Distributed Query Processing in Cloud using Canonicalization Approach

摘要

Nowadays, citations and its formats act as a vital role in scientific publication digital libraries (DLs). Due to various reasons, the citation metadata extraction process make it difficult. The citation used in the extraction process is gathered from web and there is no standard style to the citations. Data gathered from the web is difficult to process due to the erroneous nature of the data. In order to overcome these problems, this paper proposed a distributed query processing based on canonicalization approach. Here, NoSQL database is used for distributed query support and to enhance sustained data throughput. The query is retrieved from the user and then verified with the metadata created. In case of match occurs, the index prediction is performed and the results are evaluated. Semantic query matching is used to match the given query with the database. The Citation Parser (CP) method is used to convert the query string into simple symbols to minimize the query processing time and to enhance the system performance. The experimental results shows that the proposed method results higher accuracy with lesser query processing time, response time and lesser computation cost.