作者： H Jair Escalante , Manuel Montes y Gómez , Luis Enrique Sucar
DOI:
关键词:
摘要: Management of large and growing information collections is one of the main challenges of modern computer science. Technological innovations have caused the existence of databases of text, images, sounds and astronomical information. In consequence effective methods for organization, retrieval and exploration of such databases is evidently necessary. Among these collections we can distinguish those containing multi-type information (MTI). Also known as annotated collections in which one data type is the annotation of another data type. Examples of these collections include, images and their captions, video and its audio, papers and their corresponding keywords and/or references, genes and their functions, etcetera. Retrieval from MTI collections is a more difficult task than that of retrieval from single-type (ST) databases. Since we should deal with different data representations simultaneously, as well as …