artificial intelligence

Using artificial intelligence, Pinterest is modifying the function of its associated image pins. An important part of the curation service, Pinterest provides approximately 75 billion image pins on its platform to its 150 million customers.

Pinterest is progressively moving to deeper knowledge to help users discover new products, recipes, and ideas that connect to their importance – an AI that feeds vast amounts of information into neural networks that affect how How does the human mind move until this autonomy will accept the other. Information.

This function of this associated image pin of the platform pops up under an image pin with comparable suggestions. Those ideas have emerged in the past from an image board from which other customers saved similar pins. While this technique offered an endless pool of ideas to play with, it rarely caused the loss of an image pin reference.

Now, in addition to your current board co-event design, Pintrest uses information from your activity session (such as image pins that you have saved or clicked on), with their latest interactions with image lapel pins, less – Except for image pins. Pinterest suggests that complex deep knowledge planning, called Pin2Vec, helps produce more appropriate user behavior-based results. For example, by comparing the two methods for a bottle of wine with related image pins, Pinterest states: “The co-occurrence only discovered pictures of bottled wine, while Pin2v offered suggestions for wine-based drinks Discovered. This indicates that in the same. Time series, Pinners effectively saved bottled Pin wine and Pins wine cocktails. Pin 2 contributes to reducing the vetch gap. ”

The firm said some of the new technology was already showing success, with initial pins in associated trials indicating a 5% worldwide increase. Overall, nearly half of all Pinterest pin commitments are now for related pins, again showing how important the design process is on the platform.

“Pin2Vec is a major step towards increasing the relevance of related pins,” Pinterest wrote in its blog post. “We are now rapidly advancing our designs and analyzing more signals to better optimize Pinner’s suggestions around the world.”


Please enter your comment!
Please enter your name here