Syndicate content

Thanksgiving Woes? IBM and Big Data May Help.

Tanya Gupta's picture

In just about a week, on Thursday November 28, people all over the United States will kick off the "holiday season" with the celebration of Thanksgiving Day. While the day's significance is both historical and profound, in modern times it consists of a lot of shopping and a big meal with family and friends gathered around the dinner table. Pre-thanksgiving is a time to be on the lookout for creative new recipes.  Sure, we can get recipes from magazines, websites and friends and while they may be special, they will not be unique.  Wouldn’t it be nice to have an app that would create a special unique recipe just for you? A delightful recipe that has never been executed before.  Well the idea is not as futuristic as it sounds. It may be here sooner than you think.  IBM and big data have a lot to do with this particular innovation.
Can computers be creative?  IBM thinks they can.  IBM scientists Lav R. Varshney and other members of an IBM team, have used data sets and proprietary algorithms in the daunting field of the culinary arts to develop a computational creativity system. The data sets they have used are recipes, molecular level food related data and data about the compounds, ingredients and dishes that people like and dislike.  They then developed an algorithm that produces thousands or millions of new ideas from the recipes.  The recipes are then evaluated to select the best ones that combine ingredients in a way that has never been attempted before.  Humans can interact with the system by choosing a key ingredient and the kind of cuisine.

Of course in order to achieve all this, a template for the creative process is also needed.  To do so the researchers used Sawyers’s  model for creativity. The model has the following stages:

  1. Find the problem
  2. Acquire knowledge:
  3. Gather related information
  4. Incubation
  5. Generate ideas
  6. Combine ideas
  7. Select best ideas:
  8. Externalize the idea

So what were the results? Their research  paper shows that recipes created by the computational creativity system have been rated as more creative than existing recipes in online repositories.  In addition, professional chefs at various hotels, restaurants, and culinary schools have said that “the system helps them explore new vistas in food”.
This is a first for computational creativity applied in a scientific area. Recipes are instructions for preparing (process) a particular dish (output) using specified ingredients (inputs). Therefore recipes are within the realm of science. 
Post-Renaissance, creativity has been seen as a uniquely human characteristic.  Before the Renaissance, the concept of creativity was closely associated with creation and it was argued that God was the ultimate creator as only he could create from nothing (creatio ex nihilo).  The vision of creativity as a quality that has set us apart from other co-inhabitants of this earth is likely to be shattered with IBM’s latest research, not by pitting us against animals but against a computer program.  So why did this happen only now?  Big data is making this and other hard problems solvable.  A recipe is nothing more than a program.  If programs can be creative about creating new programs - are we at the cusp of automating the computer evolution?
Photo Credit: ccho
Follow PublicSphereWB on Twitter



Submitted by itoctopus on

Hi Tanya,

The problem with this algorithm is that it only provides you with ingredients, and not with cooking instructions. It's not hard to gather data from all the recipes in the world, find which ingredients each recipe has, and find which recipes are successful, and then combine compatible ingredients from successful recipes together. What's really hard it so provide the cooking instructions, and not just the ingredients.

By the way, as a programmer myself, I read the paper and I discovered that it definitely doesn't take cooking instructions into consideration - perhaps that's why they didn't, in their paper, provide any mix of ingredients that includes meat.

PS: This same concept can be applied, perhaps with more success, in the medical field. In fact, it was the topic of an R&D paper we submitted last year.

Add new comment