an experiment on single, short query searches where VIEWER, used as an interactive ranking system, markedly outperformed AltaVista. We then concentrate on a more realistic searching scenario, involving free query formulation, unconstrained selection of retrieval results, and possibility of query reformulation. We report the results of an experiment where the use of VIEWER, compared to AltaVista, seemed to shift the user effort from inspection to evaluation of results, increasing retrieval effectiveness, and user satisfaction. In particular, we found that the VIEWER users retrieved half as many nonrelevant documents as the AltaVista users while retrieving a comparable number of relevant documents. Published online: 22 September 2000     

个性化搜索引擎中用户模型智能调整算法的研究

个性化信息服务越来越成为信息检索领域中研究的热点。针对用户模型构造问题,文章结合手工定制建模与自动分析建模技术,利用空间向量模型表示法,提出了一种用户模型智能调整算法。模拟实验表明,该结构和算法能够有效地提高检索结果的准确度,并且具有良好的可适应性。     

Influence of social presence on eye movements in visual search tasks

Abstract

This study employed an eye-tracking technique to investigate the influence of social presence on eye movements in visual search tasks. A total of 20 male subjects performed visual search tasks in a 2 (target presence: present vs. absent) × 2 (task complexity: complex vs. simple) × 2 (social presence: alone vs. a human audience) within-subject experiment. Results indicated that the presence of an audience could evoke a social facilitation effect on response time in visual search tasks. Compared with working alone, the participants made fewer and shorter fixations, larger saccades and shorter scan path in simple search tasks and more and longer fixations, smaller saccades and longer scan path in complex search tasks when working with an audience. The saccade velocity and pupil diameter in the audience-present condition were larger than those in the working-alone condition. No significant change in target fixation number was observed between two social presence conditions.

Practitioner Summary: This study employed an eye-tracking technique to examine the influence of social presence on eye movements in visual search tasks. Results clarified the variation mechanism and characteristics of oculomotor scanning induced by social presence in visual search.     

具有扩充检索功能的石油化工专业信息搜索引擎的设计与开发

网络信息的日益增加迫切需要适宜的检索工具,特别是进行专业信息的检索,需要体现专业词汇特点的搜索引擎。本文在对搜索引擎核心技术进行研究的基础上,提出了石油化工信息搜索引擎的设计方案,开发了网络机器人模块,实现了海量网页的自动获取;采用最短路径分词和正向最大匹配相结合的算法,实现了中文自动分词;开发了信息索引模块,实现了网页的批量索引和增量索引;开发了信息检索模块,提供布尔逻辑查询,实现摘要自动生成。通过系统集成,初步建立了体现石油化工专业特点的搜索引擎。     

Dimensionality reduction for similarity search with the Euclidean distance in high-dimensional applications

In multimedia information retrieval, multimedia data are represented as vectors in high-dimensional space. To search these vectors efficiently, a variety of indexing methods have been proposed. However, the performance of these indexing methods degrades dramatically with increasing dimensionality, which is known as the dimensionality curse. To resolve the dimensionality curse, dimensionality reduction methods have been proposed. They map feature vectors in high-dimensional space into vectors in low-dimensional space before the data are indexed. This paper proposes a novel method for dimensionality reduction based on a function that approximates the Euclidean distance based on the norm and angle components of a vector. First, we identify the causes of, and discuss basic solutions to, errors in angle approximation during the approximation of the Euclidean distance. Then, this paper propose a new method for dimensionality reduction that extracts a set of subvectors from a feature vector and maintains only the norm and the approximated angle for every subvector. The selection of a good reference vector is crucial for accurate approximation of the angle component. We present criteria for being a good reference vector, and propose a method that chooses a good reference vector. Also, we define a novel distance function using the norm and angle components, and formally prove that the distance function consistently lower-bounds the Euclidean distance. This implies information retrieval with this function does not incur any false dismissals. Finally, the superiority of the proposed approach is verified via extensive experiments with synthetic and real-life data sets.

大规模搜索引擎检索系统框架与实现要点

随着Web规模的不断扩大，搜索引擎正成为因特网上最常用的应用之一。本文以天网搜索为实例，分析了大规模通用型中文搜索引擎检索系统的设计与实现技术。围绕检索效率和检索效果两个方面，本文介绍天网检索系统的集成框架结构和分布式架构，并分析了索引创建和索引检索中的相关实现技术。     

On transitions in the behaviour of tabu search algorithm TabuCol for graph colouring

Even though tabu search is one of the most popular metaheuristic search strategies, its understanding in terms of behavioural transitions and parameter tuning is still very limited. In this paper, we present a theoretical and experimental study of a popular tabu search algorithm TabuCol for graph colouring. We show that for some instances, there are sharp transitions in the behaviour of TabuCol, depending on the value of tabu tenure parameter. The location of this transition depends on graph structure and may also depend on its size. This is further supported by an experimental study of success rate profiles, which we define as an empirical measure of these transitions. We study the success rate profiles for a range of graph colouring instances, from 2-colouring of trees and forests to several instances from the DIMACS benchmark. These reveal that TabuCol may exhibit a spectrum of different behaviours ranging from simple transitions to highly complex probabilistic behaviour.     

