Yanying Mao

Title: Yanying Mao: Innovator in Few-Shot Relation Classification

Introduction

Yanying Mao is a prominent inventor based in Hunan, China. He has made significant contributions to the field of artificial intelligence, particularly in relation classification. His innovative work has led to the development of a unique method and apparatus that enhances the efficiency of few-shot relation classification and filtering.

Latest Patents

Yanying Mao holds a patent titled "Method and apparatus for few-shot relation classification and filtering." This patent describes a comprehensive method that includes constructing a coarse-grained filter to process an unlabeled corpus. The goal is to identify candidate instances that share the same entity as a seed instance and possess similar semantics. Additionally, a fine-grained filter is utilized to refine these candidates, ensuring that the selected instances align with the same relation concept as the seed instance. The invention also incorporates a false positive instance correction module, which adjusts the proportion of negative samples used during classifier training. This innovative approach allows for effective training based on a limited number of labeled instances.

Career Highlights

Yanying Mao is affiliated with the National University of Defense Technology, where he continues to advance his research in artificial intelligence. His work has garnered attention for its practical applications in various domains, showcasing his expertise in the field.

Collaborations

Yanying has collaborated with notable colleagues, including Fei Cai and Peihong Li, who contribute to his research endeavors. Their combined efforts enhance the impact of their work in the realm of relation classification.

Conclusion

Yanying Mao's innovative contributions to few-shot relation classification demonstrate his commitment to advancing artificial intelligence. His patent reflects a significant step forward in the field, showcasing the potential for improved classification methods.