Online Social Network Adoption: A Cross-Cultural Study

This research explores the extent to which users across cultures adopt the technology of online social networks (OSNs) in order to promote or support a social cause. By surveying graduate-level university students at institutions in the United States, China and India, this research builds on prior work in technology acceptance to model and explain how three elements of the task domain – the cultural aspects of the user, the social nature of the technology, and the social nature of the task – combine to influence the constructs and relationships within a modified Technology Acceptance Model (TAM) framework as well as the behavior that flows from it. This study contributes to our understanding of technology adoption by showing how OSNs are adopted by users across cultures in promoting and supporting social causes.     

Visualization and debugging of distributed multiagent systems

Visualizing the behavior of systems with distributed data, control, and process is a notoriously difficult task. Each component in the distributed system has only a local view of the whole setup, and the onus is on the user to integrate, into a coherent whole, the large amounts of limited information they provide. In this article, we describe an architecture and an implemented system for visualizing and controlling distributed multiagent applications. The system comprises a suite of tools, with each tool providing a different perspective of the application being visualized . Each tool interrogates the components of the distributed application, collates the returned information, and presents this information to users in an appropriate manner. This in essence, shifts the burden ofinference from the user to the visualizer. Our visualizer has been evaluated on four distributed multiagent systems: a travel management application, a telecommunications network management application, a business process management demonstrator, and an electronic commerce application. Lastly, we briefly show how the suite of tools can be used together for debugging multiagent applications - a process we refer to as debugging via corroboration.     

Gaze-based prediction of pen-based virtual interaction tasks

In typical human–computer interaction, users convey their intentions through traditional input devices (e.g. keyboards, mice, joysticks) coupled with standard graphical user interface elements. Recently, pen-based interaction has emerged as a more intuitive alternative to these traditional means. However, existing pen-based systems are limited by the fact that they rely heavily on auxiliary mode switching mechanisms during interaction (e.g. hard or soft modifier keys, buttons, menus). In this paper, we describe how eye gaze movements that naturally occur during pen-based interaction can be used to reduce dependency on explicit mode selection mechanisms in pen-based systems. In particular, we show that a range of virtual manipulation commands, that would otherwise require auxiliary mode switching elements, can be issued with an 88% success rate with the aid of users׳ natural eye gaze behavior during pen-only interaction.     

SocoTraveler: Travel-package recommendations leveraging social influence of different relationship types

The immense amount of data generated and collected on e-commerce platforms provides opportunities and challenges for big data analytics to create business value. E-tourism platforms collect not only users’ travel information but also users’ social connection information and need effective personalized recommendation systems for target marketing. In this paper, we aim to study how different types of social relationships such as colleague, schoolmate, and relative between co-travelers influence a user’s travel behavior and how to use this influence to enhance recommendation quality. To this end, we develop a probabilistic topic model leveraging individual travel history and social influence of co-travelers to capture personal interests and propose a recommendation method to utilize the proposed model. Experiments on a real travel dataset show that the proposed approach significantly outperforms benchmarks. The result highlights useful findings for travel agencies.     

Commitment-based software development

During the development of a system, software modules can be viewed in terms of their commitments: the constraints imposed by their own structure and behavior, and by their relationships with other modules (in terms of resource consumption, data requirements. etc.). The Comet system uses explicit representation and reasoning with commitments to aid the software design and development process-in particular, to lead software developers to make decisions that result in reuse. Developers can examine the commitments that must be met in order to include an existing module, and can explore how commitments change when modules are modified. Comet has been applied to the domain of sensor-based tracker software     

Understanding the locality effect in Twitter: measurement and analysis

Twitter is one of the most popular applications in the current Internet with more than 500 M registered users across the world. In this paper, we conduct a comprehensive analysis to understand the geographical characteristics of Twitter using cross-community mining techniques. Specifically, we study the locality level shown by the three main elements of Twitter, namely users, relationships and information flow. For this purpose, we rely on a dataset including the geolocation information of more than 17, 100 and 3.5 M users, relationships and tweets, respectively. Our main findings are: (1) most of the Twitter users perform their activity from an area of at most few hundred kms covering few cities within a unique country; (2) the location (i.e., country), and in particular factors such as language or Twitter popularity within a country, dictates the level of locality in the relationships of users and Twitter conversations originated in that country. The combination of these factors reveals the presence of four types of country locality profiles that we carefully analyze and compare in the paper.     

