Recommender system helps people to find information or items that they needed. Collaborative Filtering (CF) is an eminent technique in recommender systems. CF uses relationships between users and recommends items to the active user based on the ratings of his/her neighbors. But, there are several drawbacks in CF like data sparsity problem, where users only rate a small set of items. This drawback makes the computation of similarity between users imprecise and consequently minimizes the accuracy of CF algorithms. Recently, clustering is bound to produce very good accurate results in the recommendation results. In this paper, a constraint based clustering approach depending on the social information of users to obtain the recommendations is proposed. The application of this approach is studied in two application scenarios: academic venue recommendation based on collaboration information and trust-based recommendation. Using the data from DBLP digital library, the evaluation shows that the proposed constrained clustering technique based CF performs better than clustering based CF algorithms.