Development of Automated Methods for the Domain Ontology Population with the Help of a Virtual Assistant

Organisations use data in different formats: Word documents, Excel spreadsheets, databases, HTML pages and so on. It is not easy to make decisions with such data due to the lack of integration between the different sources and built-in decision-making rules. Decisions can be reached with knowledge bases, which, unlike databases, make it possible to store not only objects, facts and attributes but also more sophisticated patterns such as rules and axioms. The article proposes an ontology-based method for knowledge base creation that allows for the simultaneous integration of semi-structured data sources and extendibility while remaining context independent. At the initial steps of the method, data specification should be performed with the Data Sources Ontology developed by the authors. This ontology provides data structure description that forms supportive knowledge graph. The graph’s schema should be mapped with the domain ontology to be populated. Finally, the data are inserted into the domain ontology according to the mapping rules. Manual input is needed during data specification and data-to-ontology schema mapping.

Download Full-text

WEB PAGES CLASSIFICATION USING DOMAIN ONTOLOGY AND CLUSTERING

International Journal of Pattern Recognition and Artificial Intelligence ◽

10.1142/s0218001409007065 ◽

2009 ◽

Vol 23 (01) ◽

pp. 17-29

Author(s):

SIMA SOLTANI ◽

AHMAD ABDOLLAHZADEH BARFOROUSH

Keyword(s):

Research Area ◽

Domain Ontology ◽

Training Data ◽

Training Dataset ◽

Classification Algorithms ◽

Web Pages ◽

Main Role ◽

Single Level ◽

Common Domain ◽

Ontology Population

Transferring the current Websites to Semantic Websites, using ontology population, is a research area within which classification takes the main role. The existing classification algorithms and single level execution of them are insufficient on web data. Moreover, because of the variety in the context and structure of even common domain Websites, there is a lack of training data for these classification algorithms. In this paper, we present three experiments: (1) use information in domain ontology on the layers of classes to train classifiers (layered classification) with improvement up to 10% on accuracy of classification, (2) experiment on the problem of training dataset and use clustering as a preprocess, (3) use ensembles to benefit from both methods. Beside the improvement of accuracy in these experiments, we found that for the ensemble we can dispense with the algorithm of classification and use a simple classification like Bayes and achieve the accuracy of complex algorithms like SVM.

Download Full-text

Follow-Up Question Generation Using Neural Tensor Network-Based Domain Ontology Population in an Interview Coaching System

10.21437/interspeech.2019-1300 ◽

2019 ◽

Cited By ~ 2

Author(s):

Ming-Hsiang Su ◽

Chung-Hsien Wu ◽

Yi Chang

Keyword(s):

Domain Ontology ◽

Question Generation ◽

Tensor Network ◽

Ontology Population

Download Full-text

Automation of Two-Stage Factor VIII Assay

Thrombosis and Haemostasis ◽

10.1055/s-0038-1646705 ◽

1978 ◽

Vol 39 (02) ◽

pp. 455-465 ◽

Cited By ~ 5

Author(s):

Yvonne Stirling ◽

D J Howarth ◽

Marguerite Vickers ◽

W R S North ◽

T W Meade

Keyword(s):

Factor Viii ◽

Optical Density ◽

Electric Circuit ◽

Clotting Time ◽

Two Stage ◽

Laboratory Error ◽

Premature Closure ◽

Factor Viii Concentrates ◽

Experimental Findings ◽

Automated Methods

SummaryTwo automated methods for two-stage factor VIII assays have been compared with one another, and evaluated in practice. The Depex method records the clotting time when an electric circuit is completed by the formation of a fibrin thread across a hook-type electrode; the Electra method is based on an optical density technique of clot detection. The two methods gave comparable results for measured levels of factor VIII when haemophilic or “normal” plasmas were assayed. Results from the two methods in practice also suggest that both are valid at low and “normal” factor VIII levels. The Electra method is also probably suitable for assays of concentrates; however, the Depex method appears to give falsely high values in these circumstances, and experimental findings suggest that the reason may be that increased viscosity due to the high fibrinogen levels in factor VIII concentrates causes premature closure of the circuit between the two ends of the Depex electrode. The main advantage of the Depex method is that, provided 3 or 4 machines are available, a given number of assays can be completed more quickly than on Electra. The main advantages of Electra are that it is probably subject to less laboratory error than Depex, and that it is suitable for assaying concentrates as well as haemophilic and “normal” plasmas.

Download Full-text