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.

Exploring users' within-site navigation behavior:A case study based on clickstream data

Purpose:The goal of our research is to suggest specific Web metrics that are useful for evaluating and improving user navigation experience on informational websites.Design/methodology/approach:We revised metrics in a Web forensic framework proposed in the literature and defined the metrics of footprint,track and movement.Data were obtained from user clickstreams provided by a real estate site’s administrators.There were two phases of data analysis with the first phase on navigation behavior based on user footprints and tracks,and the second phase on navigational transition patterns based on user movements.Findings:Preliminary results suggest that the apartment pages were heavily-trafficked while the agent pages and related information pages were underused to a great extent.Navigation within the same category of pages was prevalent,especially when users navigated among the regional apartment listings.However,navigation of these pages was found to be inefficient.Research limitations:The suggestions for navigation design optimization provided in the paper are specific to this website,and their applicability to other online environments needs to be verified.Preference predications or personal recommendations are not made during the current stage of research.Practical implications:Our clickstream data analysis results offer a base for future research.Meanwhile,website administrators and managers can make better use of the readily available clickstream data to evaluate the effectiveness and efficiency of their site navigation design.Originality/value:Our empirical study is valuable to those seeking analysis metrics for evaluating and improving user navigation experience on informational websites based on clickstream data.Our attempts to analyze the log file in terms of footprint,track and movement will enrich the utilization of such trace data to engender a deeper understanding of users’within-site navigation behavior.

Dynamic data-sharing based user recruitment in mobile crowdsensing

Mobile crowdsensing(MCS) has become an emerging paradigm to solve urban sensing problems by leveraging the ubiquitous sensing capabilities of the crowd. One critical issue in MCS is how to recruit users to fulfill more sensing tasks with budget restriction, while sharing data among tasks can be a credible way to improve the efficiency. The data-sharing based user recruitment problem under budget constraint in a realistic scenario is studied, where multiple tasks require homogeneous data but have various spatio-temporal execution ranges, meanwhile users suffer from uncertain future positions. The problem is formulated in a manner of probability by predicting user mobility, then a dynamic user recruitment algorithm is proposed to solve it. In the algorithm a greedy-adding-and-substitution(GAS) heuristic is repeatedly implemented by updating user mobility prediction in each time slot to gradually achieve the final solution. Extensive simulations are conducted using a real-world taxi trace dataset, and the results demonstrate that the approach can fulfill more tasks than existing methods.

Towards Developing Privacy-Preserved Data Security Approach(PP-DSA)in Cloud Computing Environment

In the present scenario of rapid growth in cloud computing models,several companies and users started to share their data on cloud servers.However,when the model is not completely trusted,the data owners face several security-related problems,such as user privacy breaches,data disclosure,data corruption,and so on,during the process of data outsourcing.For addressing and handling the security-related issues on Cloud,several models were proposed.With that concern,this paper develops a Privacy-Preserved Data Security Approach(PP-DSA)to provide the data security and data integrity for the out-sourcing data in Cloud Environment.Privacy preservation is ensured in this work with the Efficient Authentication Technique(EAT)using the Group Signature method that is applied with Third-Party Auditor(TPA).The role of the auditor is to secure the data and guarantee shared data integrity.Additionally,the Cloud Service Provider(CSP)and Data User(DU)can also be the attackers that are to be handled with the EAT.Here,the major objective of the work is to enhance cloud security and thereby,increase Quality of Service(QoS).The results are evaluated based on the model effectiveness,security,and reliability and show that the proposed model provides better results than existing works.

