Learning Entity Research Articles

DACHA: A Dual Graph Convolution Based Temporal Knowledge Graph Representation Learning Method Using Historical Relation

Temporal knowledge graph (TKG) representation learning embeds relations and entities into a continuous low-dimensional vector space by incorporating temporal information. Latest studies mainly aim at learning entity representations by modeling entity interactions from the neighbor structure of the graph. However, the interactions of relations from the neighbor structure of the graph are neglected, which are also of significance for learning informative representations. In addition, there still lacks an effective historical relation encoder to model the multi-range temporal dependencies. In this article, we propose a d ual gr a ph c onvolution network based TKG representation learning method using h istorical rel a tions (DACHA). Specifically, we first construct the primal graph according to historical relations, as well as the edge graph by regarding historical relations as nodes. Then, we employ the dual graph convolution network to capture the interactions of both entities and historical relations from the neighbor structure of the graph. In addition, the temporal self-attentive historical relation encoder is proposed to explicitly model both local and global temporal dependencies. Extensive experiments on two event based TKG datasets demonstrate that DACHA achieves the state-of-the-art results.

Perturbation-based methods for explaining deep neural networks: A survey

• Overview of the black box problem in DNNs and XAI research. • Survey of perturbation-based attribution methods for different input data types. • Future directions for research on perturbation paradigm. Deep neural networks (DNNs) have achieved state-of-the-art results in a broad range of tasks, in particular the ones dealing with the perceptual data. However, full-scale application of DNNs in safety-critical areas is hindered by their black box-like nature, which makes their inner workings nontransparent. As a response to the black box problem, the field of explainable artificial intelligence (XAI) has recently emerged and is currently rapidly growing. The present survey is concerned with perturbation-based XAI methods, which allow to explore DNN models by perturbing their input and observing changes in the output. We present an overview of the most recent research focusing on the differences and similarities in the applications of perturbation-based methods to different data types, from extensively studied perturbations of images to the just emerging research on perturbations of video, natural language, software code, and reinforcement learning entities.

Soft-self and Hard-cross Graph Attention Network for Knowledge Graph Entity Alignment

Knowledge Graph (KG) entity alignment aims to identify entities across different KGs that refer to the same real world object, and it is the key step towards KG integration and KG complement. Recently, Graph Attention Network (GAT) based models become a popular paradigm in entity alignment community owing to its ability in modeling structural data. But current GAT based models either ignore relation semantics and edge directions when learning entity neighbor representations or make no distinction between incoming neighbors and outgoing neighbors when calculating their attention scores. Furthermore, softmax functions utilized in soft attention mechanisms of current models always assign small but nonzero probabilities to trivial elements, which is unsuitable for learning alignment oriented entity embeddings. Taking these issues into account, this paper proposes a novel GAT based entity alignment model SHEA (Soft-self and Hard-cross Graph Attention Networks for Knowledge Graph Entity Alignment), which takes both relation semantics and edge directions into consideration when modeling single KG, and distinguishes prior aligned neighbors from the general ones to take full advantage of prior aligned information. Specifically, a type of four-channels graph attention layer is conceived to aggregate information from entity neighbors in different cases. The first two channels teach entities to aggregate information from their neighbors with soft-self attention, where both neighboring entities and the linked relations are used to obtain attention values. The other two channels teach entities to aggregate information from their neighbors with hard-cross graph attention, where tf_idf11tf_idf (term frequency–inverse document frequency) is a commonly used weighting technology in statistics based information retrieval. It measures the importance of a word with regard to a document. If we think of a pre-aligned neighbor as a word and look all pre-aligned neighbors for a center entity as a document, tf_idf is suitable for measuring the importance of each pre-aligned neighborhood. is utilized to measure the importance of entity neighbors. Extensive experiments on five publicly available datasets demonstrate our superior performances.

