Using Genetic Algorithms for Personalized Recommendation 107 2.2( Crossover] With a crossover probability cross over the parents to form a new offspring. If no crossover was performed, offspring is an exact copy of parents 2.3[Mutation] With a mutation probability mutate new offspring at each locus(po sition in chromosome) 2.4Accepting] Place new offspring in a new population 3. Evaluation] Compute the fitness values for the new population of N solutions 4. [Test] If the stopping criterion is met, stop, and return the best solution in current opulation 5. Loop] Go to step 2 3 System Architecture In this paper, we propose a hybrid recommender system that uses genetic algorithms for feature weighting to find similar who may share the same interests as the active customers, and capture the potential needs of customers. The proposed system consists of three modules, as shown in Fig. I Product Customer Customer files Transacton Profle generatior Module candidate table Nei ghbomood Top N neigh Recommendation Recommend it Fig. 1. System architecture 3.1 Profile Generation Module (PGm) The goal of PGM is to create the preference profile for each customer. The first step in PGM is to build the product profile from the product database. Each product profile is characterized by its feature values and defined as a binary vector asUsing Genetic Algorithms for Personalized Recommendation 107 2.2[Crossover] With a crossover probability cross over the parents to form a new offspring. If no crossover was performed, offspring is an exact copy of parents. 2.3[Mutation] With a mutation probability mutate new offspring at each locus (po￾sition in chromosome). 2.4[Accepting] Place new offspring in a new population 3. [Evaluation] Compute the fitness values for the new population of N solutions 4. [Test] If the stopping criterion is met, stop, and return the best solution in current population 5. [Loop] Go to step 2. 3 System Architecture In this paper, we propose a hybrid recommender system that uses genetic algorithms for feature weighting to find similar customers who may share the same interests as the active customers, and capture the potential needs of customers. The proposed system consists of three modules, as shown in Fig. 1. Fig. 1. System architecture 3.1 Profile Generation Module ( PGM ) The goal of PGM is to create the preference profile for each customer. The first step in PGM is to build the product profile from the product database. Each product profile is characterized by its feature values and defined as a binary vector as