基于知识的电子商务智能推荐系统平台设计

分析了传统推荐技术存在的不足,阐述了基于知识的推荐技术的特点及其发展。针对现有基于知识的电子商务推荐系统中存在的不足,提出了基于知识的电子商务智能推荐需要解决的基本问题,设计了基于知识的电子商务智能推荐平台的逻辑框架,并阐述了其工作原理。     

基于协同过滤的电子商务智能推荐方法研究

为获得更加理想的电子商务推荐结果,提出一种基于协同过滤的电子商务智能推荐方法.该方法收集电子商务用户相关信息,并对信息进行预处理,计算电子商务用户对项目评分,构建电子商务用户评分矩阵,采用余弦算法根据用户评分矩阵计算用户之间的相性度,基于用户相似度进行电子商务智能推荐.为了与其他方法进行比较,开展仿真实验.实验结果表明...     

网格中基于本体的服务和知识发现服务

管理和利用由网格应用和网格操作所产生的海量数据,以及对网格资源和网格服务的智能使用是新一代网格的两个主要特征。为此,新一代网格应当具有被称为网格智能的知识发现和知识管理的功能。本体和元数据是实现网格智能的两个基本要素。此外,语义建模是网格支持高层服务和动态服务发现与组成的必要条件。本文对这些新兴服务进行了描述．并在知识网格上实现了一个基于本体的知识发现服务。     

基于大数据的电商数据信息智能推荐服务及应用研究

随着大数据技术、5G移动网络、物联网和人工智能技术的快速发展,移动互联网和智能手机的普及,数据信息出现了过载现象,如何将数据信息有效地实现智能推荐已成为当前人工智能和大数据领域研究的重点。通过对大数据和电商数据信息智能推荐以及电子商务的概念进行梳理,给出了基于大数据的电商数据信息智能推荐服务及应用系统架构,使大数据在电商领域的应用价值得到了充分体现,对电子商务的进一步发展起到了推动作用。     

会展简报

宋斐 9月6日,“IBM智能化电子商务基础设施高峰论坛”在北京隆重召开。继去年对电子商务的未来进行了广泛深入的讨论和交流之后,今年的大会主要集中在如何针对电子商务新时代用户的新需求、新应用,打造面向未来的电子商务基础设施。来自IBM亚太地区各个部门的高层人士就网格计算、Linux和未来的存储蓝图等热点问题进行了深入的探讨。此次大会上最令人振奋的是IBM向业界展现了飞速发展的网格计算技术。据     

Blackboard推理方法的应用与实现

针对电子商务订单实时智能处理控制问题，采用人工智能领域复杂问题推理的黑板(blackboard)方法，组织电子商务订单实时智能处理控制与推理。考虑物流配送能力的电子商务订单实时智能处理是基于知识的多任务处理。为此，提出基于blackboard的推理处理方法，解决订单处理过程中涉及到的人工处理问题的经验和知识的运用和调度。该研究有助于提高电子商务订单处理的实时性、科学性和智能性。     

语义Web应用研究综述

介绍了语义Web的关键技术XML、RDF(S)和本体,并指出了语义Web技术的众多应用领域:知识管理、语义搜索、P2P、电子商务、电子政务、语义网格、Web挖掘、语义Web服务,智能信息Agent和语义门户等.从这些领域当前面临的困难和挑战出发,分析和论述了语义Web技术在这些应用领域的作用、应用前景和部分研究成果.     

大数据环境下智能推荐系统中协同过滤算法研究

随着互联网的普及以及电子商务的发展,各类网络应用需要大数据的支持,而信息过载是大数据应用最严重的问题之一,需要来自智能推荐系统的支持.分析了大数据环境下智能推荐系统的概念和应用,对协同过滤算法这一智能推荐经典算法进行了阐述,并介绍了其在大数据环境中的应用领域.     