简论宪法惯例

宪法惯例是宪法制度的重要组成部分 ,在立宪制国家它是宪政运行中对宪法的重要创造性补充。起源于政治惯例的宪法惯例的构成必须具备三方面的重要特征。在中国法治化拓进的今天 ,深入探讨宪法惯例的作用具有较强的现实意义     

基于图卷积神经网络的在线社区行为预测

在线知识社区中,问题的回答可以看作多个回答者用户(领域专家)之间的协作行为。协作行为在知识社区中通常是大规模地发生,协作行为预测对在线社交中领域专家的推荐有重要意义。基于在线知识社区中回答者用户之间的协作行为,构建以领域专家为节点,以他们之间的协作回答关系为边的协作网络。由于协作行为网络的构建与社交关系网络的构建上结构的相似性,可以将协作行为预测构建为协作网络中的链接预测问题。通过构建基于图卷积神经网络的链接预测模型,对在线知识社区中回答者用户的协作行为进行预测。基于“知乎”数据集的实验验证,与其他经典的预测方法进行比较时,发现提出的方法能够更加有效地预测在线知识社区中回答者用户之间的协作行为。     

Relating ontologies with a fuzzy information model

More people than ever before have access to information with the World Wide Web; information volume and number of users both continue to expand. Traditional search methods based on keywords are not effective, resulting in large lists of documents, many of which unrelated to users’ needs. One way to improve information retrieval is to associate meaning to users’ queries by using ontologies, knowledge bases that encode a set of concepts about one domain and their relationships. Encoding a knowledge base using one single ontology is usual, but a document collection can deal with different domains, each organized into an ontology. This work presents a novel way to represent and organize knowledge, from distinct domains, using multiple ontologies that can be related. The model allows the ontologies, as well as the relationships between concepts from distinct ontologies, to be represented independently. Additionally, fuzzy set theory techniques are employed to deal with knowledge subjectivity and uncertainty. This approach to organize knowledge and an associated query expansion method are integrated into a fuzzy model for information retrieval based on multi-related ontologies. The performance of a search engine using this model is compared with another fuzzy-based approach for information retrieval, and with the Apache Lucene search engine. Experimental results show that this model improves precision and recall measures.     

‘Long autonomy or long delay?’ The importance of domain in opinion mining

Nowadays, people do not only navigate the web, but they also contribute contents to the Internet. Among other things, they write their thoughts and opinions in review sites, forums, social networks, blogs and other websites. These opinions constitute a valuable resource for businesses, governments and consumers. In the last years, some researchers have proposed opinion extraction systems, mostly domain-independent ones, to automatically extract structured representations of opinions contained in those texts. In this work, we tackle this task in a domain-oriented approach, defining a set of domain-specific resources which capture valuable knowledge about how people express opinions on a given domain. These resources are automatically induced from a set of annotated documents. Some experiments were carried out on three different domains (user-generated reviews of headphones, hotels and cars), comparing our approach to other state-of-the-art, domain-independent techniques. The results confirm the importance of the domain in order to build accurate opinion extraction systems. Some experiments on the influence of the dataset size and an example of aggregation and visualization of the extracted opinions are also shown.     

Using semantic components to search for domain-specific documents: An evaluation from the system perspective and the user perspective

We seek to leverage an expert user's knowledge about how information is organized in a domain and how information is presented in typical documents within a particular domain-specific collection, to effectively and efficiently meet the expert's targeted information needs. We have developed the semantic components model to describe important semantic content within documents. The semantic components model for a given collection (based on a general understanding of the type of information needs expected) consists of a set of document classes, where each class has an associated set of semantic components. Each semantic component instance consists of segments of text about a particular aspect of the main topic of the document and may not correspond to structural elements in the document. The semantic components model represents document content in a manner that is complementary to full text and keyword indexing. This paper describes how the semantic components model can be used to improve an information retrieval system. We present experimental evidence from a large interactive searching study that compared the use of semantic components in a system with full text and keyword indexing, where we extended the query language to allow users to search using semantic components, to a base system that did not have semantic components. We evaluate the systems from a system perspective, where semantic components were shown to improve document ranking for precision-oriented searches, and from a user perspective. We also evaluate the systems from a session-based perspective, evaluating not only the results of individual queries but also the results of multiple queries during a single interactive query session.     

