logo

For the Inventor, By the Inventor

New York, NY, United States of America

Melinda Han Williams

This inventor holds 13 USPTO granted patents and 4 published patent applications. Top assignee: Dstillery, Inc.. Active years: 2018-2025.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Melinda Han Williams. Retrieved from https://idiyas.com/inventor/melinda-han-williams

Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile

USPTO Granted Patents = 13 

% Patents Active = 100.0

Average Co-Inventor Count = 4.4

ph-index = 3

Forward Citations = 33(Granted Patents)


Location History:

  • New York, NY (US) (2018 - 2021)
  • Brooklyn, NY (US) (2021 - 2024)

Company Filing History:

goldMedal13 out of 24 
 
Dstillery, Inc.
 patents

Years Active: 2018-2025

Loading Chart...
13 patents (USPTO):Explore Patents
1. 12373738 - Transformers adapted for predictive targeting based on web browsing history (2025-07-29),
2. 12182723 - Iterative online learning to improve targeted advertising (2024-12-31),
3. 12045703 - Artificial intelligence and/or machine learning models trained to predict user actions based on an embedding of network locations (2024-07-23),
4. 11921732 - Artificial intelligence and/or machine learning systems and methods for evaluating audiences in an embedding space based on keywords (2024-03-05),
5. 11768844 - Artificial intelligence and/or machine learning systems and methods for evaluating audiences in an embedding space based on keywords (2023-09-26),
6. 11727313 - Unsupervised machine learning for identification of audience subpopulations and dimensionality and/or sparseness reduction techniques to facilitate identification of audience subpopulations (2023-08-15),
7. 11699109 - Artificial intelligence and/or machine learning models trained to predict user actions based on an embedding of network locations (2023-07-11),
8. 11580117 - Machine learning system and method to map keywords and records into an embedding space (2023-02-14),
9. 11128722 - Evaluating the authenticity of geographic data based on user activity correlations (2021-09-21),
10. 11068935 - Artificial intelligence and/or machine learning models trained to predict user actions based on an embedding of network locations (2021-07-20),
11. 10902009 - Machine learning system and method to map keywords and records into an embedding space (2021-01-26),
12. 10212239 - Evaluating the authenticity of geographic data based on spatial distribution (2019-02-19),
13. 9948733 - Evaluating authenticity of geographic data associated with media requests (2018-04-17),

Title: Melinda Han Williams: Innovator in Predictive Targeting Technologies

Introduction

Melinda Han Williams is a prominent inventor based in New York, NY (US). She has made significant contributions to the field of predictive targeting technologies, holding a total of 13 patents. Her work focuses on enhancing targeted advertising through innovative methods that leverage web browsing history.

Latest Patents

One of her latest patents involves transformers adapted for predictive targeting based on web browsing history. This invention relates to computer-implemented methods and memory systems that store code designed to be executed by a processor. The method includes training a first transformer with a variety of website visitation records to predict a missing website in a sequence of websites. The first transformer comprises a foundation model and a first prediction head. Additionally, a second transformer model is trained with conversion event data to predict the probability of a conversion event occurring, also utilizing a foundation model and a second prediction head.

Another notable patent focuses on iterative online learning to improve targeted advertising. This method involves accessing web browsing history for multiple users and generating embedding vectors based on this data. The model calculates the probability of conversion events for various websites and selects a subset of websites based on their associated probabilities. The model parameters are then updated using the obtained conversion rates, enhancing the effectiveness of targeted advertising strategies.

Career Highlights

Melinda is currently employed at Dstillery, Inc., where she continues to develop innovative solutions in the realm of digital advertising. Her expertise in predictive modeling and data analysis has positioned her as a key player in her field.

Collaborations

Throughout her career, Melinda has collaborated with notable colleagues, including Amelia Grieve White and Christopher Allen Jenness. These partnerships have contributed to her success and the advancement of her projects.

Conclusion

Melinda Han Williams is a trailblazer in the field of predictive targeting technologies, with a strong portfolio of patents that reflect her innovative spirit. Her work at Dstillery, Inc. continues to shape the future of targeted advertising, making her a significant figure in the industry.

Profile summary based on public USPTO records.
Data Sources: USPTO Patent Grant XML, Patent Center, EPO & CIPO • Normalized by IDiyas Innovation Graph. Methodology & provenance architecturePlease report any incorrect information to [email protected]
Loading…