6533b862fe1ef96bd12c642a
RESEARCH PRODUCT
A Survey on how to Cross-Reference Web Information Sources
Aurélie BertauxChristophe CruzJoe Raadsubject
Measure (data warehouse)Information retrievalEvent (computing)Computer scienceSimilarity Definition02 engineering and technologyDocumentary ResearchParaphraseCross-referenceIdentification (information)[ INFO.INFO-IT ] Computer Science [cs]/Information Theory [cs.IT]Semantic similaritySemantic Relatedness[INFO.INFO-IT]Computer Science [cs]/Information Theory [cs.IT]020204 information systemsSemantic Measures0202 electrical engineering electronic engineering information engineering020201 artificial intelligence & image processingSemantic integrationMeaning (existential)Cross-Reference Web Information SourcesEvent ExtractionParaphrase Identificationdescription
International audience; The goal of giving information a well-defined meaning is currently shared by different research communities. Once information has a well-defined meaning, it can be searched and retrieved more effectively. Therefore, this paper is a survey about the methods that compare different textual information sources in order to determine whether they address a similar information or not. The improvement of the studied methods will eventually lead to increase the efficiency of documentary research. In order to achieve this goal, the first category of methods focuses on semantic measure definitions. A second category of methods focuses on paraphrase identification techniques, and the last category deals with improving event extraction techniques. A general discussion is given at the end of the paper presenting the advantages and disadvantages of these methods.
| year | journal | country | edition | language |
|---|---|---|---|---|
| 2015-07-28 |