Learning entity type structured embeddings with trustworthiness on noisy knowledge graphs

Knowledge graphs (KGs) automatic construction generally involves fine-grained entity typing, i.e., assigning the types to given entities as (entity, entity type). Since the non-negligible inaccuracy of entity typing systems and lack of sufficient human supervision, KGs inevitably face entity type noises. However, most conventional entity type embedding models unreasonably assume that all entity type instances in existing KGs are completely correct, which ignore noises and could lead to potential errors for down-stream tasks. To address this issue, we propose TrustE to build trustworthiness-aware entity type structured embeddings, which takes possible entity type noises into consideration for learning better representations. Specifically, since entities and entity types are completely distinct objects, we encode them in separate entity space and entity type space with a structural projecting matrix, and learn entity type embeddings with tuple trustworthiness. To make the trustworthiness more universal, we only utilize the internal structural knowledge in existing KGs and build two tuple trustworthiness considering the local tuple and global triple information respectively, which correspondingly makes it more challenging due to the limited knowledge. We evaluate our models on three tasks: entity type noise detection, entity type prediction and classification. Experimental results on real-world datasets (FB15kET and YAGO43kET) show that our models outperform all baselines on all tasks, which verify the capability of TrustE in learning better entity type structural embeddings on noisy KGs. The source code and data of this paper can be obtained from https://github.com/Quan-SWUFE/TrustE

Reasoning over temporal knowledge graph with temporal consistency constraints

Knowledge graph reasoning or completion aims at inferring missing facts by reasoning about the information already present in the knowledge graph. In this work, we explore the problem of temporal knowledge graph reasoning that performs inference on the graph over time. Most existing reasoning models ignore the time information when learning entities and relations representations. For example, the fact (Scarlett Johansson, spouse Of, Ryan Reynolds) was true only during 2008 - 2011. To facilitate temporal reasoning, we present TA-TransRILP, which involves temporal information by utilizing RNNs and takes advantage of Integer Linear Programming. Specifically, we utilize a character-level long short-term memory network to encode relations with sequences of temporal tokens, and combine it with common reasoning model. To achieve more accurate reasoning, we further deploy temporal consistency constraints to basic model, which can help in assessing the validity of a fact better. We conduct entity prediction and relation prediction on YAGO11k and Wikidata12k datasets. Experimental results demonstrate that TA-TransRILP can make more accurate predictions by taking time information and temporal consistency constraints into account, and outperforms existing methods with a significant improvement about 6-8% on Hits@10.

SDT: An integrated model for open-world knowledge graph reasoning

Knowledge graphs (KGs) have a wide range of applications, such as recommender systems, relation extraction, and intelligent question answering systems. However, existing KGs are far from complete. Knowledge graph reasoning (KGR) has been studied to complete KGs by inferring missing entities or relations. But most previous methods require that all entities should be seen during training, which is impractical for real-world KGs with new entities emerging daily. In this paper, we address the open-world KGR task: how to perform reasoning when entities are not observed at training time. The description-embodied knowledge representation learning (DKRL) model attempts to study the open-world KGR task. We find that DKRL does not consider hierarchical type information contained in entities when learning entities and relations representations, which results in poor performance. To address this problem, we propose a novel model, SDT, that incorporates the structural information, entity descriptions, and hierarchical type information of entities into a unified framework to learn more representative embeddings for KGs. Specifically, for entity descriptions, we explore continuous bag-of-words and convolutional neural networks models to encode the semantics of entity representations. For hierarchical types, we utilize a recursive hierarchy encoder and a weighted hierarchy encoder to construct the projection matrices of hierarchical types. We evaluate the SDT model on both open-world and closed-world reasoning tasks, including entity prediction and relation prediction. Experimental results on large-scale datasets show that SDT achieves a lower mean rank and higher Hits@10 than the baseline methods.

