Microsoft technology licensing, llc (20240346533). SELF-SUPERVISED SYSTEM GENERATING EMBEDDINGS REPRESENTING SEQUENCED ACTIVITY simplified abstract

SELF-SUPERVISED SYSTEM GENERATING EMBEDDINGS REPRESENTING SEQUENCED ACTIVITY

Organization Name

microsoft technology licensing, llc

Inventor(s)

Mayank Shrivastava of Woodinville WA (US)

Sagar Goyal of Vancouver (CA)

Sahil Bhatnagar of Vancouver (CA)

Pin-Jung Chen of Bellevue WA (US)

Pushpraj Shukla of Dublin CA (US)

Arko P. Mukherjee of Issaquah WA (US)

SELF-SUPERVISED SYSTEM GENERATING EMBEDDINGS REPRESENTING SEQUENCED ACTIVITY - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240346533 titled 'SELF-SUPERVISED SYSTEM GENERATING EMBEDDINGS REPRESENTING SEQUENCED ACTIVITY

The disclosure describes a system for generating embeddings representing sequential human activity using self-supervised deep learning models.

• An encoder-decoder creates user-specific journeys based on human activity data from various sources.
• Sequential feature vectors represent events in the user activities.
• User-specific embeddings are created for each user, updating in real-time as new data is received.
• The embeddings can be fine-tuned with labeled data for specific predictive models.
• Predictive models utilize the embeddings for product recommendations and predictions.

1. 1. 1. Potential Applications:

This technology can be applied in customer relationship management, personalized marketing, and recommendation systems.

1. 1. 1. Problems Solved:

This technology addresses the challenge of accurately predicting user behavior and providing personalized recommendations based on sequential human activity data.

1. 1. 1. Benefits:

The system enables more accurate predictions, personalized recommendations, and improved customer engagement.

1. 1. 1. Commercial Applications:

"Predictive Customer Engagement System Utilizing Sequential Human Activity Embeddings"

1. 1. 1. Prior Art:

Prior research in deep learning models for sequential data analysis and recommendation systems can provide insights into similar technologies.

1. 1. 1. Frequently Updated Research:

Stay updated on advancements in deep learning models for sequential data analysis and personalized recommendation systems.

1. 1. 1. 1. Questions about Sequential Human Activity Embeddings:

1. How does this technology improve customer engagement? 2. What are the potential challenges in implementing this system in real-world applications?

Original Abstract Submitted

the disclosure herein describes a system for generating embeddings representing sequential human activity by self-supervised, deep learning models capable of being utilized by a variety of machine learning prediction models to create predictions and recommendations. an encoder-decoder is provided to create user-specific journeys, including sequenced events, based on human activity data from a plurality of tables, a customer data platform, or other sources. events are represented by sequential feature vectors. a user-specific embedding representing user activities in relationship to activities of one or more other users is created for each user in a plurality of users. the embeddings are updated in real-time as new activity data is received. the embeddings can be fine-tuned using labeled data to customize the embeddings for a specific predictive model. the embeddings are utilized by predictive models to create product recommendations and predictions, such as customer churn, next steps in a customer journey, etc.