摘要
限制选择集的尺寸,往往会得到更好的结果。针对于网络约会而言,给予用户少量且多样的约会对象推荐,有利于网恋成功率的提高。首先根据采集的用户信息进行分类,包括表面信息:照片、姓名、身高、体重、年龄、月收入、住址、是否接受异地恋、ELO评分、择偶标准;深度信息:学历、性格;多样性信息:爱好;然后构建K均值聚类模型和协同过滤模型相结合的在线约会对象推荐模型,考虑到推荐对象多样性,所以先利用K均值聚类模型将所有用户分类并判断待匹配用户所属类别,然后,考虑推荐对象深度,所以运用协同过滤模型分别计算不同信息特征下,待匹配用户与其他用户之间的相似度,将三种特征相似度赋权重为0.2,、0.4、0.4,得到最后的总相似度。最后根据不同用户特征信息不同,得到不同的选择集尺寸,根据多次模拟得到临界相似度,具体原则为:小于临界相似度的排除,在待匹配用户所属类别中选择5个,其他类别中各选1个。通过此种方法便可得到最优的约会对象集。
Limiting the size of the selection set tends to yield better results.For online dating,give users a small number and various dating recommendations is advantageous to the net love success rate enhancement.Firstly,classify the user information collected,including surface information:photo,name,height,weight,age,monthly income,address,whether to accept long-distance relationship,ELO score and mate selection criteria;In-depth information:education background,personality;Diversity of information:hobbies;Then,an online dating object recommendation model is constructed,which is a combination of k-means clustering model and collaborative filtering model.Considering the diversity of the recommended object,so first classifies all users and determines the category of the user to be matched by using k-means clustering model,then considering the depth of the recommended object,so calculated respectively the similarity between user to be matched and other users using by using collaborative filtering model.The similarity of the three features is weighted as 0.2,0.4 and 0.4,calculate and get the final total similarity.Finally,according to different user's feature information,different selection set sizes are obtained,and critical similarity is obtained based on multiple simulations.The specific principle is:exclude the users whose similarity are less than critical similarity,select 5 users from categories of the user to be matched,and choose 1 users from each of the other categories.In this way,the optim al set of dating objects can be obtained.
作者
陈一鑫
汪风传
吴宇晗
CHEN Yi-xin;WANG Feng-chuan;WU Yu-han(School of Metallurgy and Energy,North China University of Science and Technology,Tangshan Hebei 063210,China;School of Mining Engineering,North China University of Science and Technology,Tangshan Hebei 063210,China;College of Science,North China University of Science and Technology,Tangshan Hebei 063210,China)
出处
《新一代信息技术》
2019年第3期24-30,共7页
New Generation of Information Technology
关键词
在线约会对象推荐模型
K均值聚类模型
协同过滤模型
相似度
Online dating object recommendation model
K-means clustering model
Collaborative filtering model
Similarity