Geographic Named Entity Recognition (GNER) focuses on extracting geographic entity names from text and classifying them into pre-defined categories. Previous methods have not paid much attention to identifying fine-grained categories of geographic entities in sparse data situations, thus remaining limited in serving various geographic applications. To address this limitation, this paper presents a fine-grained GNER task and proposes a fine-grained GNER model, LH-FGNER, which incorporates a prototype network and hierarchical contrastive learning to improve fine-grained GNER. Specifically, the model designs label-guided sentence-level prototypes to capture the contextual semantics of geographic entities. It introduces a hierarchy tree to guide the construction of prototypes in vector space, which utilizes the hierarchy as a priori knowledge to improve the discrimination of fine-grained categories. In addition, two datasets are constructed to support the study of the fine-grained GNER task. Experimental results show that the proposed model is superior to the baseline and is robust. This work provides a methodological reference for few-shot GNER, which can be used to facilitate various geographic applications with text.

Geographical entity representation learning (GERL) is the emerging approach that manages and represents geographical entities, which advances a wide range of geographical intelligent applications by mapping entities into a latent vector space. However, previous GERL methods ignore the semantics carried by events that are closely associated with geographical entities, resulting in partially missing semantics in the learned vectors of geographical entities. To fill this gap, this paper proposes an event-enhanced geographical entity representation learning (EGERL) method to incorporate geographical events toward improving GERL. Specifically, EGERL designs an event integration strategy to bridge geographical events and geographical entities. And, it develops an anchor identification algorithm to recognize geographical anchor entities with rich event information. In addition, it augments the connections between anchor entities and non-anchor entities with an enhanced graph to enrich the information dissemination of event semantics. Finally, EGERL derives a relation-aware encoding module to encode geographical entities and their complex relations into vectors. Experimental results show that EGERL outperforms the state-of-the-art methods on knowledge representation and graph representation learning, and is robust. The study provides a new exploration of learning entity representations by fusing external semantics, and provides methodological references for various intelligent applications.

Knowledge graph embedding aims to transform the entities and relations of triplets into the low-dimensional vectors. Previous methods are oriented towards the static knowledge graphs, in which all entities and relations are assumed to be known and only some unknown triplets need to be predicted. However, the real-world knowledge graphs can grow dynamically, and some new knowledge are often added. To embed the new knowledge into the space of original knowledge graph, the classic models have to perform the entire re-embedding with including the new and original knowledge. This causes heavy computational burden for embedding. To address this problem, this study proposes a new model of anchors-based incremental embedding (ABIE) to implement the dynamical embedding for the growing knowledge graph. According to ABIE, every knowledge graph has some key entities, called anchors, which can fix the embedding space of knowledge graph. When some new knowledge is added into the graph, only a few updated entities and relations are embedded into the embedding space with the help of anchors, and the entire re-embedding on the whole graph is not necessary. By this way, the computational burden of embedding caused by the growth of knowledge graph is reduced significantly.

Collaborative filtering (CF) is initiated by representing users and items as vectors and seeks to describe the relationship between users and items at a profound level, thus predicting users’ preferred behavior. To address the issue that previous research ignored higher-order geographical interactions hidden in users’ historical behaviors, this paper proposes a location-aware neural graph collaborative filtering model (LA-NGCF), which incorporates location information of items for improving prediction performance. The model characterizes the interactions between items based on spatial decay law from a graph perspective and designs two strategies to capture the interaction effects of users and items considering node heterogeneity. An optimized loss function with spatial distances of items is also developed in the model. Extensive experiments are conducted on three publicly available real-world datasets to examine the effectiveness of our model. Results show that LA-NGCF achieves competitive performances compared with several state-of-the-art models, which suggests that location information of items is beneficial for improving the performance of personalized recommendations. This paper offers an approach to incorporate weighted interactions between items into CF algorithms and enriches the methods of utilizing geographical information for artificial intelligence applications.

We propose the first join tree (JT) propagation architecture that labels the probability information passed between JT nodes in terms of conditional probability tables (CPTs) rather than potentials. By modeling the task of inference involving evidence, we can generate three work schedules that are more time-efficient for LAZY propagation. Our experimental results, involving five real-world or benchmark Bayesian networks (BNs), demonstrate a reasonable improvement over LAZY propagation. Our architecture also models inference not involving evidence. After the CPTs identified by our architecture have been physically constructed, we show that each JT node has a sound, local BN that preserves all conditional independencies of the original BN. Exploiting inference not involving evidence is used to develop an automated procedure for building multiply sectioned BNs. It also allows direct computation techniques to answer localized queries in local BNs, for which the empirical results on a real-world medical BN are promising. Screen shots of our implemented system demonstrate the improvements in semantic knowledge.

Most approaches to mining association rules implicitly consider the utilities of the itemsets to be equal. We assume that the utilities of itemsets may differ, and identify the high utility itemsets based on information in the transaction database and external information about utilities. Our theoretical analysis of the resulting problem lays the foundation for future utility mining algorithms.

The discovery of FDs from databases has recently become a significant research problem. In this paper, we propose a new algorithm, called FD-Mine. FD-Mine takes advantage of the rich theory of FDs to reduce both the size of the dataset and the number of FDs to be checked by using discovered equivalences. We show that the pruning does not lead to loss of information. Experiments on 15 UCI datasets show that FD-Mine can prune more candidates than previous methods.