Disentangled Representations for Manipulation of Sentiment in Text
This addresses a need in fields like marketing and politics for automatic optimization of message impact and personalized language, though it is a first step and likely incremental.
The paper tackles the problem of manipulating sentiment in text while preserving semantics, presenting an algorithm that uses disentangled representations to achieve this, with validation through embedding space trajectories and analysis of transformed sentences.
The ability to change arbitrary aspects of a text while leaving the core message intact could have a strong impact in fields like marketing and politics by enabling e.g. automatic optimization of message impact and personalized language adapted to the receiver's profile. In this paper we take a first step towards such a system by presenting an algorithm that can manipulate the sentiment of a text while preserving its semantics using disentangled representations. Validation is performed by examining trajectories in embedding space and analyzing transformed sentences for semantic preservation while expression of desired sentiment shift.