In recent years, many researchers have leveraged structural information from dependency trees to improve Named Entity Recognition (NER). Most of their methods take dependency-tree labels as input features for NER model training. However, such dependency information is not inherently provided in most NER corpora, making the methods with low usability in practice. To effectively exploit the potential of word-dependency knowledge, motivated by the success of Multi-Task Learning on cross-domain NER, we investigate a novel NER learning method incorporating cross-domain Dependency Parsing (DP) as its auxiliary learning task. Then, considering the high consistency of word-dependency relations across domains, we present an unsupervised domain-adapted method to transfer word-dependency knowledge from high-resource domains to low-resource ones. With the help of cross-domain DP to bridge different domains, both useful cross-domain and cross-task knowledge can be learned by our model to considerably benefit cross-domain NER. To make better use of the cross-task knowledge between NER and DP, we unify both tasks in a shared network architecture for joint learning, using Maximum Mean Discrepancy(MMD). Finally, through extensive experiments, we show our proposed method can not only effectively take advantage of word-dependency knowledge, but also significantly outperform other Multi-Task Learning methods on cross-domain NER. Our code is open-source and available at https://github.com/xianghuisun/DADP.

Knowledge-based data service has become an emerging form of service in the world wide web (WWW). To ensure the service quality, a comprehensive knowledge base has to be constructed. Knowledge base integration is often a primary way to improve the completeness. In this paper, we focus on the fundamental problem in knowledge base integration, i.e., entity alignment (EA). EA has been studied for years. Traditional approaches focus on the symbolic features of entities and propose various similarity measures to identify equivalent entities. With recent development in knowledge graph representation learning, embedding-based entity alignment has emerged, which encodes the entities into vectors according to the semantic or structural information and computes the relatedness of entities based on the vector representation. While embedding-based approaches achieve promising results, we identify some important information that are not well exploited in existing works: 1) The neighboring entities contribute differently in the EA process, and should be carefully assigned the importance in learning the relatedness of entities; 2) The attribute values (especially the long texts) contain rich semantics that can build supplementary associations between entities. To this end, we propose SDEA - a Semantics Driven entity embedding method for Entity Alignment. SDEA consists of two modules, namely attribute embedding and relation embedding. The attribute embedding captures the semantic information from attribute values with a pre-trained transformer-based language model. The relation embedding selectively aggregates the semantic information from neighbors using a GRU model equipped with an attention mechanism. Both attribute embedding and relation embedding are driven by semantics, building bridges between entities. Experimental results show that our method significantly outperforms the state-of-the-art approaches on three benchmarks.

Record Matching in data engineering refers to searching for data records originating from the same entities across different data sources. In practice, the main challenge of record matching is that the amount of non-matches typically far exceeds the amount of matches. This is called imbalance problem, which notoriously affects efficiency and effectiveness of matching algorithms. To solve the imbalance problem, recently, density-based blocking algorithms have been studied and demonstrated an effective blocking performance. However, the efficiency of density-based blocking approaches is not good as their effectiveness. In this paper, we improve the efficiency of density-based blocking by exploiting the idea of pre-computing and pruning. Our approach optimizes the method of computing density to speed up the blocking process. Throughout experiments on real-world datasets, the proposed approach demonstrated a high performance on both blocking efficiency and blocking effectiveness.

One of the fundamental problem in data management and data integration fields is Record Matching, which refers to identifying records that relate to the same entities across different data sources. In recent literature, active learning has demonstrated to be effective for record matching. One of the key steps of active learning is to build a proper initial classifier, with which active learning algorithms can quickly locate informative examples for training accurate models. However, in this process, example labelling for model training is usually expensive. Even worse, if a weak initial classifier is used, the labelling cost can be significantly increased. In this paper, we propose an unsupervised algorithm to determine the initial classifier. The process of classifier initialization requires no labelling cost. Then on our proposed algorithm, we present an active sampling method for selecting informative examples. The experiments show that our approach achieves competitive learning performance with much less labelling cost than other approaches of active learning.

Blocking is a technique of filtering unlikely matched pairs for record matching, which aims to collect all pairs of records that relate to the same entities across different data sources. Blocking has been broadly adopted in data mining and database. However, for big data, there is no fast and effective blocking algorithm yet, because the number of candidate pairs is tremendous between large data sets. In this paper, we report on a probabilistic parallelisation of a recently proposed blocking that is a sequential algorithm for efficient record matching in single machines. Our approach runs blocking processes distributedly on partitioned input data. In order to reduce data exchange among those blocking processes, we adopt a probabilistic technique to assure that the processes can run independently and meanwhile the aggregated result is correct with respect to common metrics. Our experimental analysis endorses the advantage of our technique and shows its novel scalability on a Hadoop MapReduce system deployed physically in a cloud.