Informatics Publishing Limited

R. Saranya ¹, S. Gowri ¹, S. Monisha ¹, S. Vigneshwari ¹

Affiliations
1 Department of Information Technology, Sathyabama University, Chennai, IN

Abstract

Objective: The tremendous growth in volume of data is increasing the complexity of data handling. The existing content based search is not only complex and expensive, but also leads to poor processing and unacceptable latency for massive amount of data. Data accessed from large dataset using existing approach will slow down the query response rate. High resource cost is the severe performance bottleneck caused by query operations. The semantic based search can be carried out on the data, In order to reduce the complexity in data handling. The proposed method uses Resource Description Framework (RDF) with Ontology Web Language (OWL). Thus the RDF is used for non domain based search and OWL is used for domain based search. The non domain based search provides the composition of best data services from master RDF. The triplets from different RDF are combined to form master RDF. And the domain based search is one which provides recommendations for the user based on their search query. The main objective of this approach is to provide semantic search using RDF and recommendation using OWL. The search result will be the high similarity result sets as the method uses the combinational RDF to form large master RDF. Statistical Analysis/Methods: The user is provided with two methods of searching which is domain based search and non domain based search. The domain based search is one which provides recommendation for the user based on the search query. And the non domain based search originates the data services from master RDF. Agriculture data set is chosen as a sample data set for domain based search which can be extended further for any other domain. The web services which contain general information about the world and also related web services is chosen as a data set for non domain based search. Findings: The statistical analysis of ontology based semantic search with keyword based search produced the precision value of 0.8 out of 1.0 using the search results obtained from both semantic search and keyword based search. Applications/Improvements: The application can be further improved by adding realistic data sets and increasing the size of the database. The domain based approach is not restricted only to agriculture but it can also be extended to some other applications like Health care, pattern recognition and so on.

Keywords

Correlation, Data Services, Domain, Hazy Semantics, Ontology, Triplets.

Full Text

   

S. Monisha ¹, R. Saranya ¹, K. Vijay ¹

Affiliations
1 Department of Information Technology, Sathyabama University, Chennai – 600119, Tamil Nadu, IN

Abstract

Objective: The people all over the world are in need of gaining knowledge which is achieved by means of searching information through web services. Even though search results obtained from web services are relevant, the user is trying to obtain the best information providers. Methods: The existing approaches which is non-parametric generative model and infinite Hidden Markov Model (iHMM) of clustering technique does not yield better results for very large collections of web services. The iHMM model does not provide any security for datasets in a hierarchal structure. Thus to overcome such downsides of existing approach to a new technique can be used to help of Natural Language Processing (NLP). Findings: The new approach provides both tag recommendation and parental control. The tag recommendation is used to ping the best tags based on expert’s bookmarks, and parental control provided by group owner is used to resolve the privacy issues. The web filter denies the unauthorized access. Thus, using NLP technique in the conspiring environment enables the web user to achieve fine grained knowledge. Applications: The paper defines the efficient way to provide tag recommendation on the web for large dataset users.

Keywords

Bookmarks, NLP, Tag Recommendations, Tag Suppression, Text Mining, Web Services.

Full Text