首页>
外国专利>
METHOD FOR COLLABORATIVELY FILTERING INFORMATION IN USE OF PERSONALIZED REGRESSION WITH CONTEXT INFORMATION TO PREDICT PREFERENCE GIVEN BY USER OF ITEM TO THE ITEM AND COMPUTING APPARATUS USING THE SAME
METHOD FOR COLLABORATIVELY FILTERING INFORMATION IN USE OF PERSONALIZED REGRESSION WITH CONTEXT INFORMATION TO PREDICT PREFERENCE GIVEN BY USER OF ITEM TO THE ITEM AND COMPUTING APPARATUS USING THE SAME
The present invention relates to a method of purifying information in order to predict a user's preference given to the item, and a computing device using the method. According to the present invention, the computing device obtains the data ruic of the preference that the individual user u has given to the individual item i under the context parameter c, To minimize And the average Estimator of Where U denotes a set of individual users, I denotes a set of individual items, C denotes a set of individual context variables, and r uic denotes a set of individual users u (U, i, c) when the individual user u gives preference under the context variable c to the individual item i is defined as an element of R uic , which is a random variable indicating the preference given under the variable c. , And c denotes a set having Lt; ego, Lt; ego, And, λ U is to refer to a tuning parameter (tuning parameter) relating to U and, λ I refers to the tuning parameters of the I, λ k is refers to the tuning parameters of the context variable c k, and then, the computing estimate of the device, the average μ uic The residual And using the residual to calculate a covariance matrix for the preference of the user u Is calculated, Is dispensed the variation of each user-specific preferences related to the preferences of the user u, and then, the conditional expectation of R uic as the estimated preference data for a specific user u under the context of a variable c for the individual items at least one item of i, each of the Value .
展开▼
机译:本发明涉及一种纯化信息以预测用户对该物品的偏好的方法,以及使用该方法的计算设备。根据本发明,计算设备获得在上下文参数c下个体用户u对个体项i给予的偏好的数据规则,为了最小化并且其中U表示个体用户集合的平均估计器, I表示一组单独的项目,C表示一组单独的上下文变量,而r uic Sub>表示一组单独的用户u(U,i,c),当单个用户u在以下条件下给出偏好时单个项目i的上下文变量c被定义为R uic Sub>的 element Sub>,它是一个随机变量,指示在变量c下给出的优先级。 ,c表示具有Lt的集合;自我自我,并且,λ U Sub>是指与U相关的调整参数(调整参数),而λ I Sub>是I的调整参数,λ k Sub>是指上下文变量c k Sub>的调整参数,然后是设备的计算估计值,平均μ uic Sub>残差And使用残差来计算用户u的偏好的协方差矩阵。计算,分配与用户u的偏好相关的每个用户特定偏好的变化,然后分配R uic < / Sub>作为变量i的上下文中特定用户u的估计偏好数据,其中变量c至少i个项中的每一项,每个Value。
展开▼