Ranking objectives of advertisements on Facebook by a fuzzy TOPSIS method

Abstract

Social networking sites (SNSs) have become a vital medium for companies to place advertisements and setting an objective of advertisements on SNSs is an important issue of planning a business’s market strategy. The purpose of this work is to develop a fuzzy technique for order preference by similarity to an ideal solution (TOPSIS) method for evaluating and selecting objectives of advertisements on Facebook. In the proposed model, the fuzzy weighted ratings are defuzzified by a centroid method to generate distances of each alternative to the positive and negative ideal solutions. A fuzzy weighted normalized distances index is proposed to rank alternatives, and the centroid method is used for defuzzification. Formulas for the defuzzification of fuzzy weighted ratings and the fuzzy weighted normalized distances index are developed. A numerical example of evaluating objectives of advertisements on Facebook is used to demonstrate the feasibility of the proposed method. Example result reveals that the proposed fuzzy weighted normalized distances index is as effective as the crisp closeness coefficient in ranking objectives under the proposed fuzzy TOPSIS method. An experiment demonstrates that the rankings of objectives may be more likely to change as the gap between two linguistic weights that are assigned to fuzzy weighted normalized distances index increases.