Consumer Psychology in the Digital Age:How Online Environments Shape Purchasing Habits

The advent of the digital age has profoundly changed consumers’mindsets and habits.The rapid development of e-commerce and the widespread use of mobile applications have created enormous demand for individual recommendation systems based on mass data.This system not only increases the convenience of purchases and conversions but also alters the purchasing behavior of consumers,leading them to make choices subconsciously.Potential risks associated with large-scale data sharing and usage have heightened consumer concerns regarding privacy,thereby weakening the foundational trust in platforms and deterring them from shopping.Additionally,the rapid growth of e-commerce in the digital age,coupled with changing market circumstances,has intensified psychological pressure on consumers,making their decision-making processes more complex and difficult.Furthermore,the program will explore issues related to improving customer experience,developing individual marketing strategies,and designing customer loyalty plans.It will also address questions of privacy in a digital environment,the dilemmas of excessive or disruptive consumption behavior,and the complexity and diversity of consumer behavior in the face of digital change.The objective of this study is to develop a fear study that will enable a better understanding of the impact of online shopping on consumer behavior and provide a strategic guide for retailers to meet the challenges and opportunities presented by the digital age.

Empirical Effects of Online Retailing Recommendations

Online recommendation solves the current information overload problem in the online retailing businesses. Given relevant products by adopting recommendation algorithms, online shoppers can save time on searching and browsing for contents that they are interested in. Hence, in the increasing interests of online retailers, an empirical study was conducted to light the effectiveness of different entitled recommendations reflect on online shoppers. Working with a simulated online shopping establishment, the findings provide online retailers important guidelines regarding online customers’ behaviors.

A Survey of Online Course Recommendation Techniques

With the development of information technology, online learning has gradually become an indispensable way of knowledge acquisition. However, with the increasing amount of data information, it is increasingly difficult for people to find appropriate learning materials from a large number of educational resources. The recommender system has been widely used in various Internet applications due to its high efficiency in filtering information, helping users to quickly find personalized resources from thousands of information, thereby alleviating the problem of information overload. In addition, due to its great use value, many new researches have been proposed in the field of recommender systems in recent years, but there are not many works on online course recommendation at present. Therefore, this paper aims to sort out the existing cutting-edge recommendation algorithms and the work related to online course recommendation, so as to provide a comprehensive overview of the online course recommender system. Specifically, we will first introduce the main technologies and representative work used in the online course recommender system, explain the advantages and disadvantages of various technologies, and finally discuss the future research direction of the online course recommender system.

基于Online Judge进行程序设计类课程实践的混合推荐

针对计算机程序设计类课程中学生在Online Judge平台上进行编程实践时的个性化表现,提出混合推荐方法的思想及实现方法,探讨综合考虑教学要求和学生个性化特点的学习路径推荐方法。

Context-Aware Practice Problem Recommendation Using Learners’ Skill Level Navigation Patterns

The use of programming online judges(POJs)has risen dramatically in recent years,owing to the fact that the auto-evaluation of codes during practice motivates students to learn programming.Since POJs have greater number of pro-gramming problems in their repository,learners experience information overload.Recommender systems are a common solution to information overload.Current recommender systems used in e-learning platforms are inadequate for POJ since recommendations should consider learners’current context,like learning goals and current skill level(topic knowledge and difficulty level).To overcome the issue,we propose a context-aware practice problem recommender system based on learners’skill level navigation patterns.Our system initially performs skill level navigation pattern mining to discover frequent skill level navigations in the POJ and tofind learners’learning goals.Collaborativefiltering(CF)and con-tent-basedfiltering approaches are employed to recommend problems in the cur-rent and next skill levels based on frequent skill level navigation patterns.The sequence similarity measure is used tofind the top k neighbors based on the sequence of problems solved by the learners.The experiment results based on the real-world POJ dataset show that our approach considering the learners’cur-rent skill level and learning goals outperforms the other approaches in practice problem recommender systems.

Influence of Online Comments on Tourist' Purchase Intention Based on Questionnaire Survey among Tourists in Tai'an City

Taking tourists in Tai'an City—an outstanding tourism city of China for example, this paper explored the influence of online comments(word-of-mouth effect) on tourists' intention of purchasing services of accommodation facilities and restaurants through analyzing 502 valid questionnaires. Then measures and suggestions were proposed for tourist enterprises improving online word-of-mouth marketing.

Prediction of Online Judge Practice Passing Rate Based on Knowledge Tracing

Programming ability has become one of the most practical basic skills,and it is also the foundation of software development.However,in the daily training experiment,it is difficult for students to find suitable exercises from a large number of topics provided by numerous online judge(OJ)systems.Recommending high passing rate topics with an effective prediction algorithm can effectively solve the problem.Directly applying some common prediction algorithms based on knowledge tracing could bring some problems,such as the lack of the relationship among programming exercises and dimension disaster of input data.In this paper,those problems were analyzed,and a new prediction algorithm was proposed.Additional information,which represented the relationship between exercises,was added in the input data.And the input vector was also compressed to solve the problem of dimension disaster.The experimental results show that deep knowledge tracing(DKT)with side information and compression(SC)model has an area under the curve(AUC)of 0.7761,which is better than other models based on knowledge tracing and runs faster.