CTEA: Context and Topic Enhanced Entity Alignment for knowledge graphs

We study the problem of finding entities referring to the same real world object in multilingual knowledge graphs(KGs), i.e., entity alignment for multilingual KGs. Recently, embedding-based entity alignment methods get extended attention in this area. Most of them firstly embed the entities in low dimensional vectors space via relation structure of entities, and then align entities via these learned embeddings combined with some entity similarity function. Even achieved promising performances, these methods are defective in utilizing entity contexts and entity topic information. In this paper, we propose a novel entity alignment framework CTEA (Context and Topic Enhanced Entity Alignment), which integrates entity context information and entity topic information to help alignment. This framework learns entity topic distributions from their attributes with a specially designed topic model BTM4EA, and the learned entity topic distributions are used to filter some weakly correlated entities for each entity to be aligned. Meanwhile, we embed KGs to low dimensional vectors space via translation-based KG embedding model and mine context information from these vectors with an attention attached Convolutional Neural Network(CNN). The entity embeddings, entity contexts and entity topics are combined to get the final alignment results. Extended experiments reveal that our method achieves promising performances in most cases.

ParamE: Regarding Neural Network Parameters as Relation Embeddings for Knowledge Graph Completion

We study the task of learning entity and relation embeddings in knowledge graphs for predicting missing links. Previous translational models on link prediction make use of translational properties but lack enough expressiveness, while the convolution neural network based model (ConvE) takes advantage of the great nonlinearity fitting ability of neural networks but overlooks translational properties. In this paper, we propose a new knowledge graph embedding model called ParamE which can utilize the two advantages together. In ParamE, head entity embeddings, relation embeddings and tail entity embeddings are regarded as the input, parameters and output of a neural network respectively. Since parameters in networks are effective in converting input to output, taking neural network parameters as relation embeddings makes ParamE much more expressive and translational. In addition, the entity and relation embeddings in ParamE are from feature space and parameter space respectively, which is in line with the essence that entities and relations are supposed to be mapped into two different spaces. We evaluate the performances of ParamE on standard FB15k-237 and WN18RR datasets, and experiments show ParamE can significantly outperform existing state-of-the-art models, such as ConvE, SACN, RotatE and D4-STE/Gumbel.

Integration of Examination Strategies in E-Learning Platform for Assessment of Collaborative Activities

In this article, the authors propose a set of examination strategies for distributing tasks of collaborative activities. The first purpose behind this proposal is to assess fairly the learners who are involved in group or team work at the e-learning platform. Indeed, in the literature, few methods are used to assess the learners' individual contributions to the collective or collaborative work. Therefore, the proposal of this article is based mainly around this issue. This will lead to an approach to assess individuals within the learning group (or team), which in turn, will allow to assess the group (or team) as a learning entity.

A Game-Theoretic Data-Driven Approach for Pseudo-Measurement Generation in Distribution System State Estimation

In this paper, we present an efficient computational framework with the purpose of generating weighted pseudo-measurements to improve the quality of Distribution System State Estimation (DSSE) and provide observability with Advanced Metering Infrastructure (AMI) against unobservable customers and missing data. The proposed technique is based on a game-theoretic expansion of Relevance Vector Machines (RVM). This platform is able to estimate the customer power consumption data and quantify its uncertainty while reducing the prohibitive computational burden of model training for large AMI datasets. To achieve this objective, the large training set is decomposed and distributed among multiple parallel learning entities. The resulting estimations from the parallel RVMs are then combined using a game-theoretic model based on the idea of repeated games with vector payoff. It is observed that through this approach and by exploiting the seasonal changes in customers' behavior the accuracy of pseudo-measurements can be considerably improved, while introducing robustness against bad training data samples. The proposed pseudo-measurement generation model is integrated into a DSSE using a closed-loop information system, which takes advantage of a Branch Current State Estimator (BCSE) data to further improve the performance of the designed machine learning framework. This method has been tested on a practical distribution feeder model with smart meter data for verification.

