Knowledge-Enhanced Conversational Recommendation via Transformer-Based Sequential Modeling

Conversational recommender systems (CRSs) often utilize external knowledge graphs (KGs) to introduce rich semantic information and recommend relevant items through natural language dialogues. However, original KGs employed in existing CRSs are often incomplete and sparse, which limits the reasoning capability in recommendation. Moreover, only few of existing studies exploit the dialogue context to dynamically refine knowledge from KGs for better recommendation. To address the above issues, we propose the Variational Reasoning over Incomplete KGs Conversational Recommender (VRICR). Our key idea is to incorporate the large dialogue corpus naturally accompanied with CRSs to enhance the incomplete KGs; and perform dynamic knowledge reasoning conditioned on the dialogue context. Specifically, we denote the dialogue-specific subgraphs of KGs as latent variables with categorical priors for adaptive knowledge graphs refactor. We propose a variational Bayesian method to approximate posterior distributions over dialogue-specific subgraphs, which not only leverages the dialogue corpus for restructuring missing entity relations but also dynamically selects knowledge based on the dialogue context. Finally, we infuse the dialogue-specific subgraphs to decode the recommendation and responses. We conduct experiments on two benchmark CRSs datasets. Experimental results confirm the effectiveness of our proposed method.

Although paths of user interests shift in knowledge graphs (KGs) can benefit conversational recommender systems (CRS), explicit reasoning on KGs has not been well considered in CRS, due to the complex of high-order and incomplete paths. We propose CRFR, which effectively does explicit multi-hop reasoning on KGs with a conversational context-based reinforcement learning model. Considering the incompleteness of KGs, instead of learning single complete reasoning path, CRFR flexibly learns multiple reasoning fragments which are likely contained in the complete paths of interests shift. A fragments-aware unified model is then designed to fuse the fragments information from item-oriented and concept-oriented KGs to enhance the CRS response with entities and words from the fragments. Extensive experiments demonstrate CRFR’s SOTA performance on recommendation, conversation and conversation interpretability.

Recommender systems have been attracting much attention from both academia and industry because of their ability to capture user interests and generate personalized item recommendations. As the life pace in contemporary society speeds up, traditional recommender systems are inevitably limited by their disconnected interaction styles and low adaptivity to users’ evolving demands. Consequently, conversational recommender systems emerge as a prospective research area, where an intelligent dialogue agent is integrated with a recommender system. Conversational recommender systems possess the ability to accurately understand end-users’ intent or request and generate human-like dialogue responses when performing recommendations. However, existing conversational recommender systems only allow the systems to ask users for more preference information, while users’ further questions and concerns about the recommended items (e.g., enquiring the location of a recommended restaurant) can hardly be addressed. Though the recent task-oriented dialogue systems allow for two-way communications, they are not easy to train because of their high dependence on human guidance in terms of user intent recognition and system response generation. Hence, to enable two-way human-machine communications and tackle the challenges brought by manually crafted rules, we propose Conversational Recommender System with Adversarial Learning (CRSAL), a novel end-to-end system to tackle the task of conversational recommendation. In CRSAL, we innovatively design a fully statistical dialogue state tracker coupled with a neural policy agent to precisely capture each user’s intent from limited dialogue data and generate conversational recommendation actions. We further develop an adversarial Actor-Critic reinforcement learning approach to adaptively refine the quality of generated system actions, thus ensuring coherent human-like dialogue responses. Extensive experiments on two benchmark datasets fully demonstrate the superiority of CRSAL on conversational recommendation tasks.

Growing interests have been attracted in Conversational Recommender Systems (CRS), which explore user preference through conversational interactions in order to make appropriate recommendation. However, there is still a lack of ability in existing CRS to (1) traverse multiple reasoning paths over background knowledge to introduce relevant items and attributes, and (2) arrange selected entities appropriately under current system intents to control response generation. To address these issues, we propose CR-Walker in this paper, a model that performs tree-structured reasoning on a knowledge graph, and generates informative dialog acts to guide language generation. The unique scheme of tree-structured reasoning views the traversed entity at each hop as part of dialog acts to facilitate language generation, which links how entities are selected and expressed. Automatic and human evaluations show that CR-Walker can arrive at more accurate recommendation, and generate more informative and engaging responses.

