Abstract:
Semantic textual similarity detection is one of the research field in Natural Language Processing. The proposed solutions for this problem have different application area...Show MoreMetadata
Abstract:
Semantic textual similarity detection is one of the research field in Natural Language Processing. The proposed solutions for this problem have different application areas such as machine translation, semantic relatedness at sentence level, paraphrase detection or question-type classification. In this work, an application was proposed that applies Fisher encoding with using one of the distributed word representation method named as Word2Vec to find semantically similar sentences. This method was tested with a large test set of sentences to find similar sentences. The results show that the proposed method provides an effective solution to sentence similarity problem. In addition to this, it was aimed to be able to make benchmark tests for new methods by sharing Turkish corpus which is rich in text diversity.
Date of Conference: 02-05 May 2018
Date Added to IEEE Xplore: 09 July 2018
ISBN Information: