Scribd Logo
Upload

Welcome to Scribd!

  • 41 views

    Noida Institute of Engineering and Technology

    Uploaded by

    anand
    The document is a seminar presentation on machine learning given by Lalit Mohan Srivastav. It discusses key topics like defining machine learning, comparing traditional programming to machine learning, applications of machine learning, the machine learning process, algorithms like decision trees and supervised/unsupervised learning, examples of using machine learning like spam filtering, measuring algorithm performance with metrics like confusion matrices and F1 scores, popular software suites, and concludes that machine learning can solve difficult problems and is being widely implemented commercially.

    Copyright:

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

    Noida Institute of Engineering and Technology

    Uploaded by

    anand
    41 views24 pages
    The document is a seminar presentation on machine learning given by Lalit Mohan Srivastav. It discusses key topics like defining machine learning, comparing traditional programming to machine learning, applications of machine learning, the machine learning process, algorithms like decision trees and supervised/unsupervised learning, examples of using machine learning like spam filtering, measuring algorithm performance with metrics like confusion matrices and F1 scores, popular software suites, and concludes that machine learning can solve difficult problems and is being widely implemented commercially.

    Original Title

    ppt

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document is a seminar presentation on machine learning given by Lalit Mohan Srivastav. It discusses key topics like defining machine learning, comparing traditional programming to machine learning, applications of machine learning, the machine learning process, algorithms like decision trees and supervised/unsupervised learning, examples of using machine learning like spam filtering, measuring algorithm performance with metrics like confusion matrices and F1 scores, popular software suites, and concludes that machine learning can solve difficult problems and is being widely implemented commercially.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    41 views24 pages

    Noida Institute of Engineering and Technology

    Uploaded by

    anand
    The document is a seminar presentation on machine learning given by Lalit Mohan Srivastav. It discusses key topics like defining machine learning, comparing traditional programming to machine learning, applications of machine learning, the machine learning process, algorithms like decision trees and supervised/unsupervised learning, examples of using machine learning like spam filtering, measuring algorithm performance with metrics like confusion matrices and F1 scores, popular software suites, and concludes that machine learning can solve difficult problems and is being widely implemented commercially.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    of 24

    NOIDA INSTITUTE OF ENGINEERING AND TECHNOLOGY

    DEPARTMENT OF COMPUTER SCIENCE ENGINEERING


    SEMINAR
    ON
    MACHINE LEARNING

    NAME: LALIT MOHAN SRIVASTAV


    SECTION: CSE-VIIIB
    ROLL NO: 1613310107
    CONTENTS
     Why Machine learning?
     Defining Machine Learning
     Traditional Programming vs Machine Learning
     Application of Machine Learning
     Process of Learning
     Machine Learning Algorithm
     Decision Learning
     Filter Spam example
     Measuring Algorithm performance
     Software
     Conclusion
     References
    Why Machine Learning?

     Develop systems that can automatically adapt and customize themselves to


    individual users.
     Discover new knowledge from large databases (data mining).
     Ability to mimic human and replace certain monotonous tasks.
     Develop systems that are too difficult/expensive to construct manually.
    Defining Machine Learning.

     Machine learning is a method of data analysis that automates analytical model


    building.
     Machine learning (ML) is the study of computer algorithms that improve
    automatically through experience.
      Machine learning algorithms build mathematical model based on sample data,
    known as "training data“.
     Machine learning is closely related to computational statistics.
     Python language suitable for a variety of tasks in machine learning.
    Traditional Programming vs Machine
    Learning
    Traditional Programming

    Data
    Computer Output
    Program

    Machine Learning

    Data
    Computer
    Program
    Program
    Application of Machine Learning
    Process of Learning

     Gathering Data
     Data preparation
     Choosing a model
     Training
     Evaluation
     Parameter Tuning
     Prediction
    Contd…
    Contd…

     Learning = Improving with experience at some task


     Improve over task T,
     With respect to performance measure, P
     Based on experience, E
    Machine Learning Algorithm
     Supervised Learning
     Regressions: learning numbers
     Classifications: learning classes
     Unsupervised Learning
     Clustering: finding groups
     Dimensionality Reductions: finding efficient representations
     Semi-supervised Learning
     Reinforcement Learning
    Decision Learning

