(IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 10, No. 12, 2019 166 | Page www.ijacsa.thesai.org Enrichment Ontology with Updated user Data for Accurate Semantic Annotation Haytham Al-Feel 1 , Hanaa Ghareib Hendi 2 , Heba Elbeh 3 Faculty of Computers & Information Fayoum University, Fayoum, Egypt Abstract—Annotation is considered one of the main applications that semantic web applies. The idea beyond annotation focused on adding metadata to existing information which facilitates machines dealing with data that have meanings and can be readable. Semantic annotation is one of the techniques used for the enrichment of web content semantically, which facilitates writing comments and evaluate previously annotated resources that can lead to better search results. Our framework aims to enrich ontology via embedding data directly to ontology in order to have completed and accurate data. Keywords—Ontology; Semantic Web; Semantic Annotation; RSS News I. INTRODUCTION The current web has many problems and limitations where most of the web pages can’t be understood by machines, which makes the web flat, boring and also misunderstandings [1]. Semantic Web has many definitions but most of them refer to it as an extension for the current web that links information to each other. It also could link data to anything that could have a Unified Resource Identifier and in this way, the web will be enabled for accurate retrieval [2]. Absolutely, the work is not easy like this but it will include many sophisticated technologies and languages such as the Resource Description Framework (RDF), Web Ontology Language (OWL) and SPARQL as a query language. Semantic annotation can be defined as adding metadata to available documents that add value and explanation to the web, facilitates the searching process and finding any piece of information especially when these data resources linked to each other [3] [4] [5]. Annotation is a set of explanatory notes that accompanies a text [6]. Ontology considered one of the main players of the semantic web and the backbone for the annotation process that explained as the formal specification for a specific domain explaining its concepts semantically [7]. Ontology triples have main three parts they are: subject which describes a specific resource and identified by a URI, predicate which identifies a property and an object which can be another resource or value as shown in “Fig. 1”. Because of syntax and structure which has a lot of details, specifications, and recommendations, regular users do not have the full awareness of ontology to deal with or insert, update and delete data. The task of annotation could enrich content and ontologies in different domains. Our paper presents a Semantic annotation tool that facilitates the user adding data to available resources according to specific rules and privileges determined by the admin of each website. Admin reviews the user's insertion, whether true or false. If the user insertion is acceptable, in this case, data inserted will be stored in the ontology triple store to enrich ontology. Users do not need to have any background about semantic and ontology to use our tool and framework, just he/she will insert data that will be transformed in the form of subject, predicate, and object. The remainder of this paper is organized as follows: Section 2, highlights semantic annotation techniques. While Section 3, presents our framework and evaluation. Finally, Section 4 presents our conclusions and further research. Fig. 1. Ontology Triple Structure. II. SEMANTIC ANNOTATION Remarks, comments, notes and any explanations added to a part of web documents are considered annotations to these web documents and if these annotations based on ontologies and semantic web techniques; in this situation, we call them semantically annotated [8]. Semantic annotation is defined as the process of adding data related to ontological concepts to have full metadata description about concepts. The purpose of a semantic annotation is to specify the meaning and properties of an annotated information resource in a generally understandable way [9]. It recognizes the parts of the text and converts into pieces in a data processing mechanism that can be linked to a broader context of data that already exist [10]. In [11] we developed a Semantic Annotation Framework for News Feeds (SANF) which annotates RSS news titles and provides additional information for concepts on the news domain via an ontology contains different categories and concepts in this domain as shown in “Fig. 2”. In this paper, we added new features for enhancing the system via adding full authority to the user to add a comment on what retrieved from an ontology which could be later accepted or refused from the administrator. In addition to this, even if a particular user needs a brief annotation of an unannotated object the system does not stand idly; it acts intelligently by looking for an object online in order to perform an annotation process as fully as possible. Object Subject Predicate