基于深度学习的颜值估计与电商精准营销

Estimation of Facial Beauty Based on Deep Learning Algorithm and Precision Marketing in E-commerce

近年来,随着机器学习和人工智能领域的不断发展,使得人脸颜值估计的研究得到广泛关注。提出一种基于深度学习的颜值估计框架,利用深度卷积神经网络(Deep Convolutional Neural Network,DCNN)提取人脸图像的特征向量,并采用回归分析计算方法评估人脸颜值,为充分发挥深度卷积神经网络提取特征的能力,提出了优化后的人脸特征提取损失函数。最后,利用该颜值估计算法构建消费者颜值与服装购物偏好相关性模型。结果显示:消费者颜值与服装购物偏好存在一定的相关关系,即颜值越高的消费者越喜欢时尚款式和风格的服装。研究结论为电商企业设计出高度精准营销策略,输出个性化产品和服务提供可能。

In recent years,with the fast development of artificial intelligence and machine learning,the research on facial beauty has attracted great attention.The estimation framework of facial beauty based on deep learning is proposed in the article.The deep convolutional neural network(DCNN) is utilized to extract the feature vector of the face image,and the facial beauty is evaluated by the regression analysis method.In order to improve the efficiency of DCNN,an optimized face feature extraction loss function is established.Furthermore,the estimation algorithm of facial value is used to construct the correlation model between consumer′s facial beauty and clothing shopping preference.A certain correlation between the consumer′s facial beauty and the clothing shopping preference is found.Consumers who prefer fashion apparels are likely to get higher facial beauty score.With the estimation algorithm of facial beauty,e-commerce companies can effectively design high precision marketing strategy and offer personalized products and services.