基于自适应差异化图卷积的社交网络新增恶意用户检测

社交网络新增恶意用户检测作为一项分类任务,一直面临着数据样本不足、恶意用户标注稀少的问题。在数据有限的情况下,为了能够精确地检测出恶意用户,提出一种基于自适应差异化图卷积网络的检测方法。该方法通过提取社交网络中的用户特征和社交关系构建社交网络图。构建社交网络图后,计算节点与邻居的相似度,并对邻居进行优先级排序,利用优先级顺序采样关键邻居。关键邻居的特征通过自适应权重的加权平均方式聚合到节点自身,以此更新节点特征。特征更新后的节点通过特征降维和归一化计算得到恶意值,利用恶意值判断用户的恶意性。实验表明该方法和其他方法相比,具有更高的恶意用户查全率和整体查准率,并且能够快速地完成对新增用户的检测,证明了自适应差异化图卷积网络能够有效捕捉到少量样本的关键特征。     

社交信息背景下路径决策行为实验

为解决社交信息对路径选择影响缺少定量分析的问题,在行为实验理论的基础上,设计了在无交通信息、部分交通信息和完全交通信息3种情形下的受验者路径选择行为实验,并通过z-T ree和z-L eaf软件得以实现.实验表明:无交通信息情形下,受验者更倾向于选择可能最短路径;在部分交通信息情形下,受验者更倾向于选择当前节点最优路段...     

Assessing the effect of persuasive robots interactive social cues on users’ psychological reactance,liking, trusting beliefs and compliance

ABSTRACT

Research in the field of social robotics suggests that enhancing social cues in robots can elicit more social responses in users. It is however not clear how users respond socially to persuasive social robots and whether such reactions will be more pronounced when the robots feature more interactive social cues. In the current research, we examine social responses towards persuasive attempts provided by a robot featuring different numbers of interactive social cues. A laboratory experiment assessed participants’ psychological reactance, liking, trusting beliefs and compliance toward a persuasive robot that either presented users with: no interactive social cues (random head movements and random social praises), low number of interactive social cues (head mimicry), or high number of interactive social cues (head mimicry and proper timing for social praise). Results show that a persuasive robot with the highest number of interactive social cues invoked lower reactance and was liked more than the robots in the other two conditions. Furthermore, results suggest that trusting beliefs towards persuasive robots can be enhanced by utilizing praise as presented by social robots in no interactive social cues and high number of interactive social cues conditions. However, interactive social cues did not contribute to higher compliance.     

基于Nutch与元搜索引擎技术的高校网络舆情监测系统研究

针对高校网络舆情监测工作的不足,结合高校舆情信息的传播特点,提出了在校园网范围内基于Nutch搜索引擎技术进行信息检索,在互联网上采用元搜索引擎技术获取相关信息的舆情监测系统构建方案。采用关键词特征库匹配方式自动进行网络舆情监测,及时准确地发现网络舆情信息;探讨了高校对于网络舆情的监测、分析、引导和反馈等环节的处理方法,从而形成完善的应对机制。     

Exploring interactive style and user experience design for social web of things of Chinese users: A case study in Beijing

This study designed an interactive IoT (Internet of Things) service on mobile devices based upon the concept of Social Web of Things (SWoT), with which users can interact with IoT in the same way they use the social network services. In order to integrate IoT into Chinese daily use, this study also investigated how Chinese users interact with things and IoT technologies. A four-phase study among users living in Beijing was conducted with a lifecycle of user-centered design. Results revealed that SWoT could activate users׳ intuitive understanding of social network services, and make the interaction with SWoT natural in their own ways. Users living in Beijing tended to be utility-oriented and highly emphasized efficiency; they were distinctive in uncertainty avoidance and preference of hierarchical way of managing things. Our results suggested implications for the design of user-centered IoT systems in China, and shed light on improvements for the performance of IoT systems and balance of requirements between users and system design.     

一种融合社会化标注系统中主题域相似的个性化排序方法

随着网络技术的发展,互联网中越来越多的资源被应用于信息检索中,大量的研究表明,社会化标注可以用于改善信息检索。现有个性化排序的方法中,用户之间的相似度大多通过其共同使用过的标签集来计算。然而,现实中用户标注数据存在稀疏性和标签同义词等问题,导致相似度计算并不准确。在前人研究的基础上,提出了一种融合主题域相似的个性化排序方法。该方法首先通过主题域的划分,将不同主题含义的网页及标签分开,通过构建的标签相似网络找出标签同义词。然后结合用户标签和主题偏好找出兴趣相近的用户,并对用户的标注信息进行扩展,从而能够有效地改善个性化信息检索的效果。在真实数据上的实验结果表明,该方法能有效缓解标注稀疏性和标签同义词问题,有助于改善用户检索体验。     

What can Studies of e-Learning Teach us about Collaboration in e-Research? Some Findings from Digital Library Studies

e-Research is intended to facilitate collaboration through distributed access to content, tools, and services. Lessons about collaboration are extracted from the findings of two large, long-term digital library research projects. Both the Alexandria Digital Earth Prototype Project (ADEPT) and the Center for Embedded Networked Sensing (CENS) project on data management leverage scientific research data for use in teaching. Two forms of collaboration were studied: (1) direct, in which faculty work together on research projects; and (2) indirect or serial, in which faculty use or contribute content to a common pool, such as teaching resources, concepts and relationships, or research data. Five aspects of collaboration in e-Research are discussed: (1) disciplinary factors, (2) incentives to adopt e-Learning and e-Research technologies, (3) user roles, (4) information sharing, and (5) technical requirements. Collaboration varied by research domain in both projects, and appears partly to be a function of the degree of instrumentation in data collection. Faculty members were more interested in tools to manage their own research data than in tools to facilitate teaching. They also were more reflective about their research than teaching activities. The availability of more content, tools, and services to incorporate primary data in teaching was only a minimal incentive to use these resources. Large investments in a knowledge base of scientific concepts and relationships for teaching did not result in re-use by other faculty during the course of the project. Metadata requirements for research and for teaching vary greatly, which further complicates the transfer of resources across applications. Personal digital libraries offer a middle ground between private control and public release of content, which is a promising direction for the design of digital libraries that will facilitate collaboration in e-Research.