Jian Sun

Improving Cross-Domain Chinese Word Segmentation with Word Embeddings

2019 · arXiv (Cornell University)

Cross-domain Chinese Word Segmentation (CWS) remains a challenge despite recent progress in neural-based CWS. The limited amount of annotated data in the target domain has been the key obstacle to a satisfactory performance. In this …

Learning Long-text Semantic Similarity with Multi-Granularity Semantic Embedding Based on Knowledge Enhancement

2020

research-article Learning Long-text Semantic Similarity with Multi-Granularity Semantic Embedding Based on Knowledge Enhancement Share on Authors: Deguang Peng Chongqing Megalight Technology co. LTD, China Chongqing Megalight Technology co. LTD, ChinaView Profile , Bohui Hao Chongqing …

Semi-automatic construction method of power safety ontology based on AR-K-means

2021 · 2021 IEEE International Conference on Power Electronics, Computer Applications (ICPECA)

In terms of data modeling during the construction of the power safety knowledge map, the traditional manual method of constructing the power safety ontology has the problem of time-consuming and labor-intensive. Therefore, a semi-automatic construction …

Training Networks in Null Space of Feature Covariance for Continual Learning

2021 · arXiv (Cornell University)

In the setting of continual learning, a network is trained on a sequence of tasks, and suffers from catastrophic forgetting. To balance plasticity and stability of network in continual learning, in this paper, we propose …

Boosting Black-Box Adversarial Attacks with Meta Learning

2022 · arXiv (Cornell University)

Deep neural networks (DNNs) have achieved remarkable success in diverse fields. However, it has been demonstrated that DNNs are very vulnerable to adversarial examples even in black-box settings. A large number of black-box attack methods …

Research on Evaluating Teaching Quality of Accounting Courses Based on Improved K-nearest Neighbors Algorithm(KNN)

2023

Considering the differences in clustering positions between classifiable attribute data with different occurrence times, we redefined the position of the clustering center and proposed an improved KNN clustering method. The proposed method effectively solved the …