面向知识网格的本体学习研究

网格计算正在从单纯的面向大型计算的分布式资源共享发展为一种面向服务的架构,以实现透明而可靠的分布式系统集成。网格智能是指如何获取、预处理、表示和集成不同层次的网格服务(如HTML/XML/RDF/OWL文档、服务响应时间和服务质量等)的数据和信息,并最终转换为有用的智能(知识)。因为高层知识将在未来的网格应用起到越来越重要的作用,本体是知识网格实现的关键。文章提出了一种实现从Web文档中本体(半)自动构建的本体学习框架WebOntLearn,并讨论了本体学习中领域概念的抽取、概念之间关系的抽取和分类体系的自动构建等关键技术。     

基于XML的语义知识表达和智能搜索

传统电子商务平台的控制集中性、系统非动态性和非智能性等缺陷越来越成为其发展的瓶颈.基于智能代理与Web服务技术的新一代电子商务平台克服了这些不足.在平台中,智能代理作为商务的主体,在商务活动中需要拥有必要的知识,并需与顾客进行智能化的知识交互.基于该问题,提出基于XML的语义知识表达和智能搜索.该知识表达和搜索基于XML技术和语义关系而建立.分析表明,它们可以有效提高电子商务平台的智能性.     

Suslin’s algorithms for reduction of unimodular rows

A well-known lemma of Suslin says that for a commutative ring $A$ if $(v_1\left(X\right),\dots ,v_n\left(X\right))\in {\left(A\left[X\right]\right)}^n$ is unimodular where $v_1$ is monic and $n\ge 3$, then there exist ${\gamma }_1,\dots ,{\gamma }_{\ell }\in E_{n-1}\left(A\left[X\right]\right)$ such that the ideal generated by $\mathrm{Res}(v_1,e_1.{\gamma }_1{}^t(v_2,\dots ,v_n)),\dots ,\mathrm{Res}(v_1,e_1.{\gamma }_{\ell }{}^t(v_2,\dots ,v_n))$ equals $A$. This lemma played a central role in the resolution of Serre’s Conjecture. In the case where $A$ contains a set $E$ of cardinality greater than $deg{v}_1+1$ such that $y-y^{\prime }$ is invertible for each $y\ne y^{\prime }$ in $E$, we prove that the ${\gamma }_i$ can simply correspond to the elementary operations $L_1\to L_1+y_i{\sum }_{j=2}^{n-1}{u}_{j+1}{L}_j$, $1\le i\le \ell =deg{v}_1+1$, where $u_1{v}_1+\cdots +u_n{v}_n=1$. These efficient elementary operations enable us to give new and simple algorithms for reducing unimodular rows with entries in $K[X_1,\dots ,X_k]$ to ${}^t(1,0,\dots ,0)$ using elementary operations in the case where $K$ is an infinite field. Another feature of this paper is that it shows that the concrete local–global principles can produce competitive complexity bounds.     

A new approach toward stabilization in a two-species chemotaxis model with logistic source

This paper aims at providing an alternative approach to study global dynamic properties for a two-species chemotaxis model, with the main novelty being that both populations mutually compete with the other on account of the Lotka–Volterra dynamics. More precisely, we consider the following Neumann initial–boundary value problem

in a bounded domain $\Omega ?{\mathbb{R}}^n,n\ge 1$, with smooth boundary, where $d_1,d_2,d_3,{\chi }_1,{\chi }_2,{\mu }_1,{\mu }_2,a_1,a_2$ are positive constants.When $a_1\in (0,1)$ and $a_2\in (0,1)$, it is shown that under some explicit largeness assumptions on the logistic growth coefficients ${\mu }_1$ and ${\mu }_2$, the corresponding Neumann initial–boundary value problem possesses a unique global bounded solution which moreover approaches a unique positive homogeneous steady state $(u^1,v^1,w^1)$ of above system in the large time limit. The respective decay rate of this convergence is shown to be exponential.When $a_1\ge 1$ and $a_2\in (0,1)$, if ${\mu }_2$ is suitable large, for all sufficiently regular nonnegative initial data $u_0$ and $v_0$ with $u_0?0$ and $v_0?0$, the globally bounded solution of above system will stabilize toward $(0,1,1)$ as $t\to \infty$ in algebraic.     

