Quantifying emotions through quantum computations

Abstract

Recently, the popularity of using the expressive power of quantum machine newlinelearning to solve known challenging problems has increased remarkably. Emotion newlinedetection and sentiment analysis is one such field that has lots of potential to be newlineexplored utilizing quantum computations. Despite them having good syllogistic newlineprogress in recent years using both lexicon based and machine learning based newlineapproaches, albeit suffer from limitations such as quantification of emotion newlineintensity, interpretation of the context, accurate classification of semantic and newlinesentiment information from the rich content generated by the user. We propose newlineQuantum Machine Learning (QML) based solutions to overcome these research newlinechallenges. Leveraging core elements of quantum such as entanglement and newlinesuperposition, their utilization can help solve complex problems in emotion newlinedetection domain that are yet to be explored. The aim of this research proposal is newlineto build understanding of the limitations and promises of the state-of-the-art newlinequantum algorithms for machine learning and also to define directions for future newlineresearch in this field. In this regard, we present literature survey of some of the newlinededicated work done by the researchers in quantum field. Looking for areas that newlinebear more advantages using QML algorithms, we propose novel algorithms, and newlineconvey that quantum algorithms are one of the most promising tools that can be newlineapplicable in numerous applications such as sentiment analysis, emotion detection newlineand ranging from financial to healthcare industry with potentially exponential newlineadvantage over classical approaches. newline