WSAM:互联网UGC文本主观观点挖掘系统

互联网上的用户生成内容UGC(User Generated Content)中蕴含的用户主观观点信息对分析用户行为、用户需求等工作有着重要的价值。设计一套基于自然语言理解的互联网UGC文本主观观点分析系统WSAM,该系统能挖掘出用户主观观点所蕴含的关注对象和主观成分。分析了互联网UGC现象和生成原因,总结出UGC中用户主观观点中的四种主要类型。挖掘用户主观观点过程中,将用户主观观点的挖掘转化为句子中主观观点关注对象的识别和主观成分的判断。算法结合基于词语类、结构类等相关特征,采用最大熵分类器挖掘用户主观观点。实验验证,WSAM系统所采用的算法性能较好,且还能够灵活扩充出情感分析(Opin-ion Mining)等相关应用,同样也能达到较好的结果。     

基于信息内容和拓扑关系的社会媒体用户兴趣分类

随着社会的发展,信息已经成为社会发展越来越重要的部分,人类的信息传播活动越来越明显地展示出分众特征,对用户的分类成为人类信息活动的一个重要研究课题.从这一目标出发,分别基于信息内容、拓扑关系和两者综合的方法,按兴趣主题对社会媒体用户进行分类.对于基于信息内容的用户分类,采用LDA主题模型从用户所发布的内容中提取其主题分布,基于这一分布,采用支持向量机、决策树、贝叶斯等多种模型按兴趣主题对用户进行分类.对于基于拓扑关系的分类,依据相同兴趣主题的用户倾向于拥有共同的粉丝这一发现,构建分类模型来按兴趣主题对用户进行分类.然后提出综合信息内容和拓扑关系的分类方法来对用户进行分类.最后基于大规模Twitter数据的实验发现,采用综合方法对用户进行的兴趣分类性能明显高于采用单一信息内容或粉丝拓扑方法的性能.     

应用随机游走的社交网络用户分类方法

针对现有在线社交网络用户分类方法不能有效利用用户属性和关系网络信息提高分类性能的问题,设计了一种基于随机游走模型的多标签分类方法MLCMRW。该方法的分类过程包括学习用户初始化类别标签以及通过迭代推理获得用户稳定标签分布两个阶段,并且其可以同时考虑用户属性以及关系网络特征信息进行分类。多个在线社交网络数据集上进行的实验表明,MLCMRW比其它已有的代表性方法有更好的分类性能,并且更适合对现实中的在线社交网络进行用户分类。     

A framework for and empirical study of algorithms for traffic assignment

Traffic congestion is an issue in most cities worldwide. Transportation engineers and urban planners develop various traffic management projects in order to solve this issue. One way to evaluate such projects is traffic assignment (TA). The goal of TA is to predict the behaviour of road users for a given period of time (morning and evening peaks, for example). Once such a model is created, it can be used to analyse the usage of a road network and to predict the impact of implementing a potential project. The most commonly used TA model is known as user equilibrium, which is based on the assumption that all drivers minimise their travel time or generalised cost. In this study, we consider the static deterministic user equilibrium TA model.The constant growth of road networks and the need of highly precise solutions (required for select link analysis, network design, etc.) motivate researchers to propose numerous methods to solve this problem. Our study aims to provide a recommendation on what methods are more suitable depending on available computational resources, time and requirements on the solution. In order to achieve this goal, we implement a flexible software framework that maximises the usage of common code and, hence, ensures comparison of algorithms on common ground. In order to identify similarities and differences of the methods, we analyse groups of algorithms that are based on common principles. In addition, we implement and compare several different methods for solving sub-problems and discuss issues related to accumulated numerical errors that might occur when highly accurate solutions are required.     

Evaluating heuristics for tabletop user segmentation based on simultaneous interaction

Differentiating between users that interact on a tabletop could be beneficial for collaborative tasks to support territoriality-oriented features such as a more efficient space management or a better presentation of the contents. In this paper, we design a novel algorithm for the user differentiation or segmentation based on the simultaneous manipulation of the controls. This is a potential differentiating factor that has remained unexplored so far, and in combination with other factors may become relevant to successfully accomplish such differentiation task. Basically it relies on the idea that users manipulate digital elements with a single hand, and hence, if two controls are being used at the same time, they most likely belong to different users. On the generic algorithm, three different versions have been implemented that include several heuristics to address the problem. The comparison under a simulated experiment shows that the heuristic involving more knowledge on distances on user controls performed better according to different goodness functions. This shows promising to further development and refinement of the approach by expanding it with other potential factors to eventually build a robust user differentiation subsystem.     

基于好友亲密度的用户匹配

用户匹配的目的是检测来自不同社交网络的用户是否是同一个人.现有的研究主要集中在用户属性和网络嵌入上,而这些研究方法往往忽略了用户与好友间的亲密关系.因此,文中提出一种基于好友亲密度的用户匹配算法(FCUM).该算法是一种半监督、端到端的跨社交网络用户匹配算法,其中注意力机制被用于量化用户与好友之间的亲密度.好友亲密度的...     

