电商平台产品共同购买网络的影响因素研究

共同购买网络的推荐系统应用越加广泛，仅基于网络内生结构变量研究其共同购买链接的经济意义已有局限，故加入网络口碑这一外生变量，进行更全面的分析。采用了社会网络方法中指数随机图模型进行建模，主要围绕产品销售量、产品入度、差评率和评论数4个方面因素，探究其对共同购买网络中共同购买链接形成的影响。结果显示，销售量、产品入度和评论数对共同购买链接形成的影响呈正比关系，而差评率则会显著地降低产品共同购买的几率。该指数随机图构建出的共同购买网络为在线电商平台管理网络口碑和推荐系统优化设计提供有益参考。     

认同度修正下的近相邻改进推荐算法研究

如何融合多方因素准确地为用户提供个性化产品一直是关注的热点问题，由此，一个新的近相邻改进算法融入了大众化认同度和个性化认同度，利于更加高效地挖掘隐藏信息。实验结果表明，相对于传统近相邻算法，认同度修正算法虽然查全率小幅度上下波动，但其他多个评价指标都得到极大提升，假正率和深度有所减少，查准率、[F1]值和提升度得以增加，并且，受试者特征曲线和提升曲线也都说明此修正算法具有更为显著的推荐效果。     

基于LBSN动态异构网络的时间感知兴趣点推荐

随着基于位置社交网络(Location-Based Social Network,LBSN)的快速发展,兴趣点(Point-Of-Interest,POI)推荐可以帮助人们发现有趣的并吸引人的位置。针对签到数据的稀疏性和用户兴趣的动态性等挑战性问题,提出了基于LBSN动态异构网络的时间感知兴趣点推荐算法。在LBSN异构网络模式中增加会话节点类型。通过动态元路径,在用户和兴趣点语义关系之间有效地融入时间信息、位置信息和社交信息等。设置了用户-兴趣点之间的动态元路径集,并提出了动态路径实例的偏好度计算方法。采用矩阵分解模型对不同动态偏好矩阵进行矩阵分解。根据不同动态元路径的用户特征矩阵和兴趣点特征矩阵,获取用户在目标时间访问兴趣点的推荐列表。实验结果表明,与其他兴趣点推荐方法相比,所提方法在兴趣点推荐精确度上取得了较好的推荐结果,具有良好的应用前景。     

基于数据挖掘的开源电子资源访问行为统计模型

为了解决当前图书馆资源个性化推荐过程中存在推荐的准确率、召回率以及效率较低的问题,采用二维距离模型构建用户社区模型,用于描述访问用户与图书馆开源电子资源之间的关系,并对互联网用户需求和访问行为进行模糊规则推理.依据互联网用户属性和图书馆资源访问需求属性之间的模糊规则,建立图书馆开源电子资源访问行为统计模型,并利用该模型向用户提供个性化推荐服务.仿真结果表明,所建模型的推荐召回率高达98. 4%,推荐准确率为99. 2%,运行时间小于0. 04 s.所建模型能够为互联网用户提供准确、高效地图书馆资源个性化推荐服务.     

Graduate Student Literature Review: Current recommendations and scientific knowledge on dairy goat kid rearing practices in intensive production systems in Canada,the United States,and France

Dairy goat kid rearing is the foundation of future milk production, yet little is known on this topic. References available to producers are limited, making it more difficult for dairy goat farms to reach their full production potential. This review paper aimed to identify the current recommendations on dairy goat kid rearing practices for intensive production systems and to assess whether the different recommendations were based on scientific literature. Recommendations on dairy goat kid rearing practices, from birth to weaning inclusively, were presented and compared between countries under similar intensive production systems, including Canada, the United States, and France. The different areas of rearing investigated included kidding management, colostrum management, liquid and solid feeding, health management, disbudding, housing, weaning, and growth monitoring. Gaps in the literature were identified in all areas except for disbudding. More research on the topic of goat kid raising practices would be beneficial to refine and validate current recommendations.     

