Web-based Beowulf-Class parallel computing on image database indexing and retrieval system

2002 ◽  
Author(s):  
Chun-Ho Cheung ◽  
Lai-Man Po ◽  
Ka-Man Wong
Keyword(s):  
Parallel Computing ◽  
Retrieval System ◽  
Image Database ◽  
Web Based ◽  
Indexing And Retrieval ◽  
Database Indexing

Unsupervised and Interactive Semi-supervised Clustering for Large Image Database Indexing and Retrieval

2014 ◽  
Vol 130 (2) ◽  
pp. 201-218
Author(s):  
Hien Phuong Lai ◽  
Muriel Visani ◽  
Alain Boucher ◽  
Jean-Marc Ogier
Keyword(s):  
Image Database ◽  
Supervised Clustering ◽  
Indexing And Retrieval ◽  
Database Indexing

LocalContent: a personal scientific document retrieval system

2015 ◽  
Vol 33 (3) ◽  
pp. 373-385 ◽  
Author(s):  
Chih-Fong Tsai ◽  
Shih-Wen Ke ◽  
Kenneth McGarry ◽  
Ming-Yi Lin
Keyword(s):  
Language Processing ◽  
Retrieval System ◽  
Personal Information ◽  
Document Retrieval ◽  
Processing Technique ◽  
Content Type ◽  
Indexing And Retrieval ◽  
Database Indexing ◽  
Personal Archives ◽  
Indexing Term

Purpose – The purpose of this paper is to introduce a novel personal scientific document retrieval system. The most common approach taken for the storage of personal documents is to construct a hierarchical folder structure. Most users prefer searching for documents by manually traversing their organizational hierarchy until reaching the location where the target item is stored, then locating the specific documents within its directory or folder. However, this is very time-consuming, especially when the number of personal scientific documents is very large. Unfortunately, related personal information management (PIM) systems, which provide solutions for managing various types of personal information, have thus far made little progress at managing personal scientific documents. Design/methodology/approach – In this paper, we introduce the design of a personal scientific document retrieval system, namely, LocalContent. It is composed of database indexing and retrieval stages. During indexing, term feature extraction from scientific documents is performed by the natural language processing technique. The extracted terms are stored in the inverted index for later retrieval. For retrieval, a graphical user interface is provided by LocalContent, which allows users to search their personal scientific documents. Findings – The evaluation results based on 20 different personal archives taken from 20 graduate students show that LocalContent is simple to use and can facilitate the search for relevant scientific documents. Moreover, these users were willing to have a system which provides specialized search functions like LocalContent to explore their personal scientific documents in the future. Originality/value – LocalContent is a novel scientific document retrieval system and provides several particular functions of LocalContent including displaying the content summary of the query term frequency in each specific section of the retrieved documents, querying by local section specification and providing a number of recommended keywords related to the query terms.

Sign in / Sign up

Export Citation Format

Share Document