Building a user model implicitly from a cooperative advisory dialog

This paper reviews existing methods for building user models to support adaptive, interactive systems, identifies sigificant problems with these approaches, and describes a new method for implicitly acquiring user models from an ongoing user-system dialog. Existing explicit user model acquisition methods, such as user edited models or model building dialogs put additional burden on the user and introduce artificial model acquisition dialogs. Hand coding stereotypes, another explicit acquisition method, is a tedious and error-prone process. On the other hand, implicit acquisition techniques such as computing presuppositions or entailments either draw too few inferences to be generally useful, or too many to be trusted.In contrast, this paper describes GUMAC, a General User Model Acquisition Component that uses heuristic rules to make default inferences about users' beliefs from their interaction with an advisory expert system. These rules are based on features of human action and conversation that constrain people's behavior and establish expectations about their knowledge. The application of these rules is illustrated with two examples of extended dialogs between users and an investment advisory system. During the course of these conversations, GUMAC is able to acquire an extensive model of the users' beliefs about the aspects of the domain considered in the dialog. These models, in turn, provide the sort of information needed by an explanation generator to tailor explanations the advisory system gives to its users.     

3G无线网络条件下的用户行为分析

随着无线通信技术和计算机网络技术的快速发展,通过移动手持设备快速有效地接入互联网成为广大用户的迫切需求,了解和分析无线用户的行为模式显得十分必要。对一个小区内通过3G无线上网用户的数据进行记录,处理后以此为基础估计数据之间的关系,得出流量、IP数和用户数之间随着时间变化的潜在关系,进一步得出反映无线互联网用户行为习惯的活跃指数。分析结果为无线互联网的性能的改进和提高,以及无线网络服务商合理决策提供了数据依据,具有重要的商业价值。     

社交网络用户影响力的建模方法

社交网络用户影响力在舆情演化、广告营销及政治选举等领域有着广泛应用,研究者在过去的工作中,通过分析和建模,在影响力方面取得了一定的成果,但还存在着定义不明晰、技术落后和应用缺乏等问题。文中明确提出了社交网络用户影响力的研究模型,将传统技术与先进技术结合,并据此梳理了该领域的相关文献,主要从用户、内容特征和深度学习技术的角度论述了基于社交网络的用户影响力的研究方法,并进一步划分成本质和邻域属性、情感分析和元数据、面向局部网络和基于用户及内容特征,还介绍了节点识别的方法,为该领域的学者提供有效且全面的参考。其次,文中还介绍了用户影响力建模方法在预测应用方面的数据集、评价指标和实验结果等,旨在预测下一个激活节点。最后对其未来的发展趋势作出展望。     

基于短期记忆与遗忘系数的用户个性化建模方法研究

搜索引擎的一个标准是不同的用户用相同的查询条件检索时,返回的结果相同。为解决准确性问题,个性化搜索引擎被提出,它可以根据用户的不同个性化特征提供不同的搜索结果。然而,现有的方法更注重用户的长时记忆和独立的用户日志文件,从而降低了个性化搜索的有效性。获取用户短时记忆模型来提供准确有效的用户偏好的个性化搜索方法被广泛采用。首先,根据基于查询关键词的相关概念生成短期记忆模型;接着,基于用户的时序有效点击数据生成用户个性化模型;最后,在用户会话中引入了遗忘因子来优化用户个性化模型。实验结果表明,所提出的方法可以较好地表达用户信息需求,较为准确地构建用户的个性化模型。     

How should I explain? A comparison of different explanation types for recommender systems

Recommender systems help users locate possible items of interest more quickly by filtering and ranking them in a personalized way. Some of these systems provide the end user not only with such a personalized item list but also with an explanation which describes why a specific item is recommended and why the system supposes that the user will like it. Besides helping the user understand the output and rationale of the system, the provision of such explanations can also improve the general acceptance, perceived quality, or effectiveness of the system.In recent years, the question of how to automatically generate and present system-side explanations has attracted increased interest in research. Today some basic explanation facilities are already incorporated in e-commerce Web sites such as Amazon.com. In this work, we continue this line of recent research and address the question of how explanations can be communicated to the user in a more effective way.In particular, we present the results of a user study in which users of a recommender system were provided with different types of explanation. We experimented with 10 different explanation types and measured their effects in different dimensions. The explanation types used in the study include both known visualizations from the literature as well as two novel interfaces based on tag clouds. Our study reveals that the content-based tag cloud explanations are particularly helpful to increase the user-perceived level of transparency and to increase user satisfaction even though they demand higher cognitive effort from the user. Based on these insights and observations, we derive a set of possible guidelines for designing or selecting suitable explanations for recommender systems.     

Selecting users for participation in IT projects: Trading a representative sample for advocates and champions?

The selection of users for participation in IT projects involves trade-offs between multiple criteria, one of which is selecting a representative cross-section of users. This criterion is basic because trading it for other criteria means basing designs on information biased toward some user groups at the expense of others. Based on interviews in development and customer organizations we find that their criteria for user selection favor persons who can contribute to the progress of the IT project over persons who are representative of the full range of users. A highly valued contribution from participating users is the ability to advocate a vision for the system and champion its organizational implementation. A survey in one customer organization shows that respondents’ personal traits explain up to 31% of the variation in their experience of aspects of the usability of a recently introduced system. Thus, unless participating users are representative as to these personal traits, IT projects may, inadvertently, bring about systems that will fail to satisfy many users.     