Examining power through social relations in teachers’ discourse about intercultural education

ABSTRACTThis study examined power through social relations between entities in the implementation of intercultural communicative competence (ICC) in an environment where English is spoken as a foreign language. The examination involved analysing the discourse of two foreign English teachers. Specifically, subject positioning of entities, made apparent through pronominal markers and shifts, as well as discourse meaning, were analysed to establish attributes and relations pertinent to the teaching and learning entities and context. The findings showed how teachers’ epistemologies can be made visible through the examination of discourse. Steven, a participant of this study, indicated that while he seemingly practised an inclusive approach, he was restricted by his teaching context, which appeared to be stricter, and his students’ language proficiency. The second teacher, Tamara, was more democratic in her pedagogical approach, seen through her students’ involvement. This signified a more open work environment as both teacher and student could make necessary adjustments to support the learning objectives. From this study, we could see how intercultural education is still perceived differently. This difference may be due to the parameters of the teachers. This study also showed how teachers’ discourse is a feasible site for the examination of power.

Beyond word embeddings: learning entity and concept representations from large scale knowledge bases

Text representations using neural word embeddings have proven effective in many NLP applications. Recent researches adapt the traditional word embedding models to learn vectors of multiword expressions (concepts/entities). However, these methods are limited to textual knowledge bases (e.g., Wikipedia). In this paper, we propose a novel and simple technique for integrating the knowledge about concepts from two large scale knowledge bases of different structure (Wikipedia and Probase) in order to learn concept representations. We adapt the efficient skip-gram model to seamlessly learn from the knowledge in Wikipedia text and Probase concept graph. We evaluate our concept embedding models on two tasks: (1) analogical reasoning, where we achieve a state-of-the-art performance of 91% on semantic analogies, (2) concept categorization, where we achieve a state-of-the-art performance on two benchmark datasets achieving categorization accuracy of 100% on one and 98% on the other. Additionally, we present a case study to evaluate our model on unsupervised argument type identification for neural semantic parsing. We demonstrate the competitive accuracy of our unsupervised method and its ability to better generalize to out of vocabulary entity mentions compared to the tedious and error prone methods which depend on gazetteers and regular expressions.

Creating 21st-Century Laboratories and Classrooms for Improving Population Health: A Call to Action for Academic Medical Centers.

Academic medical centers (AMCs) in the United States built world-class infrastructure to successfully combat disease in the 20th century, which is inadequate for the complexity of sustaining and improving population health. AMCs must now build first-rate 21st-century infrastructure to connect combating disease and promoting health. This infrastructure must acknowledge the bio-psycho-social-environmental factors impacting health and will need to reach far beyond the AMC walls to foster community "laboratories" that support the "science of health," complementary to those supporting the "science of medicine"; cultivate community "classrooms" to stimulate learning and discovery in the places where people live, work, and play; and strengthen bridges between academic centers and these community laboratories and classrooms to facilitate bidirectional teaching, learning, innovation, and discovery.Private and public entities made deep financial investments that contributed to the AMC disease-centered approach to clinical care, education, and research in the 20th century. Many of these same funders now recognize the need to transform U.S. health care into a system that is accountable for population health and the need for a medical workforce equipped with the skills to measure and improve health. Innovative ideas about communities as centers of learning, the importance of social factors as major determinants of health, and the need for multidisciplinary perspectives to solve complex problems are not new; many are 20th-century ideas still waiting to be fully implemented. The window of opportunity is now. The authors articulate how AMCs must take bigger and bolder steps to become leaders in population health.

A neural network framework for relation extraction: Learning entity semantic and relation pattern

Relation extraction is to identify the relationship of two given entities in the text. It is an important step in the task of knowledge extraction. Most conventional methods for the task of relation extraction focus on designing effective handcrafted features or learning a semantic representation of the whole sentence. Sentences with the same relationship always share the similar expressions. Besides, the semantic properties of given entities can also help to distinguish some confusing relations. Based on the above observations, we propose a neural network based framework for relation classification. It can simultaneously learn the relation pattern’s information and the semantic properties of given entities. In this framework, we explore two specific models: the CNN-based model and LSTM-based model. We conduct experiments on two public datasets: the SemEval-2010 Task8 dataset and the ACE05 dataset. The proposed method achieves the state-of-the-art result without using any external information. Additionally, the experimental results also show that our approach can represent the semantic relationship of the given entities effectively.

