使用BERT预训练语言模型获取评论文本的向量表示,通过Bi-GRU网络学习其中的语义特征,分别采用情感权重和注意力机制来为评论向量分配权重,动态调节其对用户特征和产品特征的影响程度,并以加权求和的方式获得用户特征和产品特征,最后利用DeepFM算法对用户特征和产品特征进行深度交互,预测产品推荐的评分值。[结果]在Amazon product data数据集上的实验结果表明,本文模型比基线模型的RMSE与MAE指标最多可降低24.43%和31.44%;使用情感权重为用户评论加权的方法相比注意力机制,在RMSE与MAE指标上最多可降低2.59%和3.89%。
☆17Mar 28, 2023Updated 3 years ago
Alternatives and similar repositories for review-based-recommendation
Users that are interested in review-based-recommendation are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- code for "Deep Learning for Sequential Recommendation: Algorithms, Influential Factors, and Evaluations"☆12Sep 7, 2020Updated 6 years ago
- 基于adaboost的SVM预测股票价格☆11Mar 4, 2018Updated 8 years ago
- amazon crawler、亚马逊分布式爬虫系统,已在生产环境中稳定运行两年,支撑每天5千万+爬取量,包含亚马逊前台所有数据的爬取,如亚马逊商品爬虫、亚马逊关键词爬虫(可指定邮编爬取)、亚马逊关键词广告爬虫、亚马逊5大榜单爬虫、亚马逊评论爬虫、亚马逊类目爬虫、亚马逊商户爬…☆15Sep 21, 2023Updated 3 years ago
- 经典推荐算法的python实现或使用,涵盖协同过滤、矩阵分解、gbdt+lr,以及Wide&Deep等深度推荐模型。☆10Feb 9, 2022Updated 4 years ago
- 采用了单纯的Softmax和CNN网络进行手写体数字识别,采用Keras框架的Sequential模型,里面涉及了卷积,池化,dropout,Dense,Flatten;对于手写数据集获取方式也有说明,而不再不是用已经被官方弃用的方法;数据处理以及可视化数据和结果;模型的编…☆12Feb 19, 2019Updated 7 years ago
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- 基于卷积神经网络的4位字母数字混合的验证码识别(可以部分识别手写字体,需要手写字训练集)☆10Jun 17, 2018Updated 8 years ago
- The complete codes of the paper "Multimodal Graph Contrastive Learning for Recommendation"☆10Mar 20, 2023Updated 3 years ago
- huaweictf2020-xctf高校网络安全挑战赛☆17Dec 28, 2020Updated 5 years ago