本文基本策略：GRU + KB (Knowledge Base，利用Key-Value进行查询和更新)
具体做法：
1、利用GRU为物品会话序列建模，每一个物品都有一个初始表达q，最终输出h
2、利用KB构建物品和属性之间的实体关系表达（TRANSE模型||e_1+r-e_2||），其中e是物品实体表达，r为关系表达（关系可以理解为属性名，如导演、价格等；而实体又可以理解为对应于这个属性名的属性值）
3、将属性表达（即r）设置成Key-Value NN中的K值，被所有用户共享（同一个领域，物品属性数量和种类基本上是稳定的），而Key-Value NN中的V，是特定于单个用户的、根据用户访问过的物品e进行更新。假设有(K_a, V_a)以及用户新访问的物品e_1，且在KB中存在（e_1, r_a, e_2), 即实体e_1的r_a属性值为e_2，所以就可以用e_2的表达来更新V_a.而e_2表达的估计可以是e_1+r_a（TRANSE，为什么不直接用e_2，因为对应于(e_1,r_a)的e_2可能存在多个，例如一部电影有多个演员，而“演员”这个属性名只有一个）。具体的更新方式e_1+r_a和之前的V_a加权求和
4、解决了以上问题，就可以利用h和K来检索V，最后将V加权聚合成m
5、最终用户表达：concate(h,m)，物品表达：concate(q,e)，并采用BPR-loss学习参数，其中q, e, r(K)都是提前利用其他模型训练好的，所以主要训练的参数为GRU以及V。
| 文献题目 | 去谷歌学术搜索 | ||||||||||
| Improving Sequential Recommendation with Knowledge-Enhanced Memory Network | |||||||||||
| 文献作者 | Jin Huang | ||||||||||
| 文献发表年限 | 2018 | ||||||||||
| 文献关键字 | |||||||||||
| 序列推荐系统+ 知识图谱 | |||||||||||
| 摘要描述 | |||||||||||
| With the revival of neural networks, many studies try to adapt powerful sequential neural models, i.e., Recurrent Neural Networks (RNN), to sequential recommendation. RNN-based networks encode historical interaction records into a hidden state vector. Although the state vector is able to encode sequential dependency, it still has limited representation power in capturing complicated user preference. It is difficult to capture fine-grained user preference from the interaction sequence. Furthermore, the latent vector representation is usually hard to understand and explain. To address these issues, in this paper, we propose a novel knowledge enhanced sequential recommender. Our model integrates the RNN-based networks with Key-Value Memory Network (KV-MN). We further incorporate knowledge base (KB) information to enhance the semantic representation of KV-MN. RNN-based models are good at capturing sequential user preference, while knowledgeenhanced KV-MNs are good at capturing attribute-level user preference. By using a hybrid of RNNs and KV-MNs, it is expected to be endowed with both benefits from these two components. The sequential preference representation together with the attribute-level preference representation are combined as the final representation of user preference. With the incorporation of KB information, our model is also highly interpretable. To our knowledge, it is the first time that sequential recommender is integrated with external memories by leveraging large-scale KB information. | |||||||||||
