Browsing by Author "Nezzar Salaheddine Zaabeb Chamsseddine"

Now showing 1 - 1 of 1
  • No Thumbnail Available
    Item
    Design and Implementation of a Skyline-Based System for Recommending Offers to Customers of a Telecommunications Operator
    (2024) Nezzar Salaheddine Zaabeb Chamsseddine
    In the field of recommendation systems (RS), the primary objective is to provide users with relevant suggestions based on their preferences and past behaviours. This study explores the use of the skyline technique to enhance the efficiency and quality of recommendations. The skyline method aids in selecting recommendations by considering multiple user preferences or characteristics, such as price, category, etc. Its versatility allows it to be adapted to various decision-making scenarios involving multiple criteria, reducing complexity by quickly identifying non-dominated choices. Additionally, we examine the integration of TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) and cosine similarity into the recommendation process. TOPSIS ranks alternatives based on their proximity to ideal and anti-ideal solutions, while cosine similarity measures the degree of similarity between two vectors. By combining these methods, we develop a robust and flexible method to multi-criteria decision making, capable of identifying alternatives that are close to the ideal, distinct from the anti-ideal, and reflective of similarity relationships between criteria. The aim of this work is to design and implement an offer recommendation system for a telecommunications operator using the skyline, TOPSIS, and cosine similarity methods. Such a system will help customers select packages, services, and promotions that best meet their needs and usage patterns, while enabling the operator to maximize sales and customer loyalty. The choice of the skyline method is justified by its ability to streamline decisionmaking by focusing on a relevant and non-dominated subset of data, thereby eliminating clearly inferior choices across all considered criteria.