Trust in information sources: seeking information from people,documents, and virtual agents

The notion of trust has been virtually absent from most work on how people assess and choose their information sources. Based on two empirical cases this study shows that software engineers and users of e-commerce websites devote a lot of attention to considerations about the trustworthiness of their sources, which include people, documents, and virtual agents. In the project-based software engineering environment trust tends to be a collaborative issue and the studied software engineers normally know their sources first-hand or have them recommended by colleagues. Outside this network people are cautious and alert to even feeble cues about source trustworthiness. For example, users of e-commerce websites—generally perceived as single-user environments—react rather strongly to the visual appearance of virtual agents, though this is clearly a surface attribute. Across the two cases people need access to their sources in ways that enable them to assess source trustworthiness, access alone is not enough.     

在线社交网络下基于信任度的消息传播模型

社交网络作为一种新兴的媒体具有广泛的社会影响力,且基于社交网络的营销方式逐渐成为一种新的发展趋势,因此研究社交网络中消息的传播具有重大的现实和经济意义。通过借鉴日常生活中人与人之间的信任原理,提出了一种基于信任度的消息传播模型。该模型首先利用个体的公开信息,使用数据挖掘的算法对个体进行分类;然后,根据同类和不同类个体之间的关系计算个体之间的信任度;最后,使用消息与个体的属性相似性以及信任度来计算消息可能传播范围。给出了相应的计算方法,并与两种基准方法对比,结果表明,该模型在准确度上提升15%左右,而所用时间降低50%以上。与数据集统计结果对比,该实验的结果与统计结果相差5%左右,充分表明该模型在实际应用中有比较好的效果。     

Performing Binary-Categorization on Multiple-Record Web Documents Using Information Retrieval Models and Application Ontologies

To retrieve Web documents of interest, most of the Web users rely on Web search engines. All existing search engines provide query facility for users to search for the desired documents using search-engine keywords. However, when a search engine retrieves a long list of Web documents, the user might need to browse through each retrieved document in order to determine which document is of interest. We observe that there are two kinds of problems involved in the retrieval of Web documents: (1) an inappropriate selection of keywords specified by the user; and (2) poor precision in the retrieved Web documents. In solving these problems, we propose an automatic binary-categorization method that is applicable for recognizing multiple-record Web documents of interest, which appear often in advertisement Web pages. Our categorization method uses application ontologies and is based on two information retrieval models, the Vector Space Model (VSM) and the Clustering Model (CM). We analyze and cull Web documents to just those applicable to a particular application ontology. The culling analysis (i) uses CM to find a virtual centroid for the records in a Web document, (ii) computes a vector in a multi-dimensional space for this centroid, and (iii) compares the vector with the predefined ontology vector of the same multi-dimensional space using VSM, which we consider the magnitudes of the vectors, as well as the angle between them. Our experimental results show that we have achieved an average of 90% recall and 97% precision in recognizing Web documents belonged to the same category (i.e., domain of interest). Thus our categorization discards very few documents it should have kept and keeps very few it should have discarded.     

A Quantified Past: Toward Design for Remembering With Personal Informatics

This article questions how people will interact with a quantified past—the growing historical record generated by the increasing use of sensor-based technologies and, in particular, personal informatics tools. In a qualitative study, we interviewed 15 long-term users of different self-tracking tools about how they encountered and made meaning from historical data they had collected. Our findings highlight that even if few people are self-tracking as a form of deliberate lifelogging, many of them generate data and records that become meaningful digital possessions. These records are revealing of many aspects of people’s lives. Through considerable rhetorical data-work, people can appropriate such records to form highly personal accounts of their pasts. We use our findings to identify six characteristics of a quantified past and map an emerging design space for the long-term and retrospective use of personal informatics. Principally, we propose that design should seek to support people in making account of their data and guard against the assumption that more, or “better,” data will be able to do this for them. To this end, we speculate on design opportunities and challenges for experiencing, curating, and sharing historical personal data in new ways.     

