Qi Li

TL;DR: This paper formulate causality extraction as a sequence labeling problem based on a novel causality tagging scheme, and proposes a neural causality extractor with the BiLSTM-CRF model as the backbone, named SCITE (Self-attentive BiL STM- CRF wIth Transferred Embeddings), which can directly extract cause and effect without extracting candidate causal pairs and identifying their relations separately.

TL;DR: This article presents a client-based system framework for location privacy protection in LBS, and introduces a location privacy model to formulate the constraints that ideal cover-up ranges should satisfy, so as to improve the efficiency of location services and the security of location privacy.

TL;DR: A new pre-trained network architecture for this task, called the XM-CNN, which utilizes word embedding and position embedding information and is designed to reinforce the contextual output from the MT-DNN K D pre- trained model.

TL;DR: A novel knowledge-attention encoder which incorporates prior knowledge from external lexical resources into deep neural networks for relation extraction task and three effective ways of integrating knowledge-Attention with self-att attention to maximize the utilization of both knowledge and data are presented.

TL;DR: Experiments show that the multi-objective adaptive fast evolutionary algorithm greatly improves the accuracy of community detection in complex networks, and can discover the hierarchical structure of complex networks better.