IBM plans to offer you comfort food based on weather

IBM plans to offer you comfort food based on weather

“Ask not what you can do for your country. Ask what’s for lunch.” ― Orson Welles. The love of food is perhaps the sincerest. One cannot help but love a good piece of cake, or a pie unconditionally. People of all ages, regions, and cultures have some comfort food that is consumed with affection, love, and even nostalgia. And most often, the type of food that you crave for depends on the weather. There’s nothing quite like a hot cup of tea in the rains, a pumpkin spiced latte in the autumn, and heavenly turkey just at the advent of winters on Thanksgiving. And it is this craving that IBM plans to satiate.

Patented Technology

A recently published patent from IBM talks about presenting ads related to food and beverages, to users. The ads will be triggered by the weather. For instance, the weather around you will be used to present ads related to food & beverages to you. Your responses to those ads, and your cravings, as indicated by you through ingredients, and your responses, will be analyzed in conjunction with your local weather information by IBM’s Chef Watson, and relevant ads related to food that you want to consume will be presented to you. Chef Watson is IBM’s Artificial Intelligence software and has analyzed tons of data related to recipes, and knows the flavor that each ingredient possesses. As such, any ingredient that you wish to enjoy in a recipe will be analyzed, and relevant ads related to that ingredient and recipes including that ingredient generated by Chef Watson will be presented to you. For example, if you indicate that you crave something with a little pumpkin spice, your location, the weather around you, and other details like your age, will be analyzed by Chef Watson, and you will be presented with relevant ads, like an ad for a pumpkin spice latte that you may be able to order from your local coffee shop. Technical Details User location from user devices such as smartphones is received. Based on user location weather data related to users is received from weather databases like The Weather Channel®. Then, a relevant ad for edible products including food and beverages, local services, such as restaurants and movie theatres, is transmitted to the user devices based on the weather. For instance, if the weather is cold, an ad for hot soup available for instance at a local restaurant may be presented to the user. Further, user requests for specific products are received, and the requests are coupled with contextual data including weather data and user preferences (as determined over a few interactions). IBM’s AI Watson then analyzes all these parameters and provides a relevant ad to the user. The requests may be in natural language, such as “help me make a soup recipe with coconut.” Then, based on machine learning, multiple recipes are determined which are also influenced by the weather condition around the user. Then, relevant ads including recipes with ingredients mentioned in the ads may be presented to the user. For instance, output dialog provided to the user by an AI agent powered by Watson may state that “it’s cold outside so here are some great hearty soup recipes” or “it’s a hot day so I created some great cold soup recipes for you”.


Although all technical aspects mentioned in the patent application are well known, the implementation is somewhat unique. Relevant targeted ads may also be provided for other products required in specific weathers. For instance, it may not be farfetched to imagine ads for jackets being presented to users during winters, or ads related to swimwear presented during hot days to users living around beaches.
Patent Information Publication Number: US20180341979A1 Patent Title: Cognitive advertising triggered by weather data Publication date: 2018-11-29 Filing date: 2017-10-24 Inventors: Donna K. Byron; Carmine M. DiMascio; Brian Hull; Florian Pinel; Robert N. Redmond, JR.; Edward E. Seabolt; William S. Spann, III; Domenic A. Venuto Original Assignee: International Business Machines Corp