Exploring how users engage with hybrid process artifacts based on declarative process models: a behavioral analysis based on eye-tracking and think-aloud

Process design artifacts have been increasingly used to guide the modeling of business processes. To support users in designing and understanding process models, different process artifacts have been combined in several ways leading to the emergence of the so-called “hybrid process artifacts”. While many hybrid artifacts have been proposed in the literature, little is known about how they can actually support users in practice. To address this gap, this work investigates the way users engage with hybrid process artifacts during comprehension tasks. In particular, we focus on a hybrid representation of DCR Graphs (DCR-HR) combining a process model, textual annotations and an interactive simulation. Following a qualitative approach, we conduct a multi-granular analysis exploiting process mining, eye-tracking techniques, and verbal data analysis to scrutinize the reading patterns and the strategies adopted by users when being confronted with DCR-HR. The findings of the coarse-grained analysis provide important insights about the behavior of domain experts and IT specialists and show how user’s background and task type change the use of hybrid process artifacts. As for the fine-grained analysis, user’s behavior was classified into goal-directed and exploratory and different strategies of using the interactive simulation were identified. In addition, a progressive switch from an exploratory behavior to a goal-directed behavior was observed. These insights pave the way for an improved development of hybrid process artifacts and delineate several directions for future work.

     

基于用户-标签异构网络的社区问答专家发现方法

在Stack Overflow、Quora等社区问答网站中,日益增长的用户数使新问题数量急剧增加,传统的专家发现方法通常根据历史回答记录建立用户文档,再从中提取用户文本特征,难以及时寻找到合适的专家进行回答。针对该问题,提出一种社区问答中基于用户-标签异构网络的专家发现方法。根据用户历史回答记录和问题的附带标签构建用户-标签网络,以此得到用户的向量表示。在此基础上,使用全连接神经网络提取用户特征和问题文本特征,通过比较两者的余弦相似度得到候选专家列表。基于StackExchange的真实世界数据集进行测试,实验结果表明,与LDA、STM、RankingSVM和QR-DSSM方法相比,该方法的MRR指标值较高,能够准确寻找到可提供正确答案的专家。     

Multiple-vector user profiles in support of knowledge sharing

This paper describes an algorithm to automatically construct expertise profiles for company employees, based on documents authored and read by them. A profile consists of a series of high dimensional vectors, each describing an expertise domain, and provides a hierarchy between these vectors, enabling a structured view on an employee’s expertise. The algorithm is novel in providing this layered view, as well as in its high degree of automation and its generic approach ensuring applicability in an industrial setting.The profiles provide support for several knowledge management functionalities that are difficult or impossible to achieve using existing methods. This paper in particular presents the initialization of communities of practice, bringing together both experts and novices on a specific topic. An algorithm to automatically discover relationships between employees based on their profiles is described. These relationships can be used to initiate communities of practice. The algorithms are validated by means of a realistic dataset.     

A novel recommendation approach based on users’ weighted trust relations and the rating similarities

With the growing popularity of open social networks, approaches incorporating social relationships into recommender systems are gaining momentum, especially matrix factorization-based ones. The experiments in previous literatures indicate that social information is very effective in improving the performance of traditional recommendation algorithms. However, most of existing social recommendation methods only take one kind of social relations—trust information into consideration, which is far from satisfactory. Furthermore, most of the existing trust networks are binary, which results in the equal treatment to different users who are trusted by the same user in these methods. In this paper, based on matrix factorization methods, we propose a new approach to make recommendation with social information. Its novelty can be summarized as follows: (1) it shows how to add different weights on the social trust relationships among users based on the trustee’s competence and trustworthiness; (2) it incorporates the similarity relationships among users as a complement into the social trust relationships to enhance the computation of user’s neighborhood; (3) it can balance the influence of these two kinds of relationships based on user’s individuality adaptively. Experiments on Epinions and Ciao datasets demonstrate that our approach outperforms the state-of-the-art algorithms in terms of mean absolute error and root mean square error, in particular for the users who rated a few items.