The power to choose
.20 Oct

The power to choose

I recently red the excellent article by Caroline Martin: Will the algorithms put an end to the tyranny of choice?

Effectively GAFAM (Google, Apple, Facebook, Amazon, Microsoft) have phenomenal amounts of data on all their users. They naturally tend to adopt a “Deep Profile” strategy that seeks to monetize all available data. I have a French friend with a Democratic sensitivity who married an American, rather Republican. When he searches for economic information on his favourite search engine, he finds primarily articles of Democratic sensitivity, and his wife of rather Republican sensitivity. It is very disturbing and one wonders what will become our democracies when GAFAM will know our tastes and our political preferences better than ourselves. As a user, I appreciate that algorithms help me in my choices and decisions. But I have no desire that they reduce my choices and encloses me in my convictions.

Traditional personalization editors classify each user into a segment. The risk is therefore very high that these algorithms narrow the vision of each one to this particular segment. The Internet user is not considered as unique but as a member of a group to which a marketing strategy will be applied. The user is affiliated to a segment arbitrarily and it is very difficult to get out.

The most obvious alternative is to choose a “No Profile” strategy with non-contextual algorithms. They do not record personal data, are unable to customize content based on a search context, and are much less relevant than contextual algorithms.

Fortunately, there is a third way adopted by MyDreamMatch: “Fast Profile“. It has two complementary objectives:

  1. Personalize the site’s content so that the user finds more relevant results faster.
  2. Give the user the power to choose and discover new information or products at the right time.

This strategy is based on the following principles:

  1. Detect user preferences and immediate intentions for a particular project. The Artificial Intelligence understands very quickly when the user begins to discover the whole offer without particular preferences, when his intentions become more precise and he concentrates on particular products, when he is ready to make a decision, or when he changes his mind because he has another purchase project.
  2. Record the most complete as possible data on the interest expressed by the user during his / her navigation experience. When the user is interested in a product, he is sensitive to all its objective and subjective characteristics. He is not always aware of it, but the Artificial Intelligence has all the capacities to discover it. It’s a very intuitive process. These data are ephemeral and we use only the essential data for a decision in the identified project. In this phase, we are looking for Micro Data that concern only the user as opposed to Big Data, which concerns the greatest number of users.
  3. Update the user profile as quickly as possible and create a new profile when we identify a new project. The quality of the data analyzed in real time and the speed at which it is processed determines the relevance of the content personalization.

MyDreamMatch Fast Profile thus facilitates the life of the user, increasing his power to choose.

Software engineer from Supélec Paris, I worked as a developer, architect and then manager for many years at the Research Center of IBM and then in the Cisco European organization.
Passionate about Artificial Intelligence, I decided in 2015 to create my company MyDreamMatch in order to make personalized recommendations accessible and easy to use by the largest number of web sites.

Write a post