Incorporating knowledge graphs (KGs) into recommender systems to provide explainable recommendation has attracted much attention recently. The multi-hop paths in KGs can provide auxiliary facts for improving recommendation performance as well as explainability. However, existing studies may suffer from two major challenges: error propagation and weak explainability. Considering all paths between every user-item pair might involve irrelevant ones, which leads to error propagation of user preferences. Defining meta-paths might alleviate the error propagation, but the recommendation performance would heavily depend on the pre-defined meta-paths. Some recent methods based on graph convolution network (GCN) achieve better recommendation performance, but fail to provide explainability. To tackle the above problems, we propose a novel method named K nowledge-aware R easoning with G raph C onvolution N etwork (KR-GCN). Specifically, to alleviate the effect of error propagation, we design a transition-based method to determine the triple-level scores and utilize nucleus sampling to select triples within the paths between every user-item pair adaptively. To improve the recommendation performance and guarantee the diversity of explanations, user-item interactions and knowledge graphs are integrated into a heterogeneous graph, which is performed with the graph convolution network. A path-level self-attention mechanism is adopted to discriminate the contributions of different selected paths and predict the interaction probability, which improves the relevance of the final explanation. Extensive experiments conducted on three real-world datasets show that KR-GCN consistently outperforms several state-of-the-art baselines. And human evaluation proves the superiority of KR-GCN on explainability.

Conversational recommender systems (CRS) enable the traditional recommender systems to explicitly acquire user preferences towards items and attributes through interactive conversations. Reinforcement learning (RL) is widely adopted to learn conversational recommendation policies to decide what attributes to ask, which items to recommend, and when to ask or recommend, at each conversation turn. However, existing methods mainly target at solving one or two of these three decision-making problems in CRS with separated conversation and recommendation components, which restrict the scalability and generality of CRS and fall short of preserving a stable training procedure. In the light of these challenges, we propose to formulate these three decision-making problems in CRS as a unified policy learning task. In order to systematically integrate conversation and recommendation components, we develop a dynamic weighted graph based RL method to learn a policy to select the action at each conversation turn, either asking an attribute or recommending items. Further, to deal with the sample efficiency issue, we propose two action selection strategies for reducing the candidate action space according to the preference and entropy information. Experimental results on two benchmark CRS datasets and a real-world E-Commerce application show that the proposed method not only significantly outperforms state-of-the-art methods but also enhances the scalability and stability of CRS.

Conversational recommender systems (CRSs) aim to capture user preferences and provide personalized recommendations through interactive natural language interaction. The recent advent of large language models (LLMs) has revolutionized human engagement in natural conversation, driven by their extensive world knowledge and remarkable natural language understanding and generation capabilities. However, introducing LLMs into CRSs presents new technical challenges. Directly prompting LLMs for recommendation generation requires understanding a large and evolving item corpus, as well as grounding the generated recommendations in the real item space. On the other hand, generating recommendations based on external recommendation engines or directly integrating their suggestions into responses may constrain the overall performance of LLMs, since these engines generally have inferior representation abilities compared to LLMs. To address these challenges, we propose an end-to-end large-scale CRS model, named as ReFICR, a novel LLM-enhanced conversational recommender that empowers a retrievable large language model to perform conversational recommendation by following retrieval and generation instructions through lightweight tuning. By decomposing the complex CRS task into multiple subtasks, we formulate these subtasks into two types of instruction formats: retrieval and generation. The hidden states of ReFICR are utilized for generating text embeddings for retrieval, and simultaneously ReFICR is fine-tuned to handle generation subtasks. We optimize the contrastive objective to enhance text embeddings for retrieval and jointly fine-tune the large language model objective for generation. Our experimental results on public datasets demonstrate that ReFICR significantly outperforms baselines in terms of recommendation accuracy and response quality. Our code is publicly available at the link: https://github.com/yt556677/ReFICR.

