scholarly journals TopicKA: Generating Commonsense Knowledge-Aware Dialogue Responses Towards the Recommended Topic Fact

2020 ◽  
Author(s):  
Sixing Wu ◽  
Ying Li ◽  
Dawei Zhang ◽  
Yang Zhou ◽  
Zhonghai Wu
Keyword(s):  
Large Scale ◽  
Sampling Methods ◽  
Knowledge Bases ◽  
Generation Process ◽  
Dialogue Systems ◽  
Semantic Meaning ◽  
Commonsense Knowledge ◽  
Sampling Schemes ◽  
Teacher Student ◽  
Generation Network

Insufficient semantic understanding of dialogue always leads to the appearance of generic responses, in generative dialogue systems. Recently, high-quality knowledge bases have been introduced to enhance dialogue understanding, as well as to reduce the prevalence of boring responses. Although such knowledge-aware approaches have shown tremendous potential, they always utilize the knowledge in a black-box fashion. As a result, the generation process is somewhat uncontrollable, and it is also not interpretable. In this paper, we introduce a topic fact-based commonsense knowledge-aware approach, TopicKA. Different from previous works, TopicKA generates responses conditioned not only on the query message but also on a topic fact with an explicit semantic meaning, which also controls the direction of generation. Topic facts are recommended by a recommendation network trained under the Teacher-Student framework. To integrate the recommendation network and the generation network, this paper designs four schemes, which include two non-sampling schemes and two sampling methods. We collected and constructed a large-scale Chinese commonsense knowledge graph. Experimental results on an open Chinese benchmark dataset indicate that our model outperforms baselines in terms of both the objective and the subjective metrics.

scholarly journals UniMF: A Unified Framework to Incorporate Multimodal Knowledge Bases intoEnd-to-End Task-Oriented Dialogue Systems

2021 ◽  
Author(s):  
Shiquan Yang ◽  
Rui Zhang ◽  
Sarah M. Erfani ◽  
Jey Han Lau
Keyword(s):  
Large Scale ◽  
Knowledge Bases ◽  
Language Models ◽  
Dialogue Systems ◽  
Unified Framework ◽  
Proposed Model ◽  
Multimodal Information ◽  
Task Oriented ◽  
Novel Model ◽  
Single Modality

Knowledge bases (KBs) are usually essential for building practical dialogue systems. Recently we have seen rapidly growing interest in integrating knowledge bases into dialogue systems. However, existing approaches mostly deal with knowledge bases of a single modality, typically textual information. As today's knowledge bases become abundant with multimodal information such as images, audios and videos, the limitation of existing approaches greatly hinders the development of dialogue systems. In this paper, we focus on task-oriented dialogue systems and address this limitation by proposing a novel model that integrates external multimodal KB reasoning with pre-trained language models. We further enhance the model via a novel multi-granularity fusion mechanism to capture multi-grained semantics in the dialogue history. To validate the effectiveness of the proposed model, we collect a new large-scale (14K) dialogue dataset MMDialKB, built upon multimodal KB. Both automatic and human evaluation results on MMDialKB demonstrate the superiority of our proposed framework over strong baselines.

scholarly journals Recent Advances in Querying Probabilistic Knowledge Bases

2018 ◽  
Author(s):  
Stefan Borgwardt ◽  
İsmail İlkan Ceylan ◽  
Thomas Lukasiewicz
Keyword(s):  
Large Scale ◽  
Knowledge Bases ◽  
Closed Domain ◽  
Independence Assumption ◽  
Commonsense Knowledge ◽  
Closed World Assumption ◽  
Recent Advances ◽  
Closed World ◽  
Probabilistic Knowledge ◽  
Computational Properties

We give a survey on recent advances at the forefront of research on probabilistic knowledge bases for representing and querying large-scale automatically extracted data. We concentrate especially on increasing the semantic expressivity of formalisms for representing and querying probabilistic knowledge (i) by giving up the closed-world assumption, (ii) by allowing for commonsense knowledge (and in parallel giving up the tuple-independence assumption), and (iii) by giving up the closed-domain assumption, while preserving some computational properties of query answering in such formalisms.

