An online messaging application receives content that describes a conversation conducted via the online messaging application between a user and a business entity having a presence on the online messaging application. Features are extracted from the content and a transaction prediction model is applied to predict a likelihood that the conversation describes a transaction between the user and the business entity. Based on the predicted likelihood that a transaction has occurred, content items are identified for presentation to the user.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.