Page 145 - Artificial Intellegence_v2.0_Class_11
P. 145

Video Session

                       Scan the QR code or visit the following link to watch the video: How are weather forecasts made?
                       https://www.youtube.com/watch?v=fdErsR8_NaU
                       After watching the video, answer the following question:

                       What you learned from this video?










                        Google’s Nowcasting

                 Google calls the technology "Nowcasting" because it is set up to predict weather zero to six hours in advance and focus on
                 weather events like thunderstorms that change rapidly from clear skies to heavy rain to gusty winds and vice versa.
                 One of the biggest problems with existing forecasting systems is the amount of data and the processing power required to
                 process and understand all of that data. The U.S. National Oceanic and Atmospheric Administration (NOAA) alone collects
                 100 terabytes of data every day. The forecasting programs run on supercomputers, but they require huge computing power.
                 Google presents its method as much simpler. In essence, the method turns weather forecasting into a computer vision
                 problem. Based on progressive images of cloud formation and movement over a short period, a machine learning
                 algorithm predicts how the pattern will develop in the next few hours.
                 Specifically, Google uses a convolutional neural network (CNN), a neural network whose architecture is particularly used for
                 image analysis. This means that the neural network only learns from the training data and does not include any knowledge
                 of how the atmosphere works. The patterns are identified from the images that have been fed into the AI model.
                 Google claims that its algorithm can make predictions with a resolution of one kilometer and an expectancy of five to
                 ten minutes.

                        Price Forecast for Commodities

                 Commodity trading and pricing have become increasingly complex over the years. Price predictions predict the price
                 of a commodity/product/service by evaluating various factors such as its properties, demand, seasonal trends, prices of
                 other commodities (e.g. fuel), offers from numerous suppliers, loyalty, and commitment. This helps companies define an
                 optimal time to purchase a commodity, adjust prices for products or services that require a commodity (wood, coffee,
                 gold), or assess the investment attractiveness of fixed assets.
                 In general, price forecasting uses supervised learning and is done through descriptive and predictive analysis.


                 Descriptive Analysis
                 Descriptive analysis is based on statistical methods that include data collection, analysis, interpretation, and presentation.
                 Descriptive analysis enables you to turn raw observations into insights that can be understood and shared. In short, this
                 type of analysis helps answer the question, what happened?


                 Predictive Analysis
                 Predictive analytics consists of analysing current and historical data to predict the probability of future events, results,

                 or values  in the context of price predictions. Predictive analytics requires numerous statistical techniques, such as data
                 mining (identifying patterns in the data) and machine learning.

                                                                             AI Applications and Methodologies  143
   140   141   142   143   144   145   146   147   148   149   150