考虑评论质量的自注意力胶囊网络评分预测模型

基于评论文档的推荐系统普遍采用卷积神经网络识别评论的语义,但由于卷积神经网络存在“不变性”,即只关注特征是否存在,忽略特征的细节,卷积中的池化操作也会丢失文本中的一些重要信息;另外,使用用户项目交互的全部评论文档作为辅助信息不仅不会提升语义的质量,反而会受到其中低质量评论的影响,导致推荐结果并不准确。针对上述提到的两个问题,该文提出了自注意力胶囊网络评分预测模型(Self-Attention Capsule network Rate prediction, SACR),模型使用可以保留特征细节的自注意力胶囊网络挖掘评论文档,使用用户和项目的编号信息标记低质量评论,并且将二者的表示相融合用以预测评分。该文还改进了胶囊的挤压函数,从而得到更精确的高层胶囊。实验结果表明,SACR在预测准确性上较一些经典模型及最新模型均有显著的提升。     

Incremental deep forest for multi-label data streams learning

Multi-label learning has attracted many attentions. However, the continuous data generated in the fields of sensors, network access, etc., that is data streams, the scenario brings challenges such as real-time, limited memory, once pass. Several learning algorithms have been proposed for offline multi-label classification, but few researches develop it for dynamic multi-label incremental learning models based on cascading schemes. Deep forest can perform representation learning layer by layer, and does not rely on backpropagation, using this cascading scheme, this paper proposes a multi-label data stream deep forest (VDSDF) learning algorithm based on cascaded Very Fast Decision Tree (VFDT) forest, which can receive examples successively, perform incremental learning, and adapt to concept drift. Experimental results show that the proposed VDSDF algorithm, as an incremental classification algorithm, is more competitive than batch classification algorithms on multiple indicators. Moreover, in dynamic flow scenarios, the adaptability of VDSDF to concept drift is better than that of the contrast algorithm.

     

Analysis and Study on the Method of Extracting Words Without Dictionary

I. INTRODUCTION Automatic words extraction is always the hotspot and difficulty in methods which research Chinese information processing. Since 1980, domestic researching scholars have developed a great deal of study, and put forward many methods about automatic words extraction[1~5]. Seen from the form, phrase is the combination of the stable words, so in the context, the more times that close words appear at the same time, the more possibility they constitute a phrase. Therefore the fr…     

Online multi-label stream feature selection based on neighborhood rough set with missing labels

Pattern Analysis and Applications - Multi-label feature selection has been essential in many big data applications and plays a significant role in processing high-dimensional data. However, the...