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

