Microsoft Technology Licensing, LLC

Linkedin Corporation

Rupesh Gupta

Hsiao-Ping Tseng

Romer E Rosales

Shaunak Chatterjee

Ravi Kiran Holur Vijay

Joshua Daniel Hartman

Rishi Jobanputra

Romer E Rosales-Delmoral

Shubhanshu Nagar

Lingjie Weng

Cameron Alexander Lee

Ryan Oblak

Daniel ManHung Wong

Jeffrey Gee

Deepak Kumar

Gungor Polatkan

Rohan Ramanath

Shipeng Yu

Jinyun Yan

Xiaoyu Chen

Yiping Yuan

Konstantin Salomatin

Sara Smoot Gerrard

Onkar Anant Dalal

Alexander Ovsiankin

Guanfeng Liang

Anand Kishore

Peng Du

Jiaqi Ge

Mahdiehsadat Mirian Hosseinabadi

Mathew H Teoh

In some embodiments, a computer system generates a recommendation for a user of an online service based on user actions that have been performed by the user within a threshold amount of time before the generation of the recommendation. For each user action, the computer system determines an intent classification that identifies an activity of the user and that corresponds to different types of user actions, as well as a preference classification that identifies a target of the activity, and then stores these intent and preference classifications as part of indications of the user actions for use in generating different types of recommendations using different types of recommendation models. Additionally, the computer system may use mini-batches of data from an incoming stream of logged data to train an incremental update to one or more recommendation models.

Recommendations using session relevance and incremental learning

Embodiments include technologies to apply at least one machine learning model to features of a search query, features of a searcher user, features of a searchee content item, and features of a searchee user, produce a first outcome prediction that represents a probability of a first objective relating to engagement of the searcher user with a content item in an online system and a second outcome prediction that represents a probability of a second objective relating to engagement of the searchee user with the online system responsive to the engagement of the searcher user with the content item, apply a multi-objective optimization solver to the first objective, the second objective and an outcome prediction that is a combination of the first outcome prediction and the second outcome prediction, and generate a serving function for a search engine based on the first objective, the second objective, and the outcome prediction.

Results ranking with simultaneous searchee and searcher optimization

A method can include determining a first probability that a first member of members of a website will visit the website within a specified time window if the first member is provided an intervention at a specified time, determining a second probability that the first member will visit the website within the specified time window without being provided the intervention, determining a difference between the first and second probability, and in response to determining the difference is greater than a first specified threshold, providing the intervention at the specified time.

Predicting site visit based on intervention

A system, a machine-readable storage medium storing instructions, and a computer-implemented method are described herein are directed to a Message Diet Engine that generates a pool of messages for a plurality member accounts of a social network service. Each message being of a respective message type from a plurality of message types and targeted to a specific member account. For each respective member account, the Message Diet Engine selects a minimum number of messages, from the pool of messages, targeted to the respective member account that prompts an expected social network activity target and avoids an expected number of complaints. Based on the selected minimum number of messages for each respective member account, the Message Diet Engine identifies a total minimum number of messages, from the pool of messages, to be sent to the plurality of member accounts that prompts an expected total social network activity target and avoids a total expected number of complaints.

Active user message diet

A system and method for content generation and targeting using machine learning are provided. In example embodiments, a probability that a user will visit a webpage based on historical data is calculated. A probability that the user will engage with a particular content category based on past user engagement is calculated. In response to the probability of the user engaging with the particular content category being equal to or greater than a first threshold, the content is generated. Further, in response to the probability of the user not visiting a webpage meeting or exceeding a second threshold, the generated content is sent to the user.

Content generation and targeting using machine learning

This disclosure relates to systems and methods that include configuring a machine learning system to train on a plurality of messages, the machine learning system to output an expected number of positive responses and an expected number of negative responses based on an input message, determining a threshold differential and a weight value using responses to the plurality of messages, and sending the input message in response to a differential between the expected number of positive responses and the weight multiplied by the expected number of negative responses being above the threshold differential.

Deterministic message distribution

This disclosure relates to systems and methods that include configuring a machine learning system to train on a plurality of messages transmitted to target groups of an online social networking service, determining a threshold differential and a weight value using responses to the plurality of messages, and send the input message to the target in response to a differential between the expected number of positive responses and the weight multiplied by the expected number of negative responses being greater than the threshold differential.

Techniques for expanding a target audience for messaging

This disclosure relates to systems and methods for managing multiple messages. In one example, a method includes determining a message transmission frequency threshold for a member of an online social networking service using responses from the member; receiving a message that is to be transmitted to the member; storing the message, without transmitting the message to the member, in a digest of messages for the member; and transmitting the digest to the member in response to a send score for the digest exceeding a send score threshold, the send score calculated using the number of messages in the digest.

Managing multiple messages

Systems and methods for storing less than a threshold number of media content activity levels for media content objects at an online social networking service, identifying, using the stored media content activities, a threshold number of media content objects associated with a higher number of the media content activities occurring over a recent threshold period of time, receiving an indicator indicating that one of the identified media content objects is unprofessional, and propagating the indicator to each activity that includes the unprofessional media content object.

Managing unprofessional media content

Systems and methods for providing push notifications of social network content items include determining first and second engagement values for first and second content items, respectively, provided by first and second content item sources, respectively, based on previous interactions with content items. First and second utility values are determined for the first and second content items, respectively, based on a mathematical operation applied to the first and second engagement values, respectively, the first and second selection metrics, respectively, and, for the first utility value, the value metric. One of the first and second content items are determined as a push content item based on a difference between the first and second utility values and displayed on a user device.

Second-pass ranker for push notifications in a social network

Growing community of inventors

Sunnyvale, CA, United States of America

Rupesh Gupta