Business Matching for Event Management and Marketing in Mass Based on Predictive Algorithms
Anas Sabbani, Anass El Haddadi · 2019
This work focuses on finding an optimal solution based on Machine Learning, for computation of a matching score between a Buyer and Seller, for business-to-business event match purposes. The idea is to propose to the Seller the most relevant Buyers who can match the most with him when attending a trade show event. Currently, the match propensity is based on the syntactic analysis of the interests of the Buyer with the industries of all Sellers and propose only those who have at least one similarity with the Seller's industries. This process is manual and primitive. The customized model involving Interactions Converter Algorithm, Collaborative Filtering and Natural Language Processing, is the proposed solution in this work. This approach is based on implicit feedback derived from the user's interaction on front-end application. The user's behavior is tracked through several indicators that are used to provide an explicit rating for the material interacted with. The integration of this solution will allow matching score improvement in business-to-business match forecasting, resulting in better planning for Sellers and Buyers and more effective meetings.