Meta Knowledge Research Articles

A topic modeling framework for spatio-temporal information management

Real-time processing and learning of conflicting data, especially messages coming from different ideas, locations, and time, in a dynamic environment such as Twitter is a challenging task that recently gained lots of attention. This paper introduces a framework for managing, processing, analyzing, detecting, and tracking topics in streaming data. We propose a model selector procedure with a hybrid indicator to tackle the challenge of online topic detection. In this framework, we built an automatic data processing pipeline with two levels of cleaning. Regular and deep cleaning are applied using multiple sources of meta knowledge to enhance data quality. Deep learning and transfer learning techniques are used to classify health-related tweets, with high accuracy and improved F1-Score. In this system, we used visualization to have a better understanding of trending topics. To demonstrate the validity of this framework, we implemented and applied it to health-related twitter data from users originating in the USA over nine months. The results of this implementation show that this framework was able to detect and track the topics at a level comparable to manual annotation. To better explain the emerging and changing topics in various locations over time the result is graphically displayed on top of the United States map.

ENVISIONING CHANGE IN THE STATISTICS-EDUCATION CLIMATE

The African Data Initiative started as a crowd-sourced campaign to improve the teaching of statistics in African universities. The analysis of climate data provides one suitable context to illustrate ideas that lead to a radical new form of teaching. The problem within the context comes first, the technicalities are largely reduced – mathematics is supported by meta knowledge and backed up by modelling; calculations are reduced by user-friendly software that is also used by experts. The problems are treated similarly to research questions and the results are often easier to interpret, making sense as potential answers in their context. The criteria of this approach are compared to the framework proposed by W. G. Cobb to reform statistics education in the light of the latest developments in statistics, driven by the huge increase of data. Implementation details are presented around three components: case studies, data, and the required skills. Together, these three components describe an alternative education pathway centred around statistical problem solving. The focus on interpretations of results within a real context enables software, mathematical thinking and modelling to play a supportive role, which flattens the prerequisites of complex methods and encourages their use across all levels of education.
 First published February 2020 at Statistics Education Research Journal Archives

A Data-Driven Approach to Development of a Taxonomy Framework for Triple Bottom Line Metrics

This paper proposes a data-driven approach to develop a taxonomy in a data structure on list for triple bottom line (TBL) metrics. The approach is built from the authors reflection on the subject and review of the literature about TBL. The envisaged taxonomy framework grid to be developed through this approach will enable existing metrics to be classified, grouped, and standardized, as well as detect the need for further metrics development in uncovered domains and applications. The approach reported aims at developing a taxonomy structure that can be seen as a bi-dimensional table focusing on feature interrogations and characterizing answers, which will be the basis on which the taxonomy can then be developed. The interrogations column is designed as the stack of the TBL metrics features: What type of metric is it (qualitative, quantitative, or hybrid)? What is the level of complexity of the problems where it is used? What standards does it follow? How is the measurement made, and what are the techniques that it uses? In what kinds of problems, subjects, and domains is the metric used? How is the metric validated? What is the method used in its calculation? The column of characterizing answers results from a categorization of the range of types of answers to the feature interrogations. The approach reported in this paper is based on a screening tool that searches and analyzes information both within abstracts and full-text journal papers. The vision for this future taxonomy is that it will enable locating for any specific context, discern what TBL metrics are used in that context or similar contexts, or whether there is a lack of developed metrics. This meta knowledge will enable a conscious decision to be made between creating a new metric or using one of those that already exists. In this latter case, it would also make it possible to choose, among several metrics, the one that is most appropriate to the context at hand. In addition, this future framework will ease new future literature revisions, when these are viewed as updates of this envisaged taxonomy. This would allow creating a dynamic taxonomy for TBL metrics. This paper presents a computational approach to develop such taxonomy, and reports on the initial steps taken in that direction, by creating a taxonomy framework grid with a computational approach.