Making Data-Driven Transportation Decisions for Freight Operations

Using Louisiana’s Interstate system, this paper aims to demonstrate how data can be used to evaluate freight movement reliability, economy, and safety of truck freight operations to improve decision-making. Data mainly from the National Performance Management Research Data Set (NPMRDS) and the Louisiana Crash Database were used to analyze Truck Travel Time Reliability Index, commercial vehicle User Delay Costs, and commercial vehicle safety. The results indicate that while Louisiana’s Interstate system remained reliable over the years, some segments were found to be unreliable, which were annually less than 12% of the state’s Interstate system mileage. The User Delay Costs by commercial vehicles on these unreliable segments were, on average, 65.45% of the User Delay Cost by all vehicles on the Interstate highway system between 2016 and 2019, 53.10% between 2020 and 2021, and 70.36% in 2022, which are considerably high. These disproportionate ratios indicate the economic impact of the unreliability of the Interstate system on commercial vehicle operations. Additionally, though the annual crash frequencies remained relatively constant, an increasing proportion of commercial vehicles are involved in crashes, with segments (mileposts) that have high crash frequencies seeming to correspond with locations with recurring congestion on the Interstate highway system. The study highlights the potential of using data to identify areas that need improvement in transportation systems to support better decision-making.

大数据驱动的企业与用户互动创新:理论框架与前沿课题

大数据驱动的企业与用户互动创新已成为企业洞察市场和用户动态、提升创新敏捷性和适应性的重要模式,其关键因素和作用机制日益成为创新、营销和信息系统等领域的热点议题。结合系统性综述设计和关键词共现分析,梳理大数据情境下企业与用户互动创新的研究洞见。针对情境和工具如何影响领先用户与普通用户关系及二者的相互作用机制两个研究缺口,提出大数据驱动的企业与用户互动创新“条件—路径—绩效”理论框架。依据该理论框架,结合文献和案例证据,重点阐述在什么情境和工具条件下企业与领先用户、普通用户以何种互动创新路径影响创新产品和大数据合作资产,由此形成理论创新。分析结果表明,互动创新路径选择本质是在异质性情境和工具条件下,企业在创新新颖性—泛用性以及数据分析成本—收益之间的动态权衡;两类创新路径的动态变化由互动创新工具成熟度、创新主导权的变更所触发。最后,进一步提炼出大数据驱动的企业与用户互动创新理论研究的三类前沿课题,并讨论其理论和实践意义。

价值-制度-技术框架下图书馆用户数据治理优化路径研究

数智时代数据的开放利用是驱动社会及图书馆发展的核心资产,图书馆积累了各类有价值的用户数据,图书馆用户数据治理面临数据质量、数据伦理、数据法律等各种风险挑战。文章搭建了价值、制度、技术三维分析框架,提出了优化用户数据治理的多元化路径,以期实现图书馆用户数据的最大化价值。

细微粒度视域下国内外跨社交网络用户对齐研究综述

针对多源复杂网络用户管理所面临的冷启动、信息孤岛问题难以满足企业在不同社交网络中识别同一用户的困境,传统的仅使用单一维度特征提取源的跨社交网络用户对齐方法难以实现迅速、精准的用户识别,而使用细微粒度视域下的多维特征提取源可以有效耦合各类用户信息的优势以提升识别效率和精准度。论文首先以国内外跨社交网络用户对齐的相关文献为研究对象进行系统性分析,从细微粒度视域出发对文献主要内容进行汇总与梳理;然后逐一细化地对比分析以总结现有不同研究所采用的特征提取源之间的差异、优势和桎梏;最后从隐私数据集获取、多维数据集成和多源社交网络数据耦合3个方面来对未来用户对齐领域中的研究前沿方向与研究深入挖掘点作出展望。

用户视角下的教育数据要素交易:安全感知与建议

教育数据要素作为推动教育变革的核心动力,正通过数据要素市场交易推动教育数字化转型纵深发展。然而,目前教育数据要素市场在数据确权、市场监督以及市场交易等制度层面存在不足,阻碍教育数据要素价值的释放。用户作为数据要素市场的核心参与者,其对教育数据要素交易的安全感知与建议是推进数据要素交易有序发展的重要因素。研究面向四类不同用户群体开展了定向调查,发现用户总体对教育数据要素交易持积极态度,且具有较为全面的安全感知,不同用户群体则在教育数据要素基本态度、安全感知与安全建议上存在差异。为加快教育数据要素交易市场建设,可以从加强教育数据要素市场保障,提高高等院校用户市场参与意愿,构建教育数据要素安全体系,满足不同用户群体的安全需求,塑造多元教育数据要素交易市场,推动交易机制向需求侧转型等方面改善。