Network learning: Episodes of interorganizational learning towards a collective performance goal

Little is known about learning processes in horizontal networks. This study focuses on networks as learning entities, i.e. learning by multiple organizations as a group, and the mechanisms involved in developing and addressing a network-level performance goal. By using a narrative approach, we gather in-depth primary data from network members to examine: how do firms engage in network learning? and, how is network learning coordinated towards a performance goal in a horizontal inter-firm network? Our findings comprise two learning episodes: ‘learning how to compete’ and ‘learning how to perform’. These episodes help us to understand network learning processes; the relationship between network learning and organizational learning; and the regulatory role a hub firm provides towards a collective performance goal.

Learning Entity and Relation Embeddings for Knowledge Graph Completion

Knowledge graph completion aims to perform link prediction between entities. In this paper, we consider the approach of knowledge graph embeddings. Recently, models such as TransE and TransH build entity and relation embeddings by regarding a relation as translation from head entity to tail entity. We note that these models simply put both entities and relations within the same semantic space. In fact, an entity may have multiple aspects and various relations may focus on different aspects of entities, which makes a common space insufficient for modeling. In this paper, we propose TransR to build entity and relation embeddings in separate entity space and relation spaces. Afterwards, we learn embeddings by first projecting entities from entity space to corresponding relation space and then building translations between projected entities. In experiments, we evaluate our models on three tasks including link prediction, triple classification and relational fact extraction. Experimental results show significant and consistent improvements compared to state-of-the-art baselines including TransE and TransH.

SPARQL-based personalised E-Learning system designed using ontology (SPELSO): an architecture

Today the key issues in implementing an efficient E-Learning system are sharing and reusability of learning content, integrating heterogeneous resources available online and providing personalised content and services. Through our work we put forward SPELSO (SPARQL based personalised E-Learning system designed using ontology), an E-Learning system based on ontology of E-Learning (OntEL) as the underlying knowledge-modelling structure which is used to capture various pedagogical aspects of E-Learning like course/subject details, learning object type, learner profile information (including learner's cognitive state), access type, difficulty level of learning resources etc. We give an SPARQL based algorithm of generic and personalised search for learning content targeted on the knowledge base created through OntEL. Furthermore, we define the notion of current learning status of a learner which is used to capture his cognitive state at any point of time during learning process. Default and personalised learning path provides suggestive sequencing of learning entities to users.

Beyond the Organizational Bounds in CE Research: Exploring Personal and Relational Factors in a Flat Organizational Structure

