Recommending model management and visualize statistical results online text - Applying the analysis of trends in scientific research at Cantho University

The objective of the article is to propose a suitable management model which could be used to exploit rich and diversified data in different formats (i.e. text and spreadsheet). Besides, we also propose specific solutions based on a common Big Data platform, including: (1) HDFS (Hadoop Distributed File System) of Hadoop, which could be used in file management, (2) Lucene, which could be used to establish reversed indexing for text and (3) Apache Solr, which could be used to support reversed indexing management mechanism, full text searching and advanced searching functions. This article also presents experimental results, aggregates statistical results and displays statistical chart of applying the model into the analysis of trends in scientific research at Can Tho University.