Kamesh Madduri

Title: Kamesh Madduri and His Patent on Graph-Directed Key Phrase Recommendation

Introduction

Kamesh Madduri is an inventor based in State College, Pennsylvania. His patent work includes a method for recommending key phrases for items listed on an online platform, and he is associated with eBay, Inc.

Latest Patents

Madduri is credited with one patent: “Graph-Directed Key Phrase Recommendation Based on Item Similarity.” The invention uses a dataset connecting listed items, their titles, and associated key phrases. It builds a data structure that links title tokens to items and items to key phrases. Given a seed item’s title, the system identifies similar items by examining how often the title’s tokens occur in the data structure, then recommends key phrases associated with those similar items. This approach can help generate relevant phrases for item listings.

Career Highlights

Madduri’s patent reflects work on applying data structures and item similarity to improve key phrase recommendations in a listing platform. His association with eBay, Inc. connects this work to online commerce and product discovery.

Collaborations

Soumik Dey and Hansi Wu are identified as Madduri’s coworkers. The available information does not specify their roles in the patent or whether they collaborated on it.

Conclusion

Kamesh Madduri’s patent describes a data-driven way to recommend key phrases by finding items similar to a listing, highlighting the use of graph-directed relationships in online commerce.