Design of a blockchain-based secure and efficient ontology generation model for multiple data genres using augmented stratification in the healthcare industry

Abstract

Ontology generation is a process of relationship analysis, and representation for multiple data categories using automatic or semi-automatic approaches. This process requires a domain knowledgebase that describes given input data using entity-to-entity relations. A wide variety of approaches are proposed for this purpose, and each of them processes & converts input data using multiple relationship evaluation stages. These stages include data-preprocessing, correlation analysis, entity mapping, and ontology generation. A very few of these approaches are dataset independent, and most of them do not implement security measures during ontology generation, which limits their security, scalability & deployment capabilities during real-time implementation. Thus, in this text a blockchain based secure & e cient ontology generation model for multiple data genres using augmented strati cation (BOGMAS) is described. The BOGMAS model uses a semi-supervised approach for ontology generation from almost any structured or unstructured dataset. It uses a variance-based method (VBM) for reduction of redundant numerical features from the dataset, while textual features are converted to numerical values via standard word2vec model, and then processed using VBM. This model uses a combination of linear support vector machine (LSVM), and extra trees (ET) strati ers for variance estimation, which makes the model highly e cient, and reduces redundant features from the output ontology. These feature sets & their variances are given to a correlation engine for relationship estimation, and ontology generation.Each ontology record is secured using a mutable proof-of-work (PoW) based blockchain model, which assists in imbibing transparency, traceability, and distributed peer-to-peer processing capabilities. The generated ontology is represented using an incremental OWL (W3C Web Ontology Language) format, which assists in dynamically sizing the ontology depending upon incoming data. Performance of the proposed BOGMAS model is evaluated in terms of precision & recall of representation, memory usage, computational complexity, and accuracy of attack detection. It is observed that the proposed model is highly e cient in terms of precision, recall & accuracy performance, but has incrementally higher computational complexity & delay of ontology formation when compared with existing approaches. Due to this incremental increase in delay, the proposed model is observed to be applicable for a wide variety of real-time scenarios, which include but are not limited to, medical ontology generation, sports ontology generation, and internet of things (IoT) ontology generation with high security levels.