Meilan Ji

Title: Meilan Ji: Patent Inventor in Language Processing and Machine Learning.

Introduction

Meilan Ji is an inventor based in Beijing, China, associated with SAS Institute Inc. His patent work focuses on combining language models, natural language processing, information extraction, and machine learning to analyze and summarize documents. Ji is credited with 2 patents.

Latest Patents

Ji’s listed patents address document summarization and topic modeling.

“System and Method for Combining Language Models with Natural Language Processing for Summarization” describes a system that extracts and selects text segments from a collection of documents, creates a compressed representation, and supplies it to a language model to generate a summary of trends across the documents.

“System and Method for Combining Unsupervised Machine Learning and Information Extraction Models for Topic Modeling” describes extracting text segments from documents using rules related to entities, word lists, or grammatical patterns. An unsupervised machine learning model then processes at least some of those segments to identify topics.

Career Highlights

Ji’s listed inventions bring together rule-based information extraction and machine learning techniques. One focuses on generating summaries with language models, while the other uses unsupervised learning to identify topics in document collections. Both are associated with SAS Institute Inc.

Collaborations

Ji’s listed coworker is Teresa S. Jade. The provided information does not specify her role in the patents or describe particular collaborations.

Conclusion

Meilan Ji’s patent work centers on practical methods for extracting meaning from documents, including summarization and topic modeling. His inventions combine information extraction with language and machine learning models to support these data-processing tasks.