作者：
Stanley R. M. OliveiraOsmar R. ZaianeEmbrapa Information Technology
Andre Tosello 209 - Barao Geraldo 13083-886 - Campinas SP Brasil and Department of Computing Science University of Alberta Edmonton AB Canada T6G 2E8 s in