Computing and Information Systems

Over the last decade, task-oriented dialogue system has enabled canonical applications such as Amazon Alexa, Google Home, and Apple Siri to help humans accomplish specific tasks via natural language interfaces. Now while driving, we can simply ask Alexa to book a restaurant and check the weather forecast for us. Traditionally, these systems are built via the pipeline approach where different modules are separately trained and connected to generate responses. However, these systems suffer from the credit assignment issue, and requires labelled data for each component. Recently, end-to-end dialogue models have attracted increasing attention as they alleviate the drawbacks of the pipeline approach by directly mapping the user utterances to system responses. However, most existing end-to-end models leverage memory networks to generate responses, and they don't capture the structural information stored in the external knowledge base. In this thesis, we first propose a novel end-to-end dialogue model that incorporates graph structure from the knowledge base. Observing that existing end-to-end models are limited as it uses a single modality (e.g., text only) knowledge base and multimodal data (e.g., images) can provide more information about the entities of interest (e.g., restaurants), we next introduce a multimodal dialogue system that can leverage a multimodal knowledge base. Nowadays, most end-to-end dialogue models use pre-trained language models as the backbone and although they are found to be empirically effective, their black-box nature means these dialogue systems do not provide explicit reasoning. To this end, we propose a novel explainable neuro-symbolic architecture for improving the interpretability of response generation. Lastly, we found that pre-trained model based dialogue models exploit shortcuts in the training data, learning the dataset instead of the task. To tackle this challenge, we propose debiasing techniques based on contrastive learning and adversarial filtering to improve model robustness. Extensive experimental studies demonstrate the effectiveness and robustness of our dialogue systems.

Recommender systems play a key role in addressing information overload issues in many Web applications, such as e-commerce, social media platforms, and lifestyle apps. For example, the Amazon shopping website lists items that a user might be interested in at the top of the website. The core function of recommender systems is to predict how likely a user will select an item (e.g., purchase, click). Typical recommender systems leverage users' historical selected items (i.e., user-item interactions) to infer users' interests. For example, matrix factorization (MF), one of the most common recommendation algorithms, learns user and item representations by factorizing the user-item interaction matrix as the multiplication of a user matrix, and an item matrix. Besides user-item interactions, studies show that features (e.g., user/item attributes, contexts), especially feature interactions (e.g., the co-occurrences of features), are important side information that can significantly enhance the recommendation accuracy. Deep neural networks (DNNs) have achieved great success in many recent studies due to their powerful information analysis ability. As a result, recent recommendation models seek to leverage DNNs for better feature interaction modeling. However, they model feature interactions implicitly (i.e., model all feature interactions together in a black-box DNN model without knowing which interactions are modeled or how they are modeled). Recent studies have shown that implicit interaction learning is less effective in extracting useful information about feature interaction for accurate recommendations. In this thesis, we explore how to effectively leverage feature interactions and improve the performance of recommender systems. Specifically, we focus on modeling feature interactions in an explicit manner. Unlike the implicit modeling methods, explicit methods model each feature interaction individually, allowing to choose which feature interactions to model, and decide how to model each feature interaction for more accurate recommendations. Then, we focus on three challenging research questions in improving recommender systems through explicit feature interaction modeling. The first research question explores how to improve the performance of MF by enabling it to explicitly model feature interactions. MF learns user and item representations from user-item interactions, which has a drawback: it cannot consider feature interactions. This limits MF's potential to perform fine-grained analysis for more accurate predictions. Meanwhile, MF may encounter the cold-start problem (i.e., an item has too few historical user-item interactions to conduct an effective analysis). Focusing on this drawback, instead of factorizing the user-item interaction matrix, we propose to factorize multiple user-attribute interaction matrices to learn attribute representations. The final prediction is an aggregation of all the ratings predicted in the user-attribute matrices. Our proposed method achieves higher accuracy than MF-based methods, while resolving the cold-start problem. The second research question explores which feature interactions should be modeled in recommender systems. Existing recommendation algorithms consider all feature interactions to generate predictions. However, not all feature interactions are relevant to the recommendation prediction, and capturing irrelevant feature interactions may introduce noise and decrease the prediction accuracy. Therefore, we propose to detect a set of most relevant feature interactions (we formally define them as beneficial feature interactions in terms of prediction accuracy) and model only beneficial feature interactions for more accurate recommendation. We propose novel frameworks that leverage the relational reasoning ability of graph neural networks (GNNs) to achieve more effective explicit feature interaction modeling. Under these frameworks, beneficial feature interaction detection and recommendation prediction are achieved via an edge prediction task and a graph classification task, respectively. The third research question explores how to model different feature interactions in recommender systems. Existing recommendation algorithms model each feature interaction equally, neglecting their different impacts while performing the prediction. We explore how feature interactions can be categorized and modeled to fit their roles in recommendation. More specifically, for user attributes and item attributes (e.g., user gender, item color), we define two types of interactions: inner interactions for profile learning and cross interactions for preference matching. We propose a neural graph matching method, which is based on our proposed GNN-based interaction modeling framework, to model the two types of interactions so that can better analyze their impacts on recommendation predictions. For context features (e.g., weather, time), inspired by psychology, we leverage them to learn intrinsic factors and extrinsic factors that jointly influence users selection. Contrastive learning and disentanglement learning algorithms are leveraged to learn these factors. In summary, this thesis has made several contributions to explicit feature interaction modeling for improving recommender systems through: enabling feature interaction modeling in matrix factorization, detecting beneficial feature interactions, and categorizing and modeling different types of feature interactions to fit their roles in recommendation. All works are supported by theoretical analysis and empirical results.

Computing and Information Systems - Theses

Permanent URI for this collection

Filters

Date

Author

Subject

Type

Settings

Sort By

Results per page

Statistics

Citations

Search Results