| Autor: |
Kishorjit Nongmeikapam, B. Haripriya |
| Rok vydání: |
2020 |
| Předmět: |
|
| Zdroj: |
Lecture Notes in Electrical Engineering ISBN: 9789811570308 |
| DOI: |
10.1007/978-981-15-7031-5_73 |
| Popis: |
There are existing problems of finding relation between sentences for quite a few years. A system that can find the relation between a phrase and a document can be used for search engines, article finder and many more. This paper is focused on the problem statement of deciding whether given two sentences or two set of sentences are unrelated or related and if related whether it is mutually agreeing, disagreeing or neutral relation will be the classification outcome. Specifically, an attempt to find the title body consistency of a news article is done in this paper. For which, a deep transfer learning-based approach is proposed where the problem of detecting title body consistency is taken from the viewpoint of textual entailment (TE) where the title is considered as a hypothesis and news body is treated as a text. The framework used is bi-directional long short-term memory (LSTM) network which is a type of Recurrent Neural networks in deep learning with experiments with transfer learning technique. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