Building and regulating cognitive, linguistic, and interactional aspects of knowledge between the individual and the group

Research traditions studying knowledge conceive of it differently and place boundaries between knowledge types in different places. A first challenge is elaborating a framework that distinguishes between the co-elaboration of knowledge types (cognitive, linguistic, social, technological…), yet shows their connections. A second challenge is distinguishing co-elaboration of knowledge from self and social regulation of knowledge elaboration. Knowing what knowledge you are regulating may facilitate knowledge regulation. Meeting the first challenge allows us to combine research from different traditions in order to broaden understanding. Meeting the second helps determine how to support the regulation of knowledge elaboration. As a first step towards meeting these challenges, we propose 1) the logical space of knowledge elaboration and its regulation and 2) a multi-theoretical, interdisciplinary model called Multi-grain collaborative knowledge construction that concerns how individuals and groups co-elaborate knowledge during group interactions. We instantiate this model with two illustrations (physics learning and collaborative game learning) and their corresponding logical spaces of elaboration and regulation. Results reveal how knowledge regulation occurs through action and not just verbalization, that regulation interventions are multifunctional, that meta knowledge about learning can help regulation, that knowledge acquired during development is more difficult to regulate. A roadmap is proposed for furthering the distinction between knowledge elaboration and regulation.

Community engagement and subgroup meta-knowledge: some factors in the soul of a community

The Knight Foundation collects data to determine what factors impact community satisfaction, local GDP growth, and interest in Knight news publications. For the 2013 Data Expo at the Joint Statistical Meetings, participants created graphical explorations of these data. This article focuses on the idea of community meta-knowledge, which is essentially majority group empathy or understanding of how minorities experience their community. For example, the survey asks participants to rate their community “as a place for senior citizens,” on a 5-point Likert scale. A city where seniors rated their community in the same way as non-seniors is defined as a community with high meta-knowledge about conditions for seniors. Three minority groups were explored: seniors, families with young children, and racial minorities. In most communities, people outside the minority group tended to under-rate their community, compared to those in the minority group. However, meta knowledge about racial minorities stood out as an exception.

역량기반 교육과정 전환을 위한 조건 탐색

The purpose of this research is to define the problems and possibilities of competency curriculum in the base on Curriculum 2015.BR Competencies in education are key-words for future society. Since the Education systems in Korea operate in bureaucratic structure, it is not easy to actualize competency- based curriculum. The structural problems of 2015 National Curriculum are also based on the same bureaucratic structure, which leads to misunderstand the application of competency into classes. As a result, restructuring of the curriculum for the competency education should be changed in the systemic organization, and it should be innovative from the early stage of organizing stages. Learning should be organized based on the context of knowledge in the learner’s routines in the horizontal structure. In addition, an alternative class of competency classes will be able to make a higher cognitive domains through conceptual knowledge, meta knowledge, and procedural knowledge rather than factual knowledge which we focus to teach.

An instance-based learning recommendation algorithm of imbalance handling methods

Imbalance learning is a typical problem in domain of machine learning and data mining. Aiming to solve this problem, researchers have proposed lots of the state-of-art techniques, such as Over Sampling, Under Sampling, SMOTE, Cost sensitive, and so on. However, the most appropriate methods on different learning problems are diverse. Given an imbalance learning problem, we proposed an Instance-based Learning (IBL) recommendation algorithm to present the most appropriate imbalance handling method for it. First, the meta knowledge database is created by the binary relation 〈data characteristic measures-the rank of all candidate imbalance handling methods〉 of each data set. Afterwards, when a new data set comes, its characteristics will be extracted and compared with the example in the knowledge database, where the instance-based k-nearest neighbors algorithm is applied to identify the rank of all candidate imbalance handling methods for the new dataset. Finally, the most appropriate imbalance handling method will be derived through combining the recommended rank and individual bias. The experimental results on 80 public binary imbalance datasets confirm that the proposed recommendation algorithm can effectively present the most appropriate imbalance handling method for a given imbalance learning problem, with the hit rate of recommendation up to 95%.

Identification of research hypotheses and new knowledge from scientific literature

BackgroundText mining (TM) methods have been used extensively to extract relations and events from the literature. In addition, TM techniques have been used to extract various types or dimensions of interpretative information, known as Meta-Knowledge (MK), from the context of relations and events, e.g. negation, speculation, certainty and knowledge type. However, most existing methods have focussed on the extraction of individual dimensions of MK, without investigating how they can be combined to obtain even richer contextual information. In this paper, we describe a novel, supervised method to extract new MK dimensions that encode Research Hypotheses (an author’s intended knowledge gain) and New Knowledge (an author’s findings). The method incorporates various features, including a combination of simple MK dimensions.MethodsWe identify previously explored dimensions and then use a random forest to combine these with linguistic features into a classification model. To facilitate evaluation of the model, we have enriched two existing corpora annotated with relations and events, i.e., a subset of the GENIA-MK corpus and the EU-ADR corpus, by adding attributes to encode whether each relation or event corresponds to Research Hypothesis or New Knowledge. In the GENIA-MK corpus, these new attributes complement simpler MK dimensions that had previously been annotated.ResultsWe show that our approach is able to assign different types of MK dimensions to relations and events with a high degree of accuracy. Firstly, our method is able to improve upon the previously reported state of the art performance for an existing dimension, i.e., Knowledge Type. Secondly, we also demonstrate high F1-score in predicting the new dimensions of Research Hypothesis (GENIA: 0.914, EU-ADR 0.802) and New Knowledge (GENIA: 0.829, EU-ADR 0.836).ConclusionWe have presented a novel approach for predicting New Knowledge and Research Hypothesis, which combines simple MK dimensions to achieve high F1-scores. The extraction of such information is valuable for a number of practical TM applications.

Tool for Extracting Latent Field Overseers’ Knowledge for Risk Recognition on Eyes and Algorithm for Structuring Meta and Domain Knowledge from Utterances

We have developed a tool for extracting risk field overseers’ recognition knowledge with an eye tracking sensor and a motion sensor, and an algorithm for structuring knowledge from their utterances as propositional network.We applied the tool on risk recognition training for 12 field overseers, including experts and novices, in a construction company and collected data for their eyes and utterances during the training.Typical experts’ process for perceiving risk scenario from each scene (Meta knowledge) and individual pieces of risk knowledge including each risk scenario (Domain knowledge) were elicited on the data with the algorithm. As the results of these, we confirmed that the eyes promote verbalizing field overseers’ knowledge for risk recognition and the propositional network helps structuring their domain and meta knowledge for risk recognition.

The concept of intelligent tutoring for enterprise staff as a component of integrated manufacturing control system development

This paper introduces the development and pilot implementation of adaptive approaches to the universal computer system of individual training, functioning on the basis subject knowledge and meta knowledge bases with the latest advances in information and communication technologies and computing and artificial intelligence as an important part of intelligent control and planning manufacturing system.

Don't Like RDF Reification? Making Statements about Statements Using Singleton Property.

Statements about RDF statements, or meta triples, provide additional information about individual triples, such as the source, the occurring time or place, or the certainty. Integrating such meta triples into semantic knowledge bases would enable the querying and reasoning mechanisms to be aware of provenance, time, location, or certainty of triples. However, an efficient RDF representation for such meta knowledge of triples remains challenging. The existing standard reification approach allows such meta knowledge of RDF triples to be expressed using RDF by two steps. The first step is representing the triple by a Statement instance which has subject, predicate, and object indicated separately in three different triples. The second step is creating assertions about that instance as if it is a statement. While reification is simple and intuitive, this approach does not have formal semantics and is not commonly used in practice as described in the RDF Primer. In this paper, we propose a novel approach called Singleton Property for representing statements about statements and provide a formal semantics for it. We explain how this singleton property approach fits well with the existing syntax and formal semantics of RDF, and the syntax of SPARQL query language. We also demonstrate the use of singleton property in the representation and querying of meta knowledge in two examples of Semantic Web knowledge bases: YAGO2 and BKR. Our experiments on the BKR show that the singleton property approach gives a decent performance in terms of number of triples, query length and query execution time compared to existing approaches. This approach, which is also simple and intuitive, can be easily adopted for representing and querying statements about statements in other knowledge bases.

Ranking of Classifiers based on Dataset Characteristics using Active Meta Learning

Classification is a machine learning technique which is used to categorize the different input patterns into different classes. To select the best classifier for a given dataset is one of the critical issues in Classification. Using cross-validation approach, it is possible to apply candidate algorithms on a given dataset and best classifier is selected by considering various evaluation measures of Classification. But computational cost is significant. Meta Learning automates this process by acquiring knowledge in form of Meta-features and performance information of candidate algorithm on datasets and creates a Meta Knowledge Base. Once Meta Knowledge Base is generated, system uses k-Nearest Neighbor as a Meta Learner that identifies the most similar datasets to new dataset. But generation of Meta Example is a costly process due to a large number of candidate algorithms and datasets with different characteristics involved. So Active Learning is incorporated into Meta Learning System that reduces generation of Meta example and at the same time maintaining performance of candidate algorithms. Once the training phase is completed based on Active Meta Learning approach, ranking is provided based on Success Rate Ratio (SRR) method that considers accuracy as a performance evaluation measure.

