Synthesizing Aspect-Driven Recommendation Explanations from Reviews

@inproceedings{Le2020SynthesizingAR,
  title={Synthesizing Aspect-Driven Recommendation Explanations from Reviews},
  author={Trung-Hoang Le and Hady W. Lauw},
  booktitle={IJCAI},
  year={2020}
}
Explanations help to make sense of recommendations, increasing the likelihood of adoption. However, existing approaches to explainable recommendations tend to rely on rigid, standardized templates, customized only via fill-in-the-blank aspect sentiments. For more flexible, literate, and varied explanations covering various aspects of interest, we synthesize an explanation by selecting snippets from reviews, while optimizing for representativeness and coherence. To fit target users’ aspect… 

Figures and Tables from this paper

Graph-based Extractive Explainer for Recommendations
TLDR
A graph attentive neural network model is developed that seamlessly integrates user, item, attributes and sentences for extraction-based explanation and solves an integer linear programming problem to make the final selection of sentences.
On the Relationship between Explanation and Recommendation: Learning to Rank Explanations for Improved Performance
TLDR
This paper extends traditional item ranking to an item-explanation joint-ranking formalization to study if purposely selecting explanations could reach certain learning goals, e.g., improving recommendation performance.
Counterfactual Explainable Recommendation
TLDR
This paper proposes Counterfactual Explainable Recommendation (CountER), which takes the insights of counterfactual reasoning from causal inference for explainable recommendation and applies it on a black-box recommender system and evaluates the generated explanations on five real-world datasets.
Attention over Self-attention: Intention-aware Re-ranking with Dynamic Transformer Encoders for Recommendation
TLDR
This paper proposes a novel Intention-aware Re-ranking Model with Dynamic Transformer Encoder (RAISE), aiming to perform user-specific prediction for each target user based on her intentions, and introduces a dynamic transformer encoder (DTE) to capture user- specific intra-item relationships among item candidates.
Recommendation System with Reasoning Path Based on DQN and Knowledge Graph
TLDR
This paper uses reinforcement learning to recommend items to target users, and achieved a rather good result, and has added explanations for recommended items to give a better user experience.
From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AI
TLDR
This poster presents a meta-modelling system that automates the very labor-intensive and therefore time-consuming and expensive and therefore expensive and time-heavy and expensive process of cataloguing and annotating the text for this poster.
AmpSum: Adaptive Multiple-Product Summarization towards Improving Recommendation Captions
TLDR
An Adaptive Multiple-Product Summarization framework (AmpSum) that automatically and adaptively generates widget captions based on different recommended products to help customers to better understand the connection between the multiple recommendations and make faster product discovery is proposed.
Reinforcement Learning over Sentiment-Augmented Knowledge Graphs towards Accurate and Explainable Recommendation
TLDR
This paper proposes a novel recommendation framework based on KG integrated with sentiment analysis for more accurate recommendation as well as more convincing explanations and develops an interactive user interface presenting textual explanations.
Explainable Recommendation with Comparative Constraints on Product Aspects
TLDR
This work proposes to anchor reference items on the previously adopted items in a user's history to provide comparative explanations involving such items, and forms comparative constraints involving aspect-level comparisons between the target item and the reference items.
...
1
2
...

References

SHOWING 1-10 OF 53 REFERENCES
Justifying Recommendations using Distantly-Labeled Reviews and Fine-Grained Aspects
TLDR
This work proposes an ‘extractive’ approach to identify review segments which justify users’ intentions and designs two personalized generation models which can generate diverse justifications based on templates extracted from justification histories.
Explainable Recommendation via Multi-Task Learning in Opinionated Text Data
TLDR
A multi-task learning solution for explainable recommendation is developed via a joint tensor factorization that predicts not only a user's preference over a list of items, but also how the user would appreciate a particular item at the feature level, i.e., opinionated textual explanation.
Personalized Review Generation By Expanding Phrases and Attending on Aspect-Aware Representations
TLDR
Experimental results show that the model successfully learns representations capable of generating coherent and diverse reviews and discover those aspects that users are more inclined to discuss and bias the generated text toward their personalized aspect preferences.
Co-Attentive Multi-Task Learning for Explainable Recommendation
TLDR
This paper designs an encoder-selector-decoder architecture inspired by human’s information-processing model in cognitive psychology and proposes a hierarchical co-attentive selector to effectively model the cross knowledge transferred for both tasks.
Social Collaborative Viewpoint Regression with Explainable Recommendations
TLDR
This paper proposes a latent variable model, called social collaborative viewpoint regression (sCVR), for predicting item ratings based on user opinions and social relations, and uses so-called viewpoints, represented as tuples of a concept, topic, and a sentiment label from both user reviews and trusted social relations.
TriRank: Review-aware Explainable Recommendation by Modeling Aspects
TLDR
TriRank endows the recommender system with a higher degree of explainability and transparency by modeling aspects in reviews, and allows users to interact with the system through their aspect preferences, assisting users in making informed decisions.
Learning to Generate Product Reviews from Attributes
TLDR
An attention-enhanced attribute-to-sequence model to generate product reviews for given attribute information, such as user, product, and rating, and an attention mechanism to jointly generate reviews and align words with input attributes is presented.
Neural Rating Regression with Abstractive Tips Generation for Recommendation
TLDR
A deep learning based framework named NRT is proposed which can simultaneously predict precise ratings and generate abstractive tips with good linguistic quality simulating user experience and feelings.
Learning to Rank Features for Recommendation over Multiple Categories
TLDR
This paper proposes a novel model called LRPPM-CF, which is able to improve the performance in the tasks of capturing users' interested features and item recommendation by about 17%-24% and 7%-13%, respectively, as compared with several state-of-the-art methods.
A Reinforcement Learning Framework for Explainable Recommendation
TLDR
A reinforcement learning framework for explainable recommendation that can explain any recommendation model (model-agnostic) and can flexibly control the explanation quality based on the application scenario is designed.
...
1
2
3
4
5
...