     Decision tree learning is one of the predictive modeling approaches used


    in machine learning.
     A decision tree can be used to visually and explicitly represent decisions and
    decision making. 
     It uses a decision tree to go from observations about an item to conclusions about
    the item's target value.
     A decision tree is drawn upside down with its root at the top.
    Contd…
    In the image on the left, the bold
    text in black represents a
    condition/internal node, based
    on which the tree splits into
    branches/ edges. The end of
    the branch that doesn’t
    split anymore is the decision/leaf, in
    this case, whether the passenger
    died or survived, represented as
    red and green text
    respectively.
    Filter Spam Example
    Examples:- Spam Filtering
    Spam:  refers to unsolicited commercial email or unsolicited bulk email 
    Identify Spam Emails
    T: % of spam emails that were filtered
    P: % of non-spam emails that were incorrectly filtered-out
    E: a database of emails that were labelled by users
    The Learning Process for our example
    Measuring Algorithm performance
    There are various metrics for the measurement of algorithm performance
    of Machine Algorithm.
    1. Confusion Matrix:
    1. A confusion matrix is nothing but a table with two dimensions viz. “Actual” and
    “Predicted”.
    2. Both the dimensions have “True Positives (TP)”, “True Negatives (TN)”, “False Positives
    (FP)”, “False Negatives (FN)” as shown below −
    Terms associated with confusion matrix
     True Positives (TP) − It is the case when both actual class & predicted class of data
    point is 1.
     True Negatives (TN) − It is the case when both actual class & predicted class of data
    point is 0.
     False Positives (FP) − It is the case when actual class of data point is 0 & predicted
    class of data point is 1.
     False Negatives (FN) − It is the case when actual class of data point is 1 & predicted
    class of data point is 0.
    2. Classification Accuracy:
       1. It may be defined as the number of correct predictions made as a ratio of
    all predictions made.
    2. We can easily calculate it by confusion matrix with the help of
    following formula −

    3. Classification Report:
    This report consists of the scores of Precisions, Recall, F1 and Support. They are
    explained as follows −
    1. Precision, used in document retrievals, may be defined as the
    number of correct documents returned by our ML model.
    We can easily calculate it by confusion matrix with the help of
    following formula −
       2. Recall or sensitivity may be defined as the number of positives
    returned by our ML model.
    We can easily calculate it by confusion matrix with the help of
    following formula.

    3. Specificity, in contrast to recall, may be defined as the number of


    negatives returned by our ML model.
    We can easily calculate it by confusion matrix with the help of
    following formula −

    4. F1 Score will give us the harmonic mean of precision and recall.


    Mathematically, F1 score is the weighted average of the precision and
    recall.
    F1 = 2∗(precision∗recall)/(precision+recall)
    4. AUC (Area Under ROC Curve) is a performance metric, based on
    varying threshold values, for classification problems.
    ROC is a probability curve and AUC measure the separability. 
    Mathematically, it can be created by plotting TPR (True Positive Rate) i.e.
    Sensitivity or recall vs FPR (False Positive Rate).
    5. LOGLOSS (Logarithmic Loss) is also called Logistic regression loss or
    cross-entropy loss.
    It basically defined on probability estimates and measures the
    performance of a classification model where the input is a probability value between 0
    and 1. 
    It can be understood more clearly by differentiating it with accuracy.
    Software Suites

     Python scikit learn


     MATLAB, mlpy
     Oracle data mining
     SAS enterprise miner
     STATISTICA data miner
     Orange
     Ayasdi
     IBM SPSS Modeler
     Apache Mahout
    Conclusion

      Machine learning is quickly growing field in computer science.


     It has applications in nearly every other field of study.
     It is already being implemented commercially because machine learning can solve
    problems too difficult or time consuming for humans to solve.
     To describe machine learning in general terms, a variety models are used to learn
    patterns in data and make accurate predictions based on the patterns it observes.
    References

     https://www.geeksforgeeks.org/machine-learning/
     https://en.wikipedia.org/wiki/Machine_learning
     https://
    www.tutorialspoint.com/machine_learning_with_python/machine_learning_algori
    thms_performance_metrics.htm
     https://
    www.researchgate.net/publication/2940012_Machine_Learning_Techniques_in_S
    pam_Filtering
     https://www.analyticsvidhya.com/blog/2017/09/machine-learning
    THANK YOU

    You might also like