基于扩散模型生成数据重构的客户流失预测

在数据挖掘领域普遍存在数据不平衡影响到模型预测精度的问题,同时还存在未考虑用户隐私保护的问题.生成伪造数据是一种重要的解决方法,但在以结构化数据为主的场景中,由于存在数据特征维度多且不相关等特点,生成高质量的数据存在挑战.考虑到扩散模型在图像生成等任务中被成功应用,以客户流失预测为典型应用场景,尝试将扩散模型应用到客户流失预测任务中.针对该场景数据中的数值型特征和类别型特征,通过高斯扩散模型和多项式扩散模型获得生成数据,并对模型预测效果和数据隐私保护能力进行研究和分析.在多个领域的客户流失数据上进行了大量实验,探索应用生成数据对真实数据融合重构的可能性.实验结果表明基于扩散模型可生成高质量数据,且对多种预测方法均有一定提升,可实现缓解数据不平衡问题.同时,基于扩散模型生成的数据分布更接近真实数据,具有应用于用户隐私保护的潜在价值.

考虑数据稀疏性的图书推荐协同过滤算法仿真

图书推荐算法易忽略数据稀疏性问题,导致推荐结果与用户感兴趣内容之间存在较大的偏差。在考虑数据稀疏性的基础上提出一种图书推荐协同过滤算法,对数据预处理,通过对用户和用户之间综合信任度分析,利用分布估计算法对用户兴趣建模;构建用户兴趣簇类集,划分用户兴趣,从中选择出与检索对象最接近的邻居;计算邻近项目得分,按照从大到小的顺序排列,排名靠前的资源项即为图书推荐结果。实验结果表明,所提方法在推荐500本图书时,用时在12s内,且降低了平均绝对误差和均方根误差,实现了最精准的图书推荐。

平台企业“免费盈利”模式演变及其资源配置原理

以“淘宝系”为代表的平台企业通过持续创新探索,形成与传统企业大相径庭的独特商业模式,尤其是基于免费定价策略的盈利模式不断演变,成就了平台企业跨界转变的汹涌浪潮。既有研究主要以双边用户非对称定价来解释平台企业免费向用户提供技术服务的现象,但该理论无法涵盖平台企业同时对买卖双方免费的情境,且忽视了“市价为零时,技术服务市场的稀缺资源如何配置?”这一重要问题。可从资源配置视角出发,重新研究平台企业“免费盈利”模式,通过研究其发展演变过程,发现“以服务换流量,再通过流量资源获利”是平台企业通过服务升级、版本分层、跨界协同等模式实现“免费盈利”的共性逻辑;通过对其内在的价值创造过程进行理论解构,发现因用户流量资源集聚所形成的市场交易的总量扩张效应、结构优化效应和效率提升效应是平台企业的盈利来源;通过对免费技术服务市场的供需决策过程进行理论剖析,发现平台企业技术服务市场的稀缺资源配置是由用户流量与数据来驱动。相较于传统市场价格机制,以需求为中心的供给优化已取代以价格为中心的供需相互调整,这是一种更为高效的资源配置方式。政府部门应当转变传统价格中心范式的市场监管思路,探索适用于平台企业“免费盈利”模式的新型监管方式,通过规范平台企业流量竞争行为,建立健全数据交易市场体系,进一步释放数字经济领域资源配置的效率潜力。

基于大数据分析的计算机网页设计系统研究

针对当前计算机网页设计中缺乏对用户行为的深入分析和个性化设计反馈的问题,文中提出了一种基于大数据分析的计算机网页设计系统。该系统旨在通过综合利用用户交互数据,提高网页设计的个性化程度和用户体验质量。通过引入大数据分析,该系统能从大量的用户交互数据中提取关键信息,为设计师提供数据支持。实验结果显示,该系统有效提高了网页设计的用户交互质量和满意度,证明了大数据分析在计算机网页设计中的应用价值。其为网页设计领域提供了一种新的工作模式,有助于设计师更好地理解用户需求,创造出更具吸引力和有效性的网页设计方案。

