Pesonalising weather forecasts using AI techniques
<p>Communicating the scientific data of the weather forecasts to the general public has always been a challenge. Using computer graphics&#8217; visual representations to convey the message to television viewers and through weather apps and websites has certainly helped a lot to popularize the weather forecast consumption by the general public. However, these representations are not information rich since they are abstraction; moreover they are not always very actionable on the receiver side to help one decide how s/he will &#8220;live&#8221; the forecast weather conditions. Therefore, there is a need to personalize the forecast based on past user experience and personal needs. The forecast has to become more human- and needs-oriented and more focused to the particular requirements of each individual person. The challenge is to move from providing the abstraction of atmospheric information to a real sense of how the weather will "feel" to the individual.</p><p>We therefore propose a new co-creation process in which the audience is called on to provide a daily feedback on how they lived the weather conditions personally, so that, &#8220;my personal forecast&#8221; can be produced making the forecast more actionable on the user side. Preliminary, but more personalized, such attempts include the &#8220;feels like&#8221; temperature forecasts. To arrive at the &#8220;my personal forecast&#8221;, AI-based recommender systems need to be applied, using fuzzy logic as the appropriate method for the user to express how s/he actually lived personally lived weather conditions every day. Over time this information can then be used to transform science-based descriptions of weather conditions into a sense of how the weather will be experienced at a personal level.</p>