IntroductionYou can't always plan for innovation. The best companies create multidisciplinary cultures where a spark of innovation can occur anywhere in the organization. What's more, those companies react nimbly when the unexpected does happen and turn that momentary spark into something of lasting value.- Tom Kelley, IDEOA significant body of research in Corporate Entrepreneurship (CE) indicates that imbuing organizations with innovation, vision and creativity is a pervasive, effective means to grapple with constant environmental change and create a sustainable competitive advantage (Bettis & Hitt, 1995; Dess, Lumpkin, & McGee, 1999; Hitt, Ireland, Camp, & Sexton, 2002; Ireland, 2001). Relevant findings in the extant CE literature emphasize conditions that support the strategic transformation of rigid and static organizations into more dynamic learning entities, including the identification of pro-entrepreneurial architectures (Ireland, 2001), entrepreneurial organizational cultures (Brazeal, 1993; Hornsby, Kuratko, & Montagno, 1999; Kuratko, Montagno, & Hornsby, 1990) and strategic postures geared towards innovation and creativity (Covin & Slevin, 1989, 1991; Hornsby et al., 1999; Kuratko et al., 1990). Collectively, this work suggests that successful sustained regeneration (Dess, Ireland, Zahra, Floyd, Janney, & Lane, 2003) of novel products, processes and technologies is most often a deliberate, strategic choice orchestrated through structural and internal organizational factors such as free time for creative pursuits and monetary rewards for innovations (Kuratko, Montagno, & Hornsby, 1990; Marvel, Griffin, Hebda, & Vojak, 2007; Sathe, 1985; Sykes, 1986). Thus, what precipitates individuals to intentionally act on the belief they can contribute to the innovative capacity of their organizations is a key question of interest.The CE literature has largely considered this question through the lens of cultural ubiquity, assuming that management efforts to support innovation through factors such as monetary rewards, time, and decision-making discretion (e.g., Hornsby, Kuratko, Shepherd, & Bott, 2009; Hornsby, Kuratko, & Zahra, 2002) provide sufficient impetus for all organizational members across multiple structural tiers. Yet why it is some people are likely to engage in innovative behavior whereas others are not remains incomplete at least in part because CE theory has largely been developed in primarily hierarchical, top-down settings. Moreover, approaching the issue of individual innovation by focusing on hierarchy tends to lead to an over-structuralized theoretical view of individuals' intentionality (Hindle, Klyver, & Jennings, 2009) that underplays other potentially important personal and relational influences reflected in their innovative state of mind. We posit that considering CE factors in a flat organizational structure might be useful to the already substantial dearth of CE research through examining personal (creative self-efficacy or CSE) and relational (Leader-Member Exchange or LMX) variables in a context that allows for closer interaction and alignment with the CEO's vision.There is a small, emerging body of empirical evidence to support the conceptual underpinning of our position. Specifically, recent evidence suggests that organizational-level factors are not equally motivating, supportive to all, or effective across corporate infrastructures (Hornsby, Kuratko, Shepherd & Bott, 2009). Middle managers, for instance, arguably vary in their willingness to endorse, refine, and shepherd entrepreneurial opportunities. Because they play important shepherding (i.e., championing, nurturing, and guiding) roles (Kuratko, Ireland, Covin, & Hornsby, 2005), how they relate to subordinates appears to be a potentially fundamental aspect to generating the necessary individual volition to undertake innovative endeavors. Yet research focusing explicitly on the role and quality of the relationship between leaders and followers in corporate entrepreneurial activity remains relatively sparse (Tierney & Farmer, 2004),Finally, self-efficacy has been widely investigated in the entrepreneurship literature primarily as a person-centric antecedent variable that markedly impacts entrepreneurial intentions either directly or as a proxy for perceived feasibility of start-up activities (Krueger, 1993; Kruger & Reilly, 2000). …

Student groups as learning entities: The effect of group diversity and teamwork quality on groups' cognitive complexity

Collaborative learning has important group-level benefits, yet most studies in higher education only focus on individual benefits of collaborative learning experiences. This study extends these insights by testing a model in which teamwork quality mediates the impact of several compositional differences (gender, nationality and teamwork expertise diversity, as well as need for cognition disparity) on groups’ cognitive complexity in a sample of 159 student groups. The results support the mediating role of teamwork quality, and have important practical implications for the design of student groups in higher education.

The learning entities in organisational learning â an empirical study

The idea that an organisation could learn in ways that were independent of the individuals was a key breakthrough. This paper explores the important question ‘how are the learning entities – individual, group or organisation –related’? The answer is important for promoting learning by an organisation. This empirical study found that there is a relation between all the levels of learning – individual, group and organisational – with each other. However, a direct relation between organisational level learning and individual level learning is not sufficiently strong. That means group level learning mediates between individual level learning and organisational level learning. Thus, groups are also important learning entities. But more importantly, there is a need for special emphasis by organisations on roles and rules, routines, practices and standing operating procedures, culture, information systems, the physical structure of the workplace and sources outside the organisation.