The constraints for the thermodynamic parameters of the type order–disorder transformation

A mathematical analysis of the constraints on the thermodynamic parameters of the compound energy formalism for an ${(A,B)}_m{(A,B)}_n\leftrightarrow {(A,B)}_{m+n}$ type order–disorder transformation is studied. Both the ordered phase and the disordered phase can be described using the two-sublattice model after the formula ${\left(A_{y_A^{\prime }}{B}_{y_B^{\prime }}\right)}_m{\left(A_{y_a^{″}}{B}_{y_B^{″}}\right)}_n$, with $y_i^{\prime }=y_i^{″}(i=A,B)$ for the disordered phase and $y_i^{\prime }\ne y_i^{″}$ for the ordered phase. The constraints between the parameters for the ordered state were derived. The thermodynamic characteristics of the transformation are illustrated using a mathematical method and the validity of the model and the constraints is tested.     

Finite time blow-up for a class of parabolic or pseudo-parabolic equations

In this paper, we study the initial boundary value problem for a class of parabolic or pseudo-parabolic equations:

$u_t?a\Delta u_t?\Delta u+bu=k\left(t\right)|u{|}^{p?2}u,(x,t)\in \Omega \times (0,T),$

where $a\ge 0$, $b>?{?}_1$ with ${?}_1$ being the principal eigenvalue for $?\Delta$ on $H_0^1\left(\Omega \right)$ and $k\left(t\right)>0$. By using the potential well method, Levine’s concavity method and some differential inequality techniques, we obtain the finite time blow-up results provided that the initial energy satisfies three conditions: (i) $J(u_0;0)<0$; (ii) $J(u_0;0)\le d\left(\infty \right)$, where $d\left(\infty \right)$ is a nonnegative constant; (iii) $0<J(u_0;0)\le C\rho \left(0\right)$, where $\rho \left(0\right)$ involves the $L^2$-norm or $H_0^1$-norm of the initial data. We also establish the lower and upper bounds for the blow-up time. In particular, we obtain the existence of certain solutions blowing up in finite time with initial data at the Nehari manifold or at arbitrary energy level.     

Blow-up analysis in quasilinear reaction–diffusion problems with weighted nonlocal source

In this paper, we consider the blow-up of solutions to a class of quasilinear reaction–diffusion problems

where $\Omega$ is a bounded convex domain in ${\mathbb{R}}^n(n\ge 2)$, weighted nonlocal source satisfies $a\left(x\right)f\left(u(x,t)\right)\le a_1+a_2{\left(u(x,t)\right)}^p{\left({\int }_{\Omega }{\left(u(x,t)\right)}^l\mathrm{d}x\right)}^m,$ and $a_1,a_2,p,l,$ and $m$ are positive constants. By utilizing a differential inequality technique and maximum principles, we establish conditions to guarantee that the solution remains global or blows up in a finite time. Moreover, an upper and a lower bound for blow-up time are derived. Furthermore, two examples are given to illustrate the applications of obtained results.     

Boundedness in the higher-dimensional quasilinear chemotaxis-growth system with indirect attractant production

This paper deals with the following quasilinear chemotaxis-growth system

in a smoothly bounded domain $\Omega ?{\mathbb{R}}^n(n\ge 3)$ under zero-flux boundary conditions. The parameters $\mu ,\delta$ and $\tau$ are positive and the diffusion function $D\left(u\right)$ is supposed to generalize the prototype $D\left(u\right)\ge D_0{u}^{\theta }$ with $D_0>0$ and $\theta \in \mathbb{R}$. Under the assumption $\theta >1?\frac{4}{n}$, it is proved that whenever $\mu >0$, $\tau >0$ and $\delta >0$, for any given nonnegative and suitably smooth initial data ($u_0$, $v_0$, $w_0$) satisfying $u_0?0$, the corresponding initial–boundary problem possesses a unique global solution which is uniformly-in-time bounded. The novelty of the paper is that we use the boundedness of the $\left|\right|v(?,t)|{|}_{W^{1,s}\left(\Omega \right)}$ with $s\in [1,\frac{2n}{n?2})$ to estimate the boundedness of $\left|\right|?v(?,t)|{|}_{L^{2q}\left(\Omega \right)}(q>1)$. Moreover, the result in this paper can be regarded as an extension of a previous consequence on global existence of solutions by Hu et al. (2016) under the condition that $D\left(u\right)\equiv 1$ and $n=3$.