Multiple Emotional Profile Representation for Personalized Information Retrieval Systems

Information Retrieval Systems, often referred to by the acronym IRS, are essential tools in the field of data collection and analysis. Among IRS, Personalized Information Retrieval Systems stand out by offering the ability to personalize and adapt search results based on the user’s individual preferences. This ability to provide tailored information contributes to a more relevant and efficient search experience, meeting the specific needs of each user. A new type of personalized IRS has emerged, emotional IRS, which are distinguished by their ability to adapt search results based on the user’s emotions, represented by a structure called an emotional profile. This development is particularly notable thanks to major advances in the field of Natural Language Processing (NLP), which now allow a finer understanding of the emotions present in textual data. In this paper, we present a comparative study of different data structures for the representation of emotional profiles. The objective is to determine the optimal data structure to represent the emotional profile of the user and thus have an efficient and user-friendly emotional IRS.