A major task of research in conversational recommender systems is personalization. Critiquing is a common and powerful form of feedback, where a user can express her feature preferences by applying a series of directional critiques over the recommendations instead of providing specific preference values. Incremental Critiquing (IC) is a conversational recommender system that uses critiquing as a feedback to efficiently personalize products. The expectation is that in each cycle the system retrieves the products that best satisfy the user's soft product preferences from a minimal information input. In this paper, we present a novel technique that increases retrieval quality based on a combination of compatibility and similarity scores. Under the hypothesis that a user learns during the recommendation process, we propose two novel exponential Reinforcement Learning (RL) approaches for compatibility that take into account both the instant at which the user makes a critique and the number of satisfied critiques. Moreover, we consider that the impact of features on the similarity differs according to the preferences manifested by the user. We propose a Global Weighting (GW) approach that uses a common weight for nearest cases in order to focus on groups of relevant products. We show that our methodology significantly improves recommendation efficiency in four data sets of different sizes in terms of session length in comparison with state-of-the-art approaches. Moreover, our recommender shows higher robustness against noisy user data when compared to classical approaches.

Conversational recommender systems ( CRS s) aim to understand the information needs and preferences expressed in a dialogue to recommend suitable items to the user. Most of the existing conversational recommendation datasets are synthesized or simulated with crowdsourcing, which has a large gap with real-world scenarios. To bridge the gap, previous work contributes a dataset E-ConvRec, based on pre-sales dialogues between users and customer service staff in E-commerce scenarios. However, E-ConvRec only supplies coarse-grained annotations and general tasks for making recommendations in pre-sales dialogues. Different from it, we use real user needs as a clue to explore the E-commerce conversational recommendation in complex pre-sales dialogues, namely user needs-centric E-commerce conversational recommendation (UNECR).

In the past few years we observed a renewed interest in conversational recommender systems (CRS) that interact with users in natural language. Most recent research efforts use neural models trained on recorded recommendation dialogs between humans, supporting an end-to-end learning process. Given the user’s utterances in a dialog, these systems aim to generate appropriate responses in natural language based on the learned models. An alternative to such language generation approaches is to retrieve and possibly adapt suitable sentences from the recorded dialogs. Approaches of this latter type are explored only to a lesser extent in the current literature. In this work, we revisit the potential value of retrieval-based approaches to conversational recommendation. To that purpose, we compare two recent deep learning models for response generation with a retrieval-based method that determines a set of response candidates using a nearest-neighbor technique and heuristically reranks them. We adopt a user-centric evaluation approach, where study participants (N=60) rated the responses of the three compared systems. We could reproduce the claimed improvement of one of the deep learning methods over the other. However, the retrieval-based system outperformed both language generation based approaches in terms of the perceived quality of the system responses. Overall, our study suggests that retrieval-based approaches should be considered as an alternative or complement to modern language generation-based approaches.

In the era of information explosion, it is difficult for people to obtain their desired information effectively. In tourism, a travel recommender system based on big travel data has been developing rapidly over the last decade. However, most work focuses on click logs, visit history, or ratings, and dynamic prediction is absent. As a result, there are significant gaps in both dataset and recommender models. To address these gaps, in the first step of this study, we constructed two human-annotated datasets for the travel conversational recommender system. We provided two linked data sets, namely, interaction sequence and dialogue data sets. The usage of the former data set was done to fully explore the static preference characteristics of users based on it, while the latter identified the dynamics changes in user preference from it. Then, we proposed and evaluated BERT-based baseline models for the travel conversational recommender system and compared them with several representative non-conversational and conversational recommender system models. Extensive experiments demonstrated the effectiveness and robustness of our approach regarding conversational recommendation tasks. Our work can extend the scope of the travel conversational recommender system and our annotated data can also facilitate related research.

Conversational recommender systems aim at recommending the most relevant information for users based on textual or spoken dialogues, through which users can communicate their preferences to the system more efficiently. Argumentative conversational recommender systems represent a kind of deliberation dialogue in which participants share their specific beliefs in the respective representations of the common ground, to act towards a common goal. The goal of such systems is to present appropriate supporting arguments to their recommendations to show the interlocutor that a specific item corresponds to their manifested interests. Here, we present a cross-disciplinary argumentation-based conversational recommender model based on cognitive pragmatics. We also present a dialogue simulator to investigate the quality of the theoretical background. We produced a set of synthetic dialogues based on a computational model implementing the linguistic theory and we collected human evaluations about the plausibility and efficiency of these dialogues. Our results show that the synthetic dialogues obtain high scores concerning their naturalness and the selection of the supporting arguments.

There has been increasing research interest in inferring missing information from existing knowledge graphs (KGs) due to the emergence of a wide range of knowledge graph downstream applications such as question answering systems and search engines. Reasoning over knowledge graphs, which queries the correct entities only through a path consisting of multiple consecutive relations/hops from the starting entity, is an effective approach to do this task, but this topic has been rarely studied. As an attempt, the compositional training method equally treats the constructed multi-hop paths and one-hop relations to build training data, and then trains conventional knowledge graph completion models such as TransE in a compositional manner on the training data. However, it does not incorporate additional information along the paths during training, such as the intermediate entities and their types, which can be helpful to guide the reasoning towards the correct destination answers. Moreover, compositional training can only extend some existing models that can be composable, which significantly limits its applicability. Therefore, we design a novel model based on the recently proposed neural memory networks, which have large external memories and flexible writing/reading schemes, to address these problems. Specifically, we first introduce a single network layer, which is then used as the building block for a multi-layer neural network called TravNM, and a flexible memory updating method is developed to facilitate writing intermediate entity information during the multi-hop reasoning into memories. Finally, we conducted extensive experiments on large datasets, and the experimental results show the superiority of our proposed TravNM for reasoning over knowledge graphs with multiple hops.

A conversational recommender system iteratively shows a small set of options for its user to choose between. In order to select these options, the system may analyze the queries tried by the user to derive whether one option is dominated by others with respect to the user's preferences. This paper describes a framework for preference dominance. Two instances of the framework are developed for query suggestion in a conversational recommender system. The first instance of the framework is based on a basic quantitative preferences formalism, where products are compared using sums of weights of features. The second is a qualitative preference formalism, using a language that generalizes CP-nets, where models are a kind of generalized lexicographic order. A key feature of both methods is that deductions of preference dominance can be made efficiently, since this procedure needs to be applied for many pairs of products. We show that, by allowing the recommender to focus on undominated options, which are ones that the user is likely to be contemplating, both approaches can dramatically reduce the amount of advice the recommender needs to give to a user compared to what would be given by systems without this kind of reasoning.

Heavily pre-trained transformer models such as BERT have recently shown to be\nremarkably powerful at language modelling by achieving impressive results on\nnumerous downstream tasks. It has also been shown that they are able to\nimplicitly store factual knowledge in their parameters after pre-training.\nUnderstanding what the pre-training procedure of LMs actually learns is a\ncrucial step for using and improving them for Conversational Recommender\nSystems (CRS). We first study how much off-the-shelf pre-trained BERT "knows"\nabout recommendation items such as books, movies and music. In order to analyze\nthe knowledge stored in BERT's parameters, we use different probes that require\ndifferent types of knowledge to solve, namely content-based and\ncollaborative-based. Content-based knowledge is knowledge that requires the\nmodel to match the titles of items with their content information, such as\ntextual descriptions and genres. In contrast, collaborative-based knowledge\nrequires the model to match items with similar ones, according to community\ninteractions such as ratings. We resort to BERT's Masked Language Modelling\nhead to probe its knowledge about the genre of items, with cloze style prompts.\nIn addition, we employ BERT's Next Sentence Prediction head and\nrepresentations' similarity to compare relevant and non-relevant search and\nrecommendation query-document inputs to explore whether BERT can, without any\nfine-tuning, rank relevant items first. Finally, we study how BERT performs in\na conversational recommendation downstream task. Overall, our analyses and\nexperiments show that: (i) BERT has knowledge stored in its parameters about\nthe content of books, movies and music; (ii) it has more content-based\nknowledge than collaborative-based knowledge; and (iii) fails on conversational\nrecommendation when faced with adversarial data.\n