论文信息 - DeepEncyclolink: A Cross-Encyclopedia, Cross-language Article-Linking System Based on Deep Learning

DeepEncyclolink: A Cross-Encyclopedia, Cross-language Article-Linking System Based on Deep Learning

DeepEncyclolink is a web-based system for linking pairs of corresponding articles from different encyclopedias in different languages. The core technology of DeepEncyclolink is based on paragraph embeddings calculated by a long-short-term-memory network with attention. Compared to our previous feature-based machine learning model, DeepEncyclolink has made great strides in coverage and accuracy. DeepEncyclolink offers an easy-to-use user interface and is supported by a powerful service backend for retrieving and selecting equivalent articles between English Wikipedia and Chinese Baidu Encyclopedia. DeepEncyclolink breaks the barriers of different encyclopedias and different languages, making it easy for users to access knowledge written in various languages.

Yu-Chun Wang | Richard Tzong-Han Tsai | Chun-Kai Wu | Ka Ming Wong | Chia-Min Chuang

[1] Yu-Chun Wang,et al. Cross-language and Cross-encyclopedia Article Linking Using Mixed-language Topic Model and Hypernym Translation , 2014, ACL.

[2] Yoon Kim,et al. Convolutional Neural Networks for Sentence Classification , 2014, EMNLP.

[3] Xiang Zhang,et al. Character-level Convolutional Networks for Text Classification , 2015, NIPS.

[4] Stephen E. Robertson,et al. Okapi at TREC-5 , 1996, TREC.

[5] Ting Liu,et al. Document Modeling with Gated Recurrent Neural Network for Sentiment Classification , 2015, EMNLP.

[6] Alessandro Moschitti,et al. Learning to Rank Short Text Pairs with Convolutional Deep Neural Networks , 2015, SIGIR.