Loading...
Search for: filtering
0.079 seconds

    Incremental Learning Approach in Spam Detection

    , M.Sc. Thesis Sharif University of Technology Ghanbari, Elham (Author) ; Beygi, Hamid (Supervisor)
    Abstract
    Studies show that a large proportion of sent emails are spam. Spam is one of the major problems of e-mail users that result in wasting time and cost. To overcome this problem different ways are used, one of the best ways is detecting spam based on their contents. Separating legitimate e-mails and spam within their contents can be categorized as text classification. So machine-learning approaches are extremely applied in text classification, that machine-learning algorithms can be used for spam classification. However, in the majority of these algorithms, training phase is in a batch. Whereas using incremental learning algorithms is preferred in many applications, especially spam detections.... 

    Continuous-time Mean-Variance Portfolio Selection with Partial Information

    , M.Sc. Thesis Sharif University of Technology Amini Anarani, Ebrahim (Author) ; Moghadasi, Reza (Supervisor) ; Zamani, Shiva (Co-Advisor)
    Abstract
    In this thesis, we study a continuous time financial market of some risky assets and a risk-free asset for investment in a finite time period. We use mean-variance approach for investment in this market. In the model considered here, the mean returns of individual assets are explicitly affected by underlying Gaussian economic factors. Using past and present information of the asset prices, a partial-information stochastic optimal control problem with random coefficients is formulated. Here, the partial information is due to the fact that the economic factors can not be directly observed. In first step, by filtering and in secound step by solving the stochastic control problem, we show that... 

    Development of Compact Finite Difference Boltzmann Method for Simulating Compressible Rarefied Gas Flow

    , M.Sc. Thesis Sharif University of Technology Alemi Arani, Ali (Author) ; Hejranfar, Kazem (Supervisor) ; Fouladi, Nematollah (Co-Supervisor)
    Abstract
    In this work, a high-order accurate gas kinetic scheme based on the compact finite-difference Boltzmann method (CFDBM) is developed and applied for simulating the compressible rarefied gas flows. Here, the Shakhov model of the Boltzmann equation is considered and the spatial derivative term in the resulting equation is discretized by using the fourth-order compact finite-difference method and the time integration is performed by using the third-order TVD Runge-Kutta method. A filtering procedure with three discontinuity-detecting sensors is applied and examined for the stabilization of the solution method especially for the problems involving the discontinuity regions such as the shock. The... 

    Learning Methods of Minimization of Drive Test (Signal Fingerprinting Method)

    , M.Sc. Thesis Sharif University of Technology Sabati, Monther (Author) ; Hossein Khalaj, Babak (Supervisor)
    Abstract
    Drive Test is a known technique witch network operators use to optimize and evaluate their mobile network infra in terms of capacity, coverage, and quality of service. Conducting Drive tests in outdoor areas is time-consuming and increases CAPEX and OPEX. Also, in areas with many giant physical obstacles, like towers and buildings in cities, Drive test is nearly impossible. In this paper, based on data and measurements provided by various Drive tests and TEMS MDT solution, we use an advanced processing algorithm in our database to bring signal coverage map practically to the users' cellphones. By taking advantage of grid-based signal fingerprinting technique, and filtering geo-tagged...