Language independent semantic kernels for short-text classification

Short-text classification is increasingly used in a wide range of applications. However, it still remains a challenging problem due to the insufficient nature of word occurrences in short-text documents, although some recently developed methods which exploit syntactic or semantic information have enhanced performance in short-text classification. The language-dependency problem, however, caused by the heavy use of grammatical tags and lexical databases, is considered the major drawback of the previous methods when they are applied to applications in diverse languages. In this article, we propose a novel kernel, called language independent semantic (LIS) kernel, which is able to effectively compute the similarity between short-text documents without using grammatical tags and lexical databases. From the experiment results on English and Korean datasets, it is shown that the LIS kernel has better performance than several existing kernels.     

Risk analysis using FMEA: Fuzzy similarity value and possibility theory based approach

Fuzzy numerical technique for FMEA has been proposed to deal with the drawbacks of crisp FMEA and fuzzy rule based FMEA approaches. Fuzzy numerical approaches based on de-fuzzification also suffer from the drawback of providing arbitrary priority ranks of failure modes even when their membership functions overlap. To overcome this drawback we developed a new methodology integrating the concepts of similarity value measure of fuzzy numbers and possibility theory. Similarity value measure has been applied to group together failure modes having similar amount of risk value. The possibility theory has been used for checking for conformance guidelines. Two case studies have been shown to demonstrate the methodology thus developed. The proposed methodology is more robust in nature as it does not require arbitrary precise operations like de-fuzzification to prioritise the failure modes. Application of possibility theory is new to the domain of risk analysing using FMEA.     

Power coefficient as a similarity measure for memory-based collaborative recommender systems

E-commerce systems employ recommender systems to enhance the customer loyalty and hence increasing the cross-selling of products. However, choosing appropriate similarity measure is a key to the recommender system success. Based on this measure, a set of neighbors for the current active user is formed which in turn will be used later to recommend unseen items to this active user. Pearson correlation coefficient, the most popular similarity measure for memory-based collaborative recommender system (CRS), measures how much two users are correlated. However, statistic’s literature introduced many other coefficients for matching two sets (vectors) that may perform better than Pearson correlation coefficient. This paper explores Jaccard and Dice coefficients for matching users of CRS. A more general coefficient called a Power coefficient is proposed in this paper which represents a family of coefficients. Specifically, Power coefficient gives many degrees for emphasizing on the positive matches between users. However, CRS users have positive and negative matches and therefore these coefficients have to be modified to take negative matches into consideration. Consequently, they become more suitable for CRS research. Many experiments are carried out for all the proposed variants and are compared with the traditional approaches. The experimental results show that the proposed variants outperform Pearson correlation coefficient and cosine similarity measure as they are the most common approaches for memory-based CRS.     

一种基于区域综合特征的图像检索算法

针对基于内容的图像检索所面临的图像低级视觉特征和高级语义之间的语义鸿沟问题,提出一种基于区域的图像检索算法。在LUV颜色空间中使用K均值聚类算法进行图像分割,提取分割后各区域的颜色、形状和区域自相关特征构成区域的综合特征,采用二次型距离相似性度量方法完成图像之间相似性的计算。实验结果表明,该算法具有较好的图像检索性能,与MIRROR中各算法相比,使用平均归一化修正检索等级得到的检索性能提高了12%~47.8%。     

基于混合模型推荐算法的优化

现代电子商务系统用户和物品数目的日益增加使得User-Item矩阵变得越来越稀疏,再加上目前相似性度量方法均存在一定弊端,致使推荐系统的推荐质量降低了。针对传统混合模型推荐算法做了优化,其相似性度量方法由物品属性相似性和改进的修正余弦相似性线性组合而成,权重因子自动生成,考虑了用户评分尺度及用户活跃度对物品相似性的影响。为解决冷启动问题,使用用户基本信息获得用户间的相似度,各属性权重因子由SVDFeature计算得到。实验结果表明,该算法有效地提升了推荐系统的推荐质量,同时还有效解决了用户冷启动与物品冷启动问题。     