基于User-Ontology的图书馆用户数据挖掘研究

文摘鉴于目前图书馆用户数据挖掘精度与效率不高的问题,本文提出一种基于User-Ontology(用户本体)的图书馆用户数据挖掘的研究思路,从而在语义层面上实现对用户数据的挖掘。文章首先分析了图书馆现有用户数据挖掘中存在的主要不足,其次介绍了目前国内外不同领域用户本体的研究与构建情况,最后在构建图书馆通用用户本体的基础上,提出了基于用户本体的图书馆用户数据挖掘系统的优势、总体框架与功能构成。

动态用户需求下的产品数据体系设计与应用

目的为满足动态化用户对产品数据管理的需求,解决现有产品数据管理效率低下,底层数据杂乱,数据管理平台技术受限的问题,对现有产品数据体系进行设计,并开发产品数据平台,保证产品数据的规范化与个性化要求。方法从动态客户需求的视角出发,提出结合模糊Kano模型和模糊层次分析法(FAHP)的产品数据体系设计方法,首先通过Voice of the Customer(VoC)法对用户需求进行收集与分析;其次通过模糊Kano模型对收集到的用户需求进行因子分类,找到关键需求因子;再次过FAHP模型对关键需求因子进行重要性排序;最后通过聚类分析需求关联性对因子进行修正。结果得到产品数据体系的底层设计依据,应用于技术数据管理平台设计中,以效率指标量化评价设计成果。结论该方法能够显著提升数据清洁度与有效性,满足多方用户对于产品数据的管理需求,同时也为产品数据管理平台国产化的建设提供一种可行且可靠的思路。

面向多源异构数据的个性化搜索和推荐算法综述

高效精准的个性化搜索、推荐等服务可为人们生产生活带来极大便利,而随着互联网技术的迅猛发展,面向多源异构数据的个性化搜索和推荐任务逐渐变得日趋复杂,也是当前大数据分析及个性化服务领域的研究热点和难点.个性化搜索和推荐算法广泛收集多源异构数据,获取用户偏好信息,利用各类机器学习、深度学习等技术,构建用户兴趣偏好模型,预测用户偏好,推荐满足用户个性化需求和偏好的项目或内容,提升用户的使用体验和网站平台的商业利益.本文介绍面向多源异构数据的个性化搜索问题的数学描述,综述面向多源异构数据的个性化搜索和推荐算法的相关研究工作,包括:传统个性化搜索和推荐算法、融合多源异构数据的个性化搜索和推荐算法以及动态个性化搜索和推荐算法等相关研究现状,整理了算法常用数据集、性能评价指标及评估体系,进一步阐明了目前面向多源异构数据的个性化搜索和推荐方法的实际应用场景及今后研究的发展方向,并讨论了存在的不足及所面临的严峻挑战,期望为相关领域的研究人员提供有益帮助.

WEB远程服务器接口数据访问安全防护算法仿真

数据在远程服务接口传输的过程中,容易受到黑客的攻击,导致传输数据被盗取,为保证服务器的安全传输,提出WEB远程服务器接口数据访问安全防护算法。建立访问安全防护模型,利用安全代理和密钥授权中心,实现WEB远程服务器接口数据的安全存储。设计用户身份验证机制,保证WEB远程服务器接口数据在开放环境中的安全性。通过设计容错策略,在链路失效情况下实现WEB远程服务器接口数据的安全访问与防护。实验结果表明,所提算法的身份验证精度高、数据加解密完整度高。

无线网络智能化的探索和应用

随着5G网络的完善和5G流量的快速增长,如何满足网络连接的多样化需求,提升客户感知保障能力,成为网络运营智能化的关键点。介绍了某省联通在无线网络智能化、潜在贬损用户识别与感知提升方面所做的探索与实践,通过整合XDR话单和专业网管等多源数据,引入熵权法、分层二元评分法等科学算法模型,结合用户体验回访结果不断迭代优化算法,实现了对潜在贬损用户的有效识别,同时在感知驱动网络问题闭环解决方面也取得了较好的效果。