Data-driven models for timing feedback responses in a Map Task dialogue system

Traditional dialogue systems use a fixed silence threshold to detect the end of users’ turns. Such a simplistic model can result in system behaviour that is both interruptive and unresponsive, which in turn affects user experience. Various studies have observed that human interlocutors take cues from speaker behaviour, such as prosody, syntax, and gestures, to coordinate smooth exchange of speaking turns. However, little effort has been made towards implementing these models in dialogue systems and verifying how well they model the turn-taking behaviour in human–computer interactions. We present a data-driven approach to building models for online detection of suitable feedback response locations in the user's speech. We first collected human–computer interaction data using a spoken dialogue system that can perform the Map Task with users (albeit using a trick). On this data, we trained various models that use automatically extractable prosodic, contextual and lexico-syntactic features for detecting response locations. Next, we implemented a trained model in the same dialogue system and evaluated it in interactions with users. The subjective and objective measures from the user evaluation confirm that a model trained on speaker behavioural cues offers both smoother turn-transitions and more responsive system behaviour.     

UX Curve: A method for evaluating long-term user experience

The goal of user experience design in industry is to improve customer satisfaction and loyalty through the utility, ease of use, and pleasure provided in the interaction with a product. So far, user experience studies have mostly focused on short-term evaluations and consequently on aspects relating to the initial adoption of new product designs. Nevertheless, the relationship between the user and the product evolves over long periods of time and the relevance of prolonged use for market success has been recently highlighted. In this paper, we argue for the cost-effective elicitation of longitudinal user experience data. We propose a method called the “UX Curve” which aims at assisting users in retrospectively reporting how and why their experience with a product has changed over time. The usefulness of the UX Curve method was assessed in a qualitative study with 20 mobile phone users. In particular, we investigated how users’ specific memories of their experiences with their mobile phones guide their behavior and their willingness to recommend the product to others. The results suggest that the UX Curve method enables users and researchers to determine the quality of long-term user experience and the influences that improve user experience over time or cause it to deteriorate. The method provided rich qualitative data and we found that an improving trend of perceived attractiveness of mobile phones was related to user satisfaction and willingness to recommend their phone to friends. This highlights that sustaining perceived attractiveness can be a differentiating factor in the user acceptance of personal interactive products such as mobile phones. The study suggests that the proposed method can be used as a straightforward tool for understanding the reasons why user experience improves or worsens in long-term product use and how these reasons relate to customer loyalty.     

Towards a unified Media-User Typology (MUT): A meta-analysis and review of the research literature on media-user typologies

Considering the increasingly complex media landscape and diversity of use, it is important to establish a common ground for identifying and describing the variety of ways in which people use new media technologies. Characterising the nature of media-user behaviour and distinctive user types is challenging and the literature offers little guidance in this regard. Hence, the present research aims to classify diverse user behaviours into meaningful categories of user types, according to the frequency of use, variety of use and content preferences. To reach a common framework, a review of the relevant research was conducted. An overview and meta-analysis of the literature (22 studies) regarding user typology was established and analysed with reference to (1) method, (2) theory, (3) media platform, (4) context and year, and (5) user types. Based on this examination, a unified Media-User Typology (MUT) is suggested. This initial MUT goes beyond the current research literature, by unifying all the existing and various user type models. A common MUT model can help the Human–Computer Interaction community to better understand both the typical users and the diversification of media-usage patterns more qualitatively. Developers of media systems can match the users’ preferences more precisely based on an MUT, in addition to identifying the target groups in the developing process. Finally, an MUT will allow a more nuanced approach when investigating the association between media usage and social implications such as the digital divide.     

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。     

基于网络编码的手机用户群文件共享和同步

为了快速、准确的实现一跳范围内手机用户群的文件共享,以达到手机用户之间进行协同工作的目的,在Android操作平台上设计、实施了本系统。在系统中提出了基于网络编码的可靠广播算法来保证文件传输的可靠性和快速性,并通过基于机会编码的全网同步算法来实现任一时刻手机群中各节点文件同步。通过真实实验验证了系统中文件共享的时延并不会随着手机群中节点数目的增加而增加,文件同步时网络中数据量大幅度下降,因此相比并发共享等其它方法,系统提供了一种快速、可靠的手机群文件共享和同步方法。     

基于评分离散度的托攻击检测算法

检测托攻击的本质是对真实用户和虚假用户进行分类,现有的检测算法对于具有选择项的流行攻击、段攻击等攻击方式的检测鲁棒性较差.针对这一问题,通过分析真实用户和虚假用户的评分分布情况,结合ID3决策树提出基于用户评分离散度的托攻击检测Dispersion-C算法.算法通过用户评分极端评分比、去极端评分方差和用户评分标准差3个...