Collaborative filtering (CF) [6] approach is further classified as Memory-based CF and Model-based CF. Memory-based CF systems utilize the entire user-item rating dataset to generate the predictions. These systems employ statistical techniques to find a set of users known as neighbors whose past behavior was similar to the target user. After identifying the neighborhood of users, these systems use different algorithms to combine the preferences of neighbors to compute suggestions for the active user [15]. In contrast, Model-based CF approach partitions the whole dataset into training and test dataset. The different users in the training database are then clustered into groups based on their rating patterns to build the model. Various machine learning algorithms like Bayesian network [12], clustering algorithms [27] etc are used in literature for building the model. Consequently the trained model is used to generate recommendations for the target user. Though conventional clustering algorithms like K-means are simple and take less time in clustering large datasets, they have good probability of getting trapped in the local optima. To overcome this problem evolutionary algorithm and swarm intelligence techniques are explored for developing recommender systems like Particle Swarm Optimization (PSO) [24] algorithm was employed to learn personal preferences of users and provide tailored suggestions. This system performed better than genetic algorithm (GA) and the Pearson algorithm based recommender system. Another recommender system based on the collaborative behavior of ants [4] outperformed traditional collaborative filtering based recommender systems. Genetic algorithm-based approach [16] has been utilized to determine the weight value of each feature of a customer. This approach revealed better performance on recommendation effect. In another application Genetic algorithm [14] has been used to group users based on products categorized by Naïve Bayes classifier. Subsequently the web documents were recommended to the user, based on grouped user preferences and information of categorized items. The application of GA K-means [13] to a real-world online shopping market segmentation case exhibited that GA K-means clustering performed better as compared to K-means clustering and self-organizing maps (SOM). However in the recent years hybrid evolutionary algorithms like Genetic algorithms integrating local search are gaining popularity over the simple evolutionary algorithms (EAs). These are known as Memetic Algorithms, Baldwinian EAs, Lamarckian EAs, cultural algorithms or genetic local search in the literature.