拉普拉斯稀疏编码的图像检索算法

由于稀疏编码中过完备的码本和独立的编码处理,在编码过程中图像块相似性信息被丢失。为了保留块与块之间的相似性信息,提出了拉普拉斯稀疏编码的图像检索算法。给定编码特征的初始稀疏码,计算拉普拉斯矩阵,将相似性合并到稀疏目标函数,结合特征符号搜索算法和黄金分割线搜索算法,逐个更新每个稀疏编码系数。实验表明,拉普拉斯稀疏编码增强了稀疏编码的鲁棒性,与SPM模型算法相比,拉普拉斯稀疏编码的图像检索算法有较高的准确率。     

优化多重过滤的序列查询算法研究

序列数据是一种重要的数据类型,在诸多领域都有应用,比如说文本、生物数据库以及Web访问日志等。在对该类型数据进行分析的时候,对于相关信息的获取一般都是通过相似性查询得到的。本文首先根据序列查询算法的特点,提出了SSQ_MF,也就是多重过滤算法。并在此基础上设计了最优过滤顺序模型和过滤集大小估计的相关实验。实验结果表明,SSQ_MF算法的查询性能优于单一过滤器算法和随机过滤顺序的多过滤器算法。     

Leveraging prior ratings for recommender systems in e-commerce

User ratings are the essence of recommender systems in e-commerce. Lack of motivation to provide ratings and eligibility to rate generally only after purchase restrain the effectiveness of such systems and contribute to the well-known data sparsity and cold start problems. This article proposes a new information source for recommender systems, called prior ratings. Prior ratings are based on users’ experiences of virtual products in a mediated environment, and they can be submitted prior to purchase. A conceptual model of prior ratings is proposed, integrating the environmental factor presence whose effects on product evaluation have not been studied previously. A user study conducted in website and virtual store modalities demonstrates the validity of the conceptual model, in that users are more willing and confident to provide prior ratings in virtual environments. A method is proposed to show how to leverage prior ratings in collaborative filtering. Experimental results indicate the effectiveness of prior ratings in improving predictive performance.     

Three-dimensional physical simulation and optimization of water injection of a multi-well fractured-vuggy unit

With complex fractured-vuggy heterogeneous structures, water has to be injected to facilitate oil production. However, the effect of different water injection modes on oil recovery varies. The limitation of existing numerical simulation methods in representing fractured-vuggy carbonate reservoirs makes numerical simulation difficult to characterize the fluid flow in these reservoirs. In this paper, based on a geological example unit in the Tahe Oilfield, a three-dimensional physical model was designed and constructed to simulate fluid flow in a fractured-vuggy reservoir according to similarity criteria. The model was validated by simulating a bottom water drive reservoir, and then subsequent water injection modes were optimized. These were continuous (constant rate), intermittent, and pulsed injection of water. Experimental results reveal that due to the unbalanced formation pressure caused by pulsed water injection, the swept volume was expanded and consequently the highest oil recovery increment was achieved. Similar to continuous water injection, intermittent injection was influenced by factors including the connectivity of the fractured-vuggy reservoir, well depth, and the injection–production relationship, which led to a relative low oil recovery. This study may provide a constructive guide to field production and for the development of the commercial numerical models specialized for fractured-vuggy carbonate reservoirs.     

一种使用Harris特征点的区域图像检索算法

为了克服图像分割技术的限制，提出了一种基干特征点匹配技术的图像检索算法。手工提取图像中的一块区域作为查询图像，然后使用Harris算子提取彩色特征点，并用相应的颜色特征对特征点进行表示，最后利用特征点匹配技术实现区域图像的检索。实验表明，该方法对于图像的亮度变化和几何变换具有很强的鲁棒性，可以有效提高检索准确率。