An Efficient Social Spider Optimization Algorithm Based Multi-Document Summarization Model
| Autor: | R. Senthamizh Selvan, K. Arutchelvan |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
0209 industrial biotechnology
Computer science business.industry Process (engineering) Deep learning 02 engineering and technology computer.software_genre Automatic summarization Statistical classification 020901 industrial engineering & automation Multi-document summarization ComputingMethodologies_DOCUMENTANDTEXTPROCESSING 0202 electrical engineering electronic engineering information engineering Preprocessor 020201 artificial intelligence & image processing Data pre-processing Artificial intelligence Data mining business Representation (mathematics) computer |
| Zdroj: | 2021 6th International Conference on Inventive Computation Technologies (ICICT). |
| Popis: | At present times, text summarization is considered an effective technique used for the extraction of useful data from the massive quantity of documents. Depending upon the document count involved in the summarization process, it is classified into single or multi-document summarization. Compared to single document summarization, the multi-document summarization process remains a difficult task of finding a precise summary from many documents. A new Social Spider Optimization (SSO) algorithm based multi-document summarization model were proposed. This model comprises of involves preprocessing, representation of inputs, and representation of summary. The intention of the summary representation process is the generation of the summary of the documents comprising meaning data. By the optimum sentence selection procedure using the SSO algorithm, the essential sentences representing the summary are chosen. A detailed experimental validation process is carried out using the DUC 2006 and 2007 dataset. The experimental results verified the effectual outcome of the SSO algorithm compared to other optimization algorithms. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|