Improving Items and Contexts Understanding with Descriptive Graph for Conversational Recommendation
This addresses the problem of sub-optimal recommendations in conversational systems for users, but it is incremental as it builds on existing methods by improving alignment.
The paper tackles the misalignment between items and contextual words in conversational recommender systems by proposing KLEVER, a framework that jointly models them in the same semantic space using a descriptive graph, resulting in superior performance, particularly when user input is limited.
State-of-the-art methods on conversational recommender systems (CRS) leverage external knowledge to enhance both items' and contextual words' representations to achieve high quality recommendations and responses generation. However, the representations of the items and words are usually modeled in two separated semantic spaces, which leads to misalignment issue between them. Consequently, this will cause the CRS to only achieve a sub-optimal ranking performance, especially when there is a lack of sufficient information from the user's input. To address limitations of previous works, we propose a new CRS framework KLEVER, which jointly models items and their associated contextual words in the same semantic space. Particularly, we construct an item descriptive graph from the rich items' textual features, such as item description and categories. Based on the constructed descriptive graph, KLEVER jointly learns the embeddings of the words and items, towards enhancing both recommender and dialog generation modules. Extensive experiments on benchmarking CRS dataset demonstrate that KLEVER achieves superior performance, especially when the information from the users' responses is lacking.