职业学校线上线下混合式教学粘性研究

随着技术的不断发展和教育需求的变化,线上线下混合式教学模式将成为未来教育的重要趋势。本研究采用文献综述与实证分析相结合的研究方法,通过查阅相关文献,梳理总结混合式教学特点,分析学习粘性相关理论,研究混合式教学模式粘性设计理论,并根据丹麦克努兹教授的全视角学习理论结合职业学校学生特点,从动机、环境、内容三个维度探究混合式教学的粘性设计,最终根据职业教育教学规律和技术技能人才成长规律,给出课程设计建议。

基于机器学习算法的在线推荐系统设计与性能优化研究

传统运营方式已经难以应对日益复杂的业务场景。特别是电商领域,采用个性化推荐系统对于改善用户体验、促进商品销售、提高广告点击率具有重要意义。文章以电商场景为例,设计了基于机器学习算法的在线推荐系统,并对该系统进行了性能优化。

基于用户影响力感知的在线学习资源推荐方法

针对在线学习中推荐数据稀疏性和资源多样化需求等问题,构建了一套在线学习混合推荐策略。该研究提出了一种基于用户相似性、知识可信度和用户影响力评估的在线学习用户模型LIAM,以提高推荐效果。同时,采用动态直觉模糊DIF策略对LIAM模型进行优化,以提高模型的推荐准确性和可解释性。最后,该研究提出了一种基于自组织的推荐方法SOR,用于解决推荐结果的多样性和覆盖性问题,从而形成了一套完整的在线学习混合推荐策略。同时,采用Coursera数据集对推荐方法SOR进行性能验证,实验结果表明,该方法优于其他两种代表性推荐方法。推荐方法SOR有望为在线学习推荐系统提供更加准确和个性化的推荐服务,提升学习效果和用户体验。

基于PathSim的MOOCs知识概念推荐模型

大规模开放在线课程提供大规模开放式在线学习平台,为推进现代教育发挥关键作用。然而,减少用户学习盲区和改善用户体验方面的研究仍具有挑战性:交互数据稀疏;难以扩展到大型推荐任务上;用户需求不单由用户喜好决定,还受到不同教师、课程影响;以统一的方式对课程学习事件中不同类型实体及关系进行建模并不妥靠。基于此,引入相关性度量,依据全图结构信息计算各边权重,提出采用相关性度量算法PathSim进行邻域采样的知识概念推荐模型PathSimSage。各实体间相关性得分可在本地离线计算,将神经网络与传播过程分离,保证神经网络的堆叠层数和传播过程的独立性,大幅减少模型所需训练时间。在公开的MoocCube数据集上进行了综合实验,PathSimSage降低了不相关的信息甚至噪声的影响,解决随机游走采样所引发的高度节点偏差问题,并在一定程度上缓解了过平滑效应。

基于云计算的线上商品智能推荐系统设计与应用

受用户行为和商品属性的影响,线上商品推荐的可靠性难以得到保障,为此,设计基于云计算的线上商品智能推荐系统。将密集计算型ic5云服务器作为系统的硬件装置;在软件设计阶段,利用云计算技术对用户行为进行综合分析,并将其与商品属性进行匹配分析,确定最终的推荐结果。应用测试结果显示,该系统在不同数据集上的接受者操作特性曲线下面积(Area Under Curve,AUC)表现出了较高的稳定性,且均在0.88以上,表明该系统具有较高的应用价值。

基于在线评论信息的商业兴趣点推荐

在信息爆炸式增长的背景下,在线评论信息为消费者选择商业兴趣点决策提供了重要依据。依据数量庞大的在线评论信息,考虑消费者的主观风险偏好行为,对具有时空特征的商业兴趣点构建推荐模型,并利用Yelp平台上的真实数据集和Python软件对上述模型进行实验分析。

融合知识图谱的学习者个性化学习资源推荐

针对学习资源个性化推荐中存在的数据稀疏和冷启动问题,本文提出了一种基于知识图谱的学习者个性化学习资源推荐模型模型(LPRM).LPRM模型利用在线学习中学生与课程的历史交互信息以及在线课程的属性信息构建课程知识图谱,辅助课程资源的个性化推荐;针对RippleNet框架中实体关系传播未考虑实体影响力的问题,提出节点影响力计算模型衡量知识图谱中实体的影响力,构建LPRM模型框架得到学习者对学习资源的评分.大量对比实验结果表明,本文提出的基于知识图谱的学习者个性化学习资源推荐模型模在AUC和ACC评价指标上均表现出最优的性能,模型参数分析结果表明LPRM模型能有效地提升学习者学习资源个性化推荐性能,较好地缓解了数据稀疏和冷启动引起的学习者个性化资源推荐不准确等问题.