融合时间隐语义填充和子群划分的推荐算法

随着个性化推荐技术的发展，推荐系统面临着越来越多的挑战。传统的推荐算法通常存在数据稀疏性和推荐精度低等问题。针对以上问题，提出了一种融合时间隐语义填充和子群划分的推荐算法[K]-TLFM（Time Based Latent Factor Model Integrated with [k]-means）。该算法利用融合时间因素的隐语义模型对原始用户物品评分矩阵缺失项进行填充，避免了用全局平均值或者用户/物品平均值补全矩阵带来的误差，有效缓解了数据稀疏性问题，同时融合时间因素有效地刻画了用户偏好随时间的变化；完成评分矩阵缺失项填充后，基于二分[k]-means聚类算法将偏好、兴趣特征相似的对象划分到同一个子群中，在目标用户所属的子群中基于选定的协同过滤算法为用户产生推荐列表，提高了推荐效率和准确性。在MovieLens和Netflix数据集上对该算法的推荐性能进行了对比实验，结果表明该算法具有更高的推荐精度。     

基于多特征融合的电影推荐系统

协同过滤算法（CF）根据用户-物品的评分矩阵做推荐，未考虑物品自身属性。本文将MovieLens数据集上的电影属性，作为影响推荐结果的因素，融合电影的简介、评论、评分、导演和演员等多种因素，进行推荐。使用CNN（卷积神经网络）和Word2Vec（Word to Vector,词向量模型）处理电影简介；使用AFINN（Finn rup Nielsen情感词典）处理评论，并对结果进行映射；对导演和演员数据进行建模，得到该因素下的预测评分，最后将各因素下的结果进行加权融合，通过调整权重，得到最佳效果。经验证，该方法的推荐性能优于传统的CF算法。     

Interventions for long-term software security: Creating a lightweight program of assurance techniques for developers

Though some software development teams are highly effective at delivering security, others either do not care or do not have access to security experts to teach them how. Unfortunately, these latter teams are still responsible for the security of the systems they build: systems that are ever more important to ever more people. We propose that a series of lightweight interventions, six hours of facilitated workshops delivered over three months, can improve a team's motivation to consider security and awareness of assurance techniques, changing its security culture even when no security experts are involved. The interventions were developed after an Appreciative Inquiry and Grounded Theory survey of security professionals to find out what approaches work best. We tested the interventions in a participatory action research field study where we delivered the workshops to three software development organizations and evaluated their effectiveness through interviews beforehand, immediately afterwards, and after twelve months. We found that the interventions can be effective with teams with limited or no security experience and that improvement is long-lasting. This approach and the learning points arising from the work here have the potential to be applied in many development teams, improving the security of software worldwide.     

CineCubes: Aiding data workers gain insights from OLAP queries

In this paper we demonstrate that it is possible to enrich query answering with a short data movie that gives insights to the original results of an OLAP query. Our method, implemented in an actual system, CineCubes, includes the following steps. The user submits a query over an underlying star schema. Taking this query as input, the system comes up with a set of queries complementing the information content of the original query, and executes them. For each of the query results, we execute a set of highlight extraction algorithms that identify interesting patterns and values in the data of the results. Then, the system visualizes the query results and accompanies this presentation with a text commenting on the result highlights. Moreover, via a text-to-speech conversion the system automatically produces audio for the constructed text. Each combination of visualization, text and audio practically constitutes a movie, which is wrapped as a PowerPoint presentation and returned to the user.     

Augmented context-based recommendation service framework using knowledge over the Linked Open Data cloud

This research proposes ACARDS (Augmented-Context bAsed RecommenDation Service) framework that is able to utilize knowledge over the Linked Open Data (LOD) cloud to recommend context-based services to users. To improve the level of user satisfaction with the result of the recommendation, the ACARDS framework implements a novel recommendation algorithm that can utilize the knowledge over the LOD cloud. In addition, the noble algorithm is able to use new concepts like the enriched tags and the augmented tags that originate from the hashtags on the SNSs materials. These tags are utilized to recommend the most appropriate services in the user’s context, which can change dynamically. Last but not least, the ACARDS framework implements the context-based reshaping algorithm on the augmented tag cloud. In the reshaping process, the ACARDS framework can recommend the highly receptive services in the users’ context and their preferences. To evaluate the performance of the ACARDS framework, we conduct four kinds of experiments using the Instagram materials and the LOD cloud. As a result, we proved that the ACARDS framework contributes to increasing the query efficiency by reducing the search space and improving the user satisfaction on the recommended services.