Piloting a new approach: making use of technology to present a distance learning computer science course

Teaching projects which make use of new technology are becoming of interest to all academic institutions in the UK due to economic pressure to increase student numbers. CMC (Computer- Mediated Communication) such as computer conferencing appears an attractive solution to higher education's 'numbers' problem, with the added benefit that it is free from time and place constraints. Researchers have discussed CMC from a number of different perspectives, for example Mason and Kaye (1989) describe a CMC system as a system of interactivity between tutors, students, resources and organizational structure. Steeples et al (1993) consider CMC in terms of group cohesion, modes of discourse and intervention strategies to stimulate and structure participation. Goodyear et al (1994) discuss the Just in Time (TT-Based) Open Learning (JTTOL) model in terms of a set of educational beliefs, role definitions, working methods and learning resources, together with a definition of infrastructure requirements for CMC. Shedletsky (1993) suggests that a CMC should be viewed in terms of an 'intrapersonal communication' model, while Mayes et al (1994) identify three types of learning which is mediated by telematics, that is, learning by conceptualization, construction and dialogue. Other researchers, such as Velayo (1994), describe the teacher as 'an active agent', and present a model for computer conferencing which neglects the social aspect of CMC, while Berge (1995) mentions the importance of social activity between students and the importance of the role of the moderator. From these accounts, there appear to be a number of dimensions which can be used to evaluate CMC. Not all researchers emphasize the same dimensions; however, this paper proposes that computer conferencing systems should be designed to encourage students to participate in all three of the following dimensions. These can be summarized as: (a) a knowledge dimension (includes domain and meta knowledge); (b) a social dimension; and (c) a motivational dimension. We will focus on the design and management of a conference M205-STILE, which was employed by the Open University's Computer Science department to supplement the teaching of an undergraduate course. The conference was structured into sub-conferences to take account of the above three dimensions and to facilitate students' computer-supported collaborative learning. This paper reports on the construction of the CMC environment, but more importantly how the three dimensions can be sustained through the role of a person whom we have called an Interactive Media Facilitator (IMF). This implementation is part of the 'Students' and Teachers' Integrated Learning Environment' (STILE) project.DOI:10.1080/0968776960040110

Querying for provenance, trust, uncertainty and other meta knowledge in RDF

The Semantic Web is based on accessing and reusing RDF data from many different-sources, which one may assign different levels of authority and credibility. Existing Semantic Web query languages, like SPARQL, have targeted the retrieval, combination and re-use of facts, but have so far ignored all aspects of meta knowledge, such as origins, authorship, recency or certainty of data. In this paper, we present an original, generic, formalized and implemented approach for managing many dimensions of meta knowledge, like source, authorship, certainty and others. The approach re-uses existing RDF modeling possibilities in order to represent meta knowledge. Then, it extends SPARQL query processing in such a way that given a SPARQL query for data, one may request meta knowledge without modifying the query proper. Thus, our approach achieves highly flexible and automatically coordinated querying for data and meta knowledge, while completely separating the two areas of concern.

Querying for Provenance, Trust, Uncertainty and Other Meta Knowledge in RDF

The Semantic Web is based on accessing and reusing RDF data from many different sources, which one may assign different levels of authority and credibility. Existing Semantic Web query languages, like SPARQL, have targeted the retrieval, combination and reuse of facts, but have so far ignored all aspects of meta knowledge, such as origins, authorship, recency or certainty of data. In this paper, we present an original, generic, formalized and implemented approach for managing many dimensions of meta knowledge, like source, authorship, certainty and others. The approach re-uses existing RDF modeling possibilities in order to represent meta knowledge. Then, it extends SPARQL query processing in such a way that given a SPARQL query for data, one may request meta knowledge without modifying the query proper. Thus, our approach achieves highly flexible and automatically coordinated querying for data and meta knowledge, while completely separating the two areas of concern.

