Arama ayrıntıları kayıtlarında kullanıcı davranışlarının LSTM modelleri ile tahmin edilmesi

City planners, governors and mobile phone operators utilize quite from Call Detail Records (CDR) data in the fields like optimizations of traffic congestion, event detection, billing and advertisement policies. Both individual and crowd analyses help improving the quality of services. In this thesis, we present regression and classification analysis for three main problems. In regression analysis, we experiment on call counts and call time-sums of specified numbers of users. We propose cluster based and outlier separating models in these two tasks for the purpose of improving the results of individual user-based models comprised of various Long Short Term Memory(LSTM) layers. In the classification analysis, on the other hand, we present models that predict next locations on the trajectories of the users. We improve the results of base LSTM model with two-predictions-at-once approach. The analyses show that recurrent neural networks work well with sequential data and optimizations on top of the models yield promising results.