TODAY, SOCIAL NETWORKS HAVE PROVIDED A SUITABLE PLATFORM FOR SOCIAL RELATIONSHIP. AMONG ONLINE SOCIAL NETWORKS, TWITTER HAS BECOME A POPULAR PLATFORM FOR INFORMATION DIFFUSION AROUND THE WORLD. DUE TO POPULARITY OF TWITTER, IT HAS BEEN TARGETED BY spamMERS AND MALICIOUS ACTIVITIES. IN THIS REGARD, SEVERAL STUDIES HAVE BEEN CONDUCTED USING MACHINE LEARNING TECHNIQUES BY RESEARCHERS TO REACH PROMISING RESULTS. IN RECENT YEARS, ENSEMBLE LEARNING ALGORITHMS HAVE BEEN PRESENTED AS ONE OF THE MODERN MACHINE LEARNING TECHNIQUES, DUE TO ITS HIGH ACCURACY, FOR DATA MINING. IN THIS PAPER, WE PROPOSE A DATA MINING FRAMEWORK USING ENSEMBLE LEARNING FOR spam DETECTION IN TWITTER. IN THE PROPOSED METHOD, AFTER DATA COLLECTION, PREPROCESSING, FEATURE EXTRACTION AND FEATURE SELECTION, THE CLASSIFICATION IS CONDUCTED BY ENSEMBLE LEARNING USING THE DECISION TREE, K-NEAREST NEIGHBOR AND NAï VE BAYES. THE SIMULATION RESULTS ARE COMPARED WITH OTHER CLASSIFICATION ALGORITHMS.