采用局部子图嵌入的MOOCs知识概念推荐模型

大规模开放在线课程(MOOCs)在减少用户学习盲区和改善用户体验方面已经有大量的研究,尤其是基于图神经网络的个性化课程资源推荐,但现有工作主要集中在固定或同质图上,容易受到数据稀疏问题的影响且难以扩展。在局部子图上使用图卷积,并结合扩展的矩阵分解(MF)模型来解决这一问题。首先,将异构图分解为多个基于元路径的子图,结合随机游走采样方法实现在采样节点富有影响力邻域的同时捕获实体之间复杂的语义关系,并在局部邻域上进行图卷积平滑各节点表示,实现高可扩展性;然后,使用注意力机制适应性地融合不同子图的上下文信息,更全面地构建用户偏好;最后,通过扩展矩阵分解优化模型参数,获得推荐列表。为了验证提出模型的性能,在公开的MOOCs数据集上进行对比实验,相较于最优基线,性能提升了2%,内存计算需求降低了近500%,缓解数据稀疏问题的同时仍具有较强的可扩展性。

基于知识子图与注意力机制的在线课程推荐模型

推荐系统可以帮助用户在海量的资源中筛选出满足其需求的项目,不断发展的推荐系统为在线教育提供了新的思路。在线课程资源推荐作为在线教育领域中的重要一环,目前存在课程资源过载和课程推荐结果缺乏可解释性的问题。对此,该文提出了一种基于知识子图与注意力机制的在线课程推荐模型,以利用知识子图进行推荐。有别于直接利用知识图谱进行推荐而忽略了知识表示不准确问题的模型,该模型首先采用Node2vec随机游走方法从知识图谱中提取连接用户-课程对的连通子图,然后通过分层注意网络对子图进行编码,以生成用于用户所需课程预测的子图嵌入,最后生成Top-N推荐课程列表,并给出模型的可解释性说明。为验证模型的有效性,以“中国大学MOOC(慕课)”上的数据为样本构建数据集,实验结果表明,相较于KGCN-PN、GAT、KGAT以及POCR模型,文中模型在NDCG、HR以及MRR评价指标上分别提升了10.6%,9.41%,13.7%。

基于Spark平台的并行化谱聚类算法的在线学习资源推荐

为了提高在线学习资源推荐的准确度,采用谱聚类用于学习资源的归类,将类别相似度高的资源推荐给用户,提出Spark平台的并行化谱聚类算法,提高资源推荐效率;首先提取在线学习资源及用户特征并初始化,建立谱聚类模型,在Spark平台上分别求解无向图的顶点相似度及归一化拉普拉斯系数;然后采用归一化分割划分子集,通过归一化割集优化方式求解类别特征,并对类别特征按行输出特征点;最后采用k均值算法对特征点进行聚类,获得聚类结果。结果表明,采用谱聚类算法并借助于Spark平台的计算优势,所提推荐方法比常用的在线学习资源推荐算法的准确率和覆盖率更高,在海量学习资源的实时推荐方面具有较高适应度。

基于多维度特征融合的网络课程推荐方法构建研究

现有网络课程推荐的相关研究较少考虑用户对不同维度课程特征偏好,针对此不足提出融合多维度特征的网络课程推荐方法。利用多种数据分析技术获取课程描述文本、课程关系数据和课程类别属性的特征表示,基于深度自编码器融合各维度特征表示以构建课程与用户特征画像,并将其输入深度神经网络实现课程推荐。实验证明,相较于基于单一维度特征方法、基于双重维度特征方法以及传统机器学习方法,该方法在Precision@K和mAP@K评价指标上均有明显提高,对促进个性化课程推荐的实践应用具有一定的指导意义。

考虑期望强度与属性补偿关系的新能源汽车在线推荐方法

基于消费者属性期望,结合柔性决策和补偿原则理论,定义期望强度与属性补偿关系,明确属性期望与期望强度融合方法以及产品属性补偿规则,提出考虑期望强度和属性补偿关系的新能源汽车在线筛选方法。结合证据理论与消费者风险偏好提出新能源汽车在线排序方法,实现在线推荐满足消费者期望的新能源汽车。研究发现:与考虑属性期望严格约束的推荐方法相比,同时考虑期望强度与属性补偿关系能够避免综合高性价比车型因部分消费者属性期望严格约束而被忽略;采用证据推理结合消费者风险偏好确定车型优先级,能够更加精准地为消费者推荐满足个性化期望的新能源汽车。

基于协同过滤和特征工程的图书推荐系统研究及云图构建

面向图书馆联机公共目录检索(OPAC)系统,提出一种基于协同过滤结合特征工程的加权图书推荐算法。所提算法运用协同过滤的思想,通过分析图书借阅历史,对兴趣相似的用户进行聚类。引入图书借阅次数、上架时间、排架相邻度、专业相关度等特征项,经过特征编码后,利用加权算法计算出图书推荐指数。此外,针对推荐系统的冷启动和稀疏性问题,提出解决策略。测试结果表明,所提算法实现了完整的图书推荐系统及推荐图书的云图构建。