Semantic Enrichment in Ontologies for Matching

Matching between heterogeneous ontologies becomes crucial for interoperability in distributed and intelligent environments. Although many efforts in ontology mapping have already been conducted, most of them rely heavily on the meaning of entity names, rather than the semantics defined in ontologies. In order to deal with semantic heterogeneity, we propose a semantically enriched model of ontologies (called MetaOntoModel) where every domain concept is treated as a sort—an entity type that carries identity criteria for its instances—and classified these concepts based on three philosophical notions: identity, rigidity, and dependency. According to the classification, concept-level properties (called meta-knowledge) are embedded for each concept. Our novel idea is that if two concepts are semantically equivalent, then they have the same meta-knowledge. On the contrary, if two concepts possess different kinds of meta knowledge, then they cannot be matched. We also prove that meta-knowledge can determine not only the scope of semantic correspondences, but also which properties are the most relevant in finding correspondence between two similar concepts.

Meta knowledge of intelligent manufacturing: An overview of state-of-the-art

This paper tries to stimulate empirical research into the overall impacts of intelligent systems in manufacturing aspects. To reach this goal, a schema of intelligent applications is provided for each aspect as frame base structure, meaning the knowledge of intelligent applications in that specific aspect. Then, a semantic network is developed for intelligent manufacturing based on hierarchical structure of manufacturing systems to provide Meta knowledge of intelligent manufacturing applications. The analysis of semantic network indicates an increasing growth in application of soft computing in manufacturing aspects. Specially, fuzzy logic and its derivates have found more applications in recent years.

Computer-Automated Multiparadigm Modeling in Control Systems Technology

The use of model-based technologies has made it imperative for the development of a feedback control system to deal with many different tasks such as: plant modeling in all its variety; model reduction to achieve a complexity or level of abstraction suitable for the design task at hand; synthesis of control laws that vary from discrete event reactive control to continuous model predictive control, their analyses, and testing; design of the implementation; modeling of the computational platform and its operating system; analysis of the implementation effects; software synthesis for different platforms to facilitate rapid prototyping, hardware-in-the-loop simulation, etc. Throughout these tasks, different formalisms are used that are very domain specific (e.g., tailored to electrical circuits, multibody systems, reactive control algorithms, communication protocols) and that often need to be coupled, integrated, and transformed (e.g., a block diagram control law in the continuous domain has to be discretized and then implemented in software). Significant improvements in many aspects (performance, cost, development time) of the development process can therefore be achieved by: 1) relating and integrating these different formalisms; 2) automatic derivation of different levels of modeling abstractions; and 3) rigorous and tailored design of the different formalisms by capturing the domain (or meta) knowledge. The emerging field of computer automated multiparadigm modeling (CAMPaM), presented in this paper in the context of control system design, aims to develop a domain-independent formal framework that leverages and unifies different activities in each of these three dimensions.

An Efficient Expert System for Machine Fault Diagnosis

An efficient expert system for machine fault diagnosis is developed. A new search method is proposed in this system to improve the efficiency of the diagnostic process. First of all, a diagnostic tree (a decision tree) is built by domain experts according to the functions of the devices in the machine. Then, the diagnostic priorities of nodes (devices) in the tree are determined based on a fuzzy group multiple attribute decision making method. A meta knowledge base for fault diagnosis is generated automatically based on the determined priorities to guide the diagnostic process. After that, a domain knowledge base that hypothesises possible faults for each device in the tree is generated by domain experts and/or manuals. At last, the inference process starts based on the meta knowledge base and hypothesises which device is the possible cause of failure. To validate the system performance, an illustrative example (VCR troubleshooting) is presented for demonstration purposes.

Concepts learning with fuzzy clustering and relevance feedback

In recent years feedback approaches have been used in relating low-level image features with concepts to overcome the subjective nature of the human image interpretation. Generally, in these systems when the user starts with a new query, the entire prior experience of the system is lost. In this paper, we address the problem of incorporating prior experience of the retrieval system to improve the performance on future queries. We propose a semi-supervised fuzzy clustering method to learn class distribution (meta knowledge) in the sense of high-level concepts from retrieval experience. Using fuzzy rules, we incorporate the meta knowledge into a probabilistic feature relevance feedback approach to improve the retrieval performance. Results on synthetic and real databases show that our approach provides better retrieval precision compared to the case when no retrieval experience is used.