Page 185 - Ai_V1.0_Class9
P. 185

The kind of math needed for AI includes:
                 1.   Probability and Statistics (exploring data): Probability theory and statistics are key fundamentals for many
                    AI algorithms, particularly those involving machine learning. It is useful in tasks such as natural language
                    processing, computer vision, and decision-making.
                 2.   Linear Algebra (finding out unknown or missing values): Linear algebra is involved in large-scale data
                    processing playing a vital role in machine learning and AI. It performs operations in neural networks, image
                    processing, and data transformations.
                 3.   Calculus (training and improving AI model): Calculus is essential for understanding the best possible
                    solution algorithms used in machine learning. It minimise mistakes and maximise the parameters of machine
                    learning models.
                 4.  Graph Theory: Graph theory is used in AI representing trends using data visualisation.
                 5.  Information Theory: Information theory provides mathematical tools for data analysis.
                 6.   Logic and Set Theory: Logic and set theory are used in concepts like expert systems, database systems,
                    and knowledge graphs. Mathematics and AI are deeply integrated fields, with mathematics providing the
                    theoretical foundation for many AI algorithms and techniques.
                 7.   Algorithm design: The design of the algorithms often uses mathematical principles and structures such as
                    functions, matrices and graphs.


                         Statistics


                 Statistics is used for collecting, exploring, and analysing the data. It also helps in concluding data. It enables AI
                 systems to detect patterns, identify relationships, and infer conclusions from data.
                    • Data is collected from various sources.
                    • Data is explored and cleaned to be used.
                    • The analysis of data is done to understand it better.
                    • Conclusions and decisions can be made from the data.
                 Let’s consider an example to illustrate these steps:
                 A school wants to improve the performance of its students and decides to collect data on study habits and
                 grades.

                    • Collecting Data:  The school conducts a survey where students report the number of hours they study each
                   week and their grades in various subjects.
                    •  Exploring and Cleaning Data: The school first looks at the data to find patterns, like the range of study hours
                   and grades. They also clean the data by fixing any missing or incorrect information (e.g., if some students didn’t
                   fill in all the fields or gave unrealistic answers).
                    • Analysing Data: The school summarises the average study hours and grades to get an overview. They also
                   check if there is a significant relationship between study hours and grades.
                    •  Drawing Conclusions: The analysis might reveal that students who study for more hours tend to have higher
                   grades. Based on this conclusion, the school might decide to implement study support programs to encourage
                   students to study more. Thus, we can say statistics helps in transforming the raw data into meaningful insights,
                   enabling better decisions and strategies in various fields such as business, healthcare, education, and more.



                            a  Brainy Fact


                        As  per  the  statistics,  Netflix  makes  $1  billion  annually  from  automated  personalised
                        recommendations.



                                                                           Maths for AI (Statistics & Probability)   183
   180   181   182   183   184   185   186   187   188   189   190