SWAMI KESHVANAND INSTITUTE OF TECHNOLOGY, MANAGEMENT & GRAMOTHAN, JAIPUR Industrial Training Machine Learning & Deep Learning Department of Electronics & Communication Engineering Presented to: Mr. Abhinandan Jain (Assistant Professor SKIT Jaipur) Mr. Neeraj Jain (Assistant Professor SKIT Jaipur) SKIT/ECE/IndustrialTraining-AyushJajodia Presented By: Ayush Kumar Jajodia 7ECB-G2 18ESKEC711 1 TABLE OF CONTENTS • • • • • • • Machine Learning Python in ML Applications Algorithms for ML Models Deep Learning Neural Networks Summary 2 SKIT/ECE/IndustrialTraining-AyushJajodia MACHINE LEARNING Machine Learning is a branch of Artificial Intelligence & Computer science which focuses on the use of data and algorithms to imitate the way that humans learn. It is programming computers to optimize performance criterion using example data set. ML can detect patterns in data to predict future data or other outcomes of interest. 3 SKIT/ECE/IndustrialTraining-AyushJajodia WHY ML? Huge amount of data Need Statistics 4 SKIT/ECE/IndustrialTraining-AyushJajodia APPLICATIONS Speech Recognition used in mobile phones is an example of ML Object Recognition Predictive Models Credit Risk Modelling Product Recommendation 5 SKIT/ECE/IndustrialTraining-AyushJajodia PYTHON IN ML Open source General purpose Language Easy to understand. Widely developed Library ecosystem. Used to perform Complex tasks without extensive coding. 6 SKIT/ECE/IndustrialTraining-AyushJajodia STEPS IN BUILDING ML MODEL Problem Formulation Data Tidying Pre-Processing Train-Test Split Model Building Validation & Model Accuracy Prediction SKIT/ECE/IndustrialTraining-AyushJajodia 7 TYPES OF ML ALGORITHMS 1.Supervised ML Algorithms: Algorithms that make predictions on given set of samples. It searches for patterns within the value labels assigned to data points. Ex. Linear Regression, Logistic Regression ,Decision Tree, Support Vector Machine 8 SKIT/ECE/IndustrialTraining-AyushJajodia 2.Unsupervised ML Algorithms: There are no labels associated with data points. These ML Algorithms organize the data into a group of clusters to describe its structure and make complex data look simple and organized for analysis. It is preferable as it is easy to get unlabelled data in comparison to labelled data. Ex. KNN,K-Means clustering ,Neural Networks etc. 9 SKIT/ECE/IndustrialTraining-AyushJajodia ALGORITHMS Linear Regression: It is a linear approach to modelling the relationship between a dependent variable and one or more independent variables. It is a statistical method that is used for predictive analysis. Linear Regression makes predictions for continuous/real or numeric variables such as sales, salary, age, etc. 10 SKIT/ECE/IndustrialTraining-AyushJajodia 2.Logistic Regression: It is used for predicting the categorical dependent variable using a given set of independent variables. It gives the probabilistic values which lie and 1. Linear regression is used for solving regression whereas logistic regression is used for classification problems. between 0 problem solving the It uses sigmoid function (Logistic Function) which is a mathematical function used to map the predicted values to probabilities. SKIT/ECE/IndustrialTraining-AyushJajodia 11 DIFFERENCE B/W LINEAR & LOGISTIC REGRESSION Equation in Linear Regression=> Y= Bo+B1X whereas for logistic Regression eqn would be=> p(X)=e^(Bo + B1X)/(1+ e^(Bo+B1X)). 12 SKIT/ECE/IndustrialTraining-AyushJajodia Decision Trees: A decision tree is a decision support tool that uses a tree like model of decisions and their possible consequences ,including chance events like outcomes, resource costs ,and utility. It is one way to display an algorithm that only contains conditional control statements. Lacks a bit in accuracy. 13 SKIT/ECE/IndustrialTraining-AyushJajodia Support Vector Classifier : It is a supervised ML algorithm that is capable of performing classification,regression and event outlier detection .The Linear SVM classifier works by drawing a straight line between 2 classes. The algorithm creates a line or a hyperplane which separates the data into classes. It uses kernels to create non-linear boundaries. 14 SKIT/ECE/IndustrialTraining-AyushJajodia Deep Learning Deep learning is a type of machine learning and AI that imitates the way human gain certain type of knowledge. It is way to automate predictive analysis. While traditional ML algorithms are linear, deep learning algorithms are stacked in a hierarchy of increasing complexity. 15 SKIT/ECE/IndustrialTraining-AyushJajodia NEURAL NETWORKS Neural Networks make up the backbone of deep learning algorithms. It reflects the behaviour of the human brain, allowing computer to recognize patterns and solve common problems in the field of Artificial Intelligence , Machine Learning & Deep Learning. Types of Neural Networks:- (1) ANN,(2)CNN,(3)RNN 16 SKIT/ECE/IndustrialTraining-AyushJajodia ANN (Artificial Neural Networks) A single neuron can be considered as a Logistic Regression,and ANN is a group of multiple neurons at each layer , that is Input, Hidden ,Output. It is based on a collection of connected units or nodes called artificial neurons, which loosely model the neurons in a biological brain. Once the ANN is trained, it can be used to predict the outcome of another new set of input data. 17 SKIT/ECE/IndustrialTraining-AyushJajodia CNN (Convolutional Neural Network) It is Deep Learning algorithm that can take an input image and be able to differentiate one from another. This neural network model uses a variation of multilayer neurons and contains one or more convolutional layers that can be entirely connected. It is used in image recognition and processing that is specifically designed to process pixel data. 18 SKIT/ECE/IndustrialTraining-AyushJajodia TIME SERIES ANALYSIS A time series is simply a series of data points with time order this means we pick one or more variables and note their values over time. Time series forecasting is the use of a model to predict future values based on previously observed values. 19 SKIT/ECE/IndustrialTraining-AyushJajodia PROJECT 1. Geospatial use: Determining with the help of data sets whether Global Warming exists or not with the help of graphical analysis. SKIT/ECE/IndustrialTraining-AyushJajodia 20 21 SKIT/ECE/IndustrialTraining-AyushJajodia SUMMARY In this presentatiin we understood the meaning of machine learning and its importance. Different Machine Learning Algorithms and how these are used to build a model to predict dependent variable such as price etc. 22 SKIT/ECE/IndustrialTraining-AyushJajodia 23 SKIT/ECE/IndustrialTraining-AyushJajodia THANK YOU !! 24 SKIT/ECE/IndustrialTraining-AyushJajodia