Page 208 - Ai_417_V3.0_C9_Flipbook
P. 208

What is Modelling?


              Modelling or data modelling is defined as the process of designing decision-making algorithms that have to
              be trained on a set of data (which was acquired at the data acquisition stage for the problem you scoped in the
              problem-scoping stage) and applying that learning to recognise certain types of patterns.

              Once the data is visualised and trends are formed, we need to work with algorithms to prepare the AI model. This
              can be done by designing our models or using the existing AI models. Before we go into the details of modelling,
              let us first understand the following important terms:
                 • Artificial Intelligence: AI refers to any technique that enables computers to mimic or imitate, develop, and

                 demonstrate human intelligence. They are machines that can perform tasks that they are programmed for. AI
                 enables machines to think without any human intervention.
                 • Machine Learning: Machines need to learn the ways of humans by learning the techniques and processes. So
                 machine learning is a subset of artificial intelligence that uses statistical methods that enable machines to

                 improve with experiences. So machines learn from their mistakes and take them into consideration in the next
                 iteration,  this  way  they  keep  improving  with  experience.  For  example,  Snapchat  filters  and  Netflix
                 recommendations.
                                                                                                    Artificial
                 • Deep  Learning: Machines can draw  meaningful inferences from  large            Intelligence
                 volumes of datasets. In deep learning, the machine is trained with a huge
                 amount  of  data,  which  helps it train  itself.  Deep  learning  is a  machine    Machine
                                                                                                     Learning
                 learning algorithm that is  inspired by the functionality of our brain cells

                 called neurons. For example,  Google Translate and  image  recognition  in           Deep
                 social media apps.                                                                  Learning

              Artificial intelligence is an umbrella term that includes machine learning and
              deep learning. Deep learning follows a specific learning approach,  which is a subset of machine learning comprising
              multiple machine learning algorithms.
              Let us understand the difference between artificial intelligence, machine learning and deep learning.


                     Artificial Intelligence            Machine Learning                     Deep Learning
                It aims at making a machine that   It aims at making a machine that   It aims at building neural network
                mimics human intelligence.       can learn through data and solve   that can help in discovering patterns

                                                 complex problems.                or trends.
                It is a subset of data science.  It is a subset of AI.            It is a subset of machine learning.

                It is the simulation of intelligence   It is the training of machines to   It is the process of using artificial
                in machines.                     take decisions with experience.  neural networks for solving complex
                                                                                  problems.

                Examples: Robotics, natural      Examples: Decision trees, random  Examples: Convolutional neural
                language processing, computer    forests, support vector machines,   networks (CNNs), recurrent neural
                vision, expert systems, etc.     neural networks, etc.            networks (RNNs), generative
                                                                                  adversarial networks (GANs), etc.


                    206     Touchpad Artificial Intelligence (Ver. 3.0)-IX
   203   204   205   206   207   208   209   210   211   212   213