Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 20 views

    Sms Classificat Ion Method For Disaster Respon Se Using Enhanc ED Machine Learni NG Algorithms

    Uploaded by

    Satish Kodi

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd

    Sms Classificat Ion Method For Disaster Respon Se Using Enhanc ED Machine Learni NG Algorithms

    Uploaded by

    Satish Kodi
    20 views10 pages

    Original Title

    Untitled

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    20 views10 pages

    Sms Classificat Ion Method For Disaster Respon Se Using Enhanc ED Machine Learni NG Algorithms

    Uploaded by

    Satish Kodi

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    You are on page 1of 10

    M E T H O D FO R

    S IF I CA T IO N
    SMS CLAS U SI N G E N H A NC E D
    P O N SE
    DISASTER RES I N G A L G O R I TH M S
    L E A R N
    MACHINE
    ABSTRACT:

    The bulk of SMS that the Quick Response Team and Rescue Agencies received
    during disasters made it hard for them to categorize responses based on priorities.
    This paper provides a method that classifies SMS received by the agency as Spam,
    Invalid, Alert 1 Alert 2, and Alert 3. This method allows proper response to be
    extended to those asking for it based on prevailing needs. This also provides a
    chance to ignore insignificant messages and save precious time that may be
    incurred by merely dealing with unimportant messages.  
    The existing system implementation of Naïve Bayes Algorithm, a self-learning
    algorithm, and together with Natural Language Processing was utilized in this
    research. Extension of the method is however devised in order to cover the
    irregularity of the data to process. In the proposed system we are using SVM
    and Logistic Regression. Test results of the classification method showed
    success in its implementation and since it is a self-learning process, the method
    gets better and became more accurate through time.
    EXISTING SYSTEM:

    Data Mining Technology provides a platform for information extraction from


    unstructured information to solicit useful data. This study is focused on classifying
    SMS information sent through mobile devices that are coursed through authorized
    agencies to determine the authenticity of the message and its prevailing needs. The
    primary intention is to provide relevant and correct action towards a valid request and
    to set aside spam and insignificant messages as well.
    Disadvantages:

     Time taking process for computation.


     More complexity.
    PROPOSED SYSTEM:

    The Bayesian classifier in this research is used to represent the probability distribution
    of identified text messages of SMS. This method was formulated to solve the
    classification problem associated with this research. Considering the independence of
    the tokens among each other and also with respect to the output, the Naïve Bayesian
    Classifier is the most appropriate method that can efficiently do the job. If the input is
    considered to be a set of attributes, like words, then using a Bayesian network, we can
    calculate the probability of whether a message belongs to a specific class or not. The
    process is composed of two phases: the training phase and the classification phase.
    In the training phase, the filter is trained using a known collection of words for
    classified messages. A database of tokens appearing in each corpus and their total
    occurrences are maintained in a database. Based on their occurrences in each set of
    classified messages, each token is assigned a probability for its capacity to
    determine a message of its classification. we use the SVM and Logistic regression
    to get more exactness and scalability to signify the messages.
    Advantages:

     Robust process.
     Easy computational process.
    SYSTEM REQUIREMENTS

    H/W System Configuration:-


    • Processor - I3/Intel Processor
    • RAM - 4GB (min)
    • Hard Disk - 160GB
    • Key Board - Standard Windows Keyboard
    • Mouse - Two or Three Button Mouse
    • Monitor - SVGA
    S/W System Configuration:-

     Operating System : Windows 7/8/10

     Server side Script : HTML, CSS, JS

     IDE : Pycharm

     Framework : Flask

     Libraries Used : Numpy, Flask, Pandas, Sklearn, Matplotlib, Seaborn.

     Technology : Python 3.6+

    You might also like