knowledgegraph (KG) representation learning aims at embedding triples in the form of vectors. Their semantic similarity can be expressed through the distance of those vectors, and thus easily be computed for further ...
详细信息
ISBN:
(纸本)9781450391641
knowledgegraph (KG) representation learning aims at embedding triples in the form of vectors. Their semantic similarity can be expressed through the distance of those vectors, and thus easily be computed for further application, such as knowledge completion. Early KG embedding methods, such as the TransE model, were mainly training with relationship of each individual triple, and ignore the relationships between multiple triples. Thus their representation result is subject to the integrity of the triples. Recently, some path-enhanced methods, such as PTransE and RPJE, adopt the path information composed of multiple triples as supplement relation information for training, which achieves better effect than those triples based methods. On the other hand, the performance of path-enhanced methods is still affected by the fact that they are hard to learn the symmetric relation pattern in both triples and paths. Thus we propose Path Rotation based KG Representation Learning method (PRRL), which maps entities and relations into complex vector space and defines both relations and paths (composed by sequence of relations) as a rotation from source entity to target entity. PRRL can model and infer a variety of relation patterns, including symmetry/antisymmetry, inversion and composition relations by representing the path with Hadamard product of relations. The results of experiment on multiple datasets show that PRRL is better than the baseline in the completion of the task of KG.
The knowledgegraph completion algorithm can make the knowledgegraph more complete and is currently a research hotspot in the field of artificial intelligence. The knowledgegraph completion model is mainly defined i...
详细信息
ISBN:
(纸本)9798400707674
The knowledgegraph completion algorithm can make the knowledgegraph more complete and is currently a research hotspot in the field of artificial intelligence. The knowledgegraph completion model is mainly defined in three aspects, the way of negative example generation, the design of scoring function and the design of loss function. The previous knowledgegraph completion models only rely on factual view data to predict the missing links between entities and ignore the valuable common sense knowledge, and there is invalid negative sampling in knowledge graph embedding techniques; on the other hand, the existing graph neural network-based knowledge graph embedding models mainly consider capturing the graph structure around entities, and the relational representation is only used to update the entity embedding, which may miss the potentially useful information about the relational structure of potentially useful information. To address the above challenges, this paper proposes a two-view graph neural network-based knowledgegraph completion model combined with common sense awareness. Common knowledge is first automatically extracted from fact triples with entity concepts to facilitate high-quality negative sampling, and then positive and weighted negative triples are fed into the two-view graph neural network-based knowledge graph embedding model to capture entity- and relationship-centric graph structures and learn vector representations of entities and relationships, and then the learned entity and relationship representations are fed into a weighted score function to return the final the final score. Extensive experimental and ablation studies on four datasets, FB15K, FB15K237, NELL995, and DBpedia-242, show that the model achieves better performance compared to the state-of-the-art models.
In SPARQL, only actual connected nodes and properties are retrieved as fact. However, many knowledgegraphs expressed in RDF are incomplete, with knowledge graph embedding and graph neural networks used to compensate ...
详细信息
ISBN:
(纸本)9781450395656
In SPARQL, only actual connected nodes and properties are retrieved as fact. However, many knowledgegraphs expressed in RDF are incomplete, with knowledge graph embedding and graph neural networks used to compensate for the incompleteness. In this short research paper, we propose a query language, TranSPARQL, a SPARQL extension, enables RDF fuzzy searching that uses these neural network-based link predictions. A variable can describe one predictor in the pattern of subject, property and object in TranSPARQL. Additionally, we describe two implementations; a monotonic implementation with TransE for link predictions and a hybrid implementation using TransE and graph neural networks.
The rapid growth of biomedical publications has presented significant challenges in the field of information retrieval. Most existing work focuses on document retrieval given explicit queries. However, in real applica...
详细信息
The rapid growth of biomedical publications has presented significant challenges in the field of information retrieval. Most existing work focuses on document retrieval given explicit queries. However, in real applications such as curated biomedical database maintenance, explicit queries are missing. In this paper, we propose a two-step model for biomedical information retrieval in the case that only a small set of example documents is available without explicit queries. Initially, we extract keywords from the observed documents using large pre-trained language models and biomedical knowledgegraphs. These keywords are then enriched with domain-specific entities. Information retrieval techniques can subsequently use the collected entities to rank the documents. Following this, we introduce an iterative Positive-Unlabeled learning method to classify all unlabeled documents. Experiments conducted on the PubMed dataset demonstrate that the proposed technique outperforms the state-of-the-art positive-unlabeled learning methods. The results underscore the effectiveness of integrating large language models and biomedical knowledgegraphs in improving zero-shot information retrieval performance in the biomedical domain.
Recommendation systems are designed to uncover users’ potential preferences and make recommendations. However, they often face challenges such as data sparsity and the cold start problem. Although the introduction of...
详细信息
Recommendation systems are designed to uncover users’ potential preferences and make recommendations. However, they often face challenges such as data sparsity and the cold start problem. Although the introduction of knowledgegraphs has partially addressed the issue of data sparsity, the challenge of cold start has not been effectively resolved. In this paper, a novel approach called Social Perception with graph Attention Network (SPGAT) for Recommendation is proposed. In SPGAT, we aim to leverage social perception to solve the cold start effectively for more accurate recommendations. The approach utilizes a multi-layer graph attention network to aggregate user preference features from collaborative knowledgegraphs and social perception graphs. By analyzing the social network of a new user, associated friend users can be identified. The interaction data of these friend users is then provided as side information to recommend to the new user. To handle one-to-many and many-to-many relations, we introduce the TransD graphembedding model, which maps different types of relations and entities to different spaces. To optimize the proposed SPGAT, self-adversarial negative sampling is utilized to implement entity and relation embedding and generate negative samples. Experimental results demonstrate that SPGAT has achieved superior performance compared to several advanced methods.
暂无评论