Bullying, School Violence, and Climate in Evolving Contexts

2018 ◽  
Author(s):  
Ron Avi Astor ◽  
Rami Benbenisthty
Keyword(s):  
School Violence ◽  
School Safety ◽  
Media Coverage ◽  
Large Scale ◽  
Ecological Model ◽  
Empirical Studies ◽  
Practical Applications ◽  
Teacher Student ◽  
Area Of Interest ◽  
New Research

Since 2005, the bullying, school violence, and school safety literatures have expanded dramatically in content, disciplines, and empirical studies. However, with this massive expansion of research, there is also a surprising lack of theoretical and empirical direction to guide efforts on how to advance our basic science and practical applications of this growing scientific area of interest. Parallel to this surge in interest, cultural norms, media coverage, and policies to address school safety and bullying have evolved at a remarkably quick pace over the past 13 years. For example, behaviors and populations that just a decade ago were not included in the school violence, bullying, and school safety discourse are now accepted areas of inquiry. These include, for instance, cyberbullying, sexting, social media shaming, teacher–student and student–teacher bullying, sexual harassment and assault, homicide, and suicide. Populations in schools not previously explored, such as lesbian, gay, bisexual, transgender, and queer students and educators and military- and veteran-connected students, become the foci of new research, policies, and programs. As a result, all US states and most industrialized countries now have a complex quilt of new school safety and bullying legislation and policies. Large-scale research and intervention funding programs are often linked to these policies. This book suggests an empirically driven unifying model that brings together these previously distinct literatures. This book presents an ecological model of school violence, bullying, and safety in evolving contexts that integrates all we have learned in the 13 years, and suggests ways to move forward.

scholarly journals Yield Prediction for Tomato Greenhouse Using EFuNN

2013 ◽  
Vol 2013 ◽  
pp. 1-9 ◽  
Author(s):  
Kefaya Qaddoum ◽  
E. L. Hines ◽  
D. D. Iliescu
Keyword(s):  
Large Scale ◽  
Management Practices ◽  
Intelligent System ◽  
Vapour Pressure Deficit ◽  
Knowledge Bases ◽  
Yield Prediction ◽  
Tomato Yield ◽  
Fuzzy Neural ◽  
Rule Sets ◽  
Human Operators

In the area of greenhouse operation, yield prediction still relies heavily on human expertise. This paper proposes an automatic tomato yield predictor to assist the human operators in anticipating more effectively weekly fluctuations and avoid problems of both overdemand and overproduction if the yield cannot be predicted accurately. The parameters used by the predictor consist of environmental variables inside the greenhouse, namely, temperature, CO2, vapour pressure deficit (VPD), and radiation, as well as past yield. Greenhouse environment data and crop records from a large scale commercial operation, Wight Salads Group (WSG) in the Isle of Wight, United Kingdom, collected during the period 2004 to 2008, were used to model tomato yield using an Intelligent System called “Evolving Fuzzy Neural Network” (EFuNN). Our results show that the EFuNN model predicted weekly fluctuations of the yield with an average accuracy of 90%. The contribution suggests that the multiple EFUNNs can be mapped to respective task-oriented rule-sets giving rise to adaptive knowledge bases that could assist growers in the control of tomato supplies and more generally could inform the decision making concerning overall crop management practices.

scholarly journals Hunting for Information in Streamflow Signatures to Improve Modelled Drainage

Water ◽  
2022 ◽  
Vol 14 (1) ◽  
pp. 110
Author(s):  
Raphael Schneider ◽  
Simon Stisen ◽  
Anker Lajer Højberg
Keyword(s):  
Large Scale ◽  
Agricultural Land ◽  
Hydrological Cycle ◽  
Drainage System ◽  
Water Model ◽  
Generation Process ◽  
Hydrological Models ◽  
Drainage Patterns ◽  
Drain Flow ◽  
Set Up

About half of the Danish agricultural land is drained artificially. Those drains, mostly in the form of tile drains, have a significant effect on the hydrological cycle. Consequently, the drainage system must also be represented in hydrological models that are used to simulate, for example, the transport and retention of chemicals. However, representation of drainage in large-scale hydrological models is challenging due to scale issues, lacking data on the distribution of drain infrastructure, and lacking drain flow observations. This calls for more indirect methods to inform such models. Here, we investigate the hypothesis that drain flow leaves a signal in streamflow signatures, as it represents a distinct streamflow generation process. Streamflow signatures are indices characterizing hydrological behaviour based on the hydrograph. Using machine learning regressors, we show that there is a correlation between signatures of simulated streamflow and simulated drain fraction. Based on these insights, signatures relevant to drain flow are incorporated in hydrological model calibration. A distributed coupled groundwater–surface water model of the Norsminde catchment, Denmark (145 km2) is set up. Calibration scenarios are defined with different objective functions; either using conventional stream flow metrics only, or a combination with hydrological signatures. We then evaluate the results from the different scenarios in terms of how well the models reproduce observed drain flow and spatial drainage patterns. Overall, the simulation of drain in the models is satisfactory. However, it remains challenging to find a direct link between signatures and an improvement in representation of drainage. This is likely attributable to model structural issues and lacking flexibility in model parameterization.

scholarly journals Trajectory-User Linking via Variational AutoEncoder

2018 ◽  
Author(s):  
Fan Zhou ◽  
Qiang Gao ◽  
Goce Trajcevski ◽  
Kunpeng Zhang ◽  
Ting Zhong ◽  
...  
Keyword(s):  
Latent Variables ◽  
Large Scale ◽  
Human Mobility ◽  
Hierarchical Structures ◽  
High Dimensional ◽  
Generation Process ◽  
Mobility Patterns ◽  
Variational Autoencoder ◽  
Hidden States ◽  
Structural Semantics

Trajectory-User Linking (TUL) is an essential task in Geo-tagged social media (GTSM) applications, enabling personalized Point of Interest (POI) recommendation and activity identification. Existing works on mining mobility patterns often model trajectories using Markov Chains (MC) or recurrent neural networks (RNN) -- either assuming independence between non-adjacent locations or following a shallow generation process. However, most of them ignore the fact that human trajectories are often sparse, high-dimensional and may contain embedded hierarchical structures. We tackle the TUL problem with a semi-supervised learning framework, called TULVAE (TUL via Variational AutoEncoder), which learns the human mobility in a neural generative architecture with stochastic latent variables that span hidden states in RNN. TULVAE alleviates the data sparsity problem by leveraging large-scale unlabeled data and represents the hierarchical and structural semantics of trajectories with high-dimensional latent variables. Our experiments demonstrate that TULVAE improves efficiency and linking performance in real GTSM datasets, in comparison to existing methods.

scholarly journals The Significance of Irrational Aspect for the Formation of Relations in the “Teacher – Student – Teacher” System

2019 ◽  
Vol 62 (2) ◽  
pp. 58-76
Author(s):  
T. E. Marinosyan
Keyword(s):  
Student Teacher ◽  
Large Scale ◽  
Cognitive Activity ◽  
Educational Process ◽  
Teacher Education Programs ◽  
Pedagogical Issues ◽  
Teacher Student ◽  
Theory Of Education ◽  
Comprehensive Development ◽  
Emotional Component

The article is devoted to the significance of the irrational in education processes and to the role of teacher as an actor of psychological influence on the formation of child’s personality. Unfortunately, teacher education programs at universities do not properly introduce to the students all the aspects (including unconscious ones) of the interaction between people, in particular in the “teacher – student” system. At the same time, in the pedagogical literature there are no special works related to this issue. Psychological theories and methods of education, which are designed to form and develop a personality, can simply destroy the necessary qualities in a child that lead to his independence and self-affirmation. Teachers who are incompetent in pedagogical issues often defend themselves by appealing to the well-known names of scientists and their theories. The theories should not become a dogma and a main path in education. They should serve as an auxiliary instrument. The theory of education is obliged to revise the content and form of the organization of educational process in accordance with the understanding and sense of the rhythm of the New World, taking into account the integration of educational concepts that provide a balance of the rational and irrational in the cognitive activity of a student. This balance is necessary for the full and comprehensive development of child’s personality, for the formation of his/her identity. The author of the article believes that under the conditions of large-scale simulacrization, virtualization, digitalization of the educational space, it is necessary to strengthen the emotional component of the educational process to avoid the final elimination of the aura (W. Benjamin) of the class area from the sphere of human relations in the “teacher – student – teacher” system.

SEARCH FOR NEW METHODOLOGICAL APPROACHES AND METHODS IN RISK MANAGEMENT

2021 ◽  
Vol 54 (2) ◽  
pp. 35-47
Author(s):  
Svetlana N. Dvoryatkina ◽  
◽  
Arseny M. Lopukhin ◽  
Keyword(s):  
Risk Management ◽  
Large Scale ◽  
Quantitative Methods ◽  
Knowledge Bases ◽  
Fractal Model ◽  
Educational Process ◽  
Machine Learning Algorithms ◽  
Risk Modeling ◽  
Negative Consequences ◽  
Fractal Approach

The study actualized the complex and large-scale problem of adapting the theory of risk man-agement for the education system. A comprehensive analysis of domestic and international stud-ies revealed the lack of a theoretical framework, a general methodological vision of the problem of riskiness and risk-taking in the educational sphere. While effective management of education-al activities, ensuring the development of the competitiveness of the individual in the labor mar-ket and its potential for active participation in the life of society is possible on the basis of the modern paradigm of risk management, integrating achievements in pedagogical, economic, mathematical and computer sciences. A new methodology in the study is the fractal approach, which defines the idea of quantitative and qualitative analysis and assessment of the risk of non-formation of professional competencies, complex educational and cognitive constructs of subject activity. The fractal model of assessing the formation of knowledge and competencies, its risk landscape, taking into account the subject and cognitive divergence, will ensure the effective-ness of the structure of knowledge storage in the educational process, minimizing the time for building space and engineering knowledge bases, and the depth of solving the problem of pre-dicting educational risks. New methods of risk modeling based on machine learning algorithms and factor analysis, methods for constructing neural integrators, quantitative methods with and without taking into account the probability distribution will ensure the accuracy and speed of risk assessment and prediction, will allow one to identify new patterns of risk activity and further ways to develop the theory of risk. The presented effective strategies and innovative tools will solve the problem of minimizing unplanned chaos, the cascade of negative consequences of risky situations, including the COVID-19 epidemic.

scholarly journals A Knowledge-Enhanced Pretraining Model for Commonsense Story Generation

2020 ◽  
Vol 8 ◽  
pp. 93-108 ◽  
Author(s):  
Jian Guan ◽  
Fei Huang ◽  
Zhihao Zhao ◽  
Xiaoyan Zhu ◽  
Minlie Huang
Keyword(s):  
Long Range ◽  
Temporal Order ◽  
State Of The Art ◽  
Knowledge Bases ◽  
Fine Tuning ◽  
Causal Relationships ◽  
Commonsense Knowledge ◽  
Language Generation ◽  
Story Generation ◽  
Global Coherence

Story generation, namely, generating a reasonable story from a leading context, is an important but challenging task. In spite of the success in modeling fluency and local coherence, existing neural language generation models (e.g., GPT-2) still suffer from repetition, logic conflicts, and lack of long-range coherence in generated stories. We conjecture that this is because of the difficulty of associating relevant commonsense knowledge, understanding the causal relationships, and planning entities and events with proper temporal order. In this paper, we devise a knowledge-enhanced pretraining model for commonsense story generation. We propose to utilize commonsense knowledge from external knowledge bases to generate reasonable stories. To further capture the causal and temporal dependencies between the sentences in a reasonable story, we use multi-task learning, which combines a discriminative objective to distinguish true and fake stories during fine-tuning. Automatic and manual evaluation shows that our model can generate more reasonable stories than state-of-the-art baselines, particularly in terms of logic and global coherence.

Sign in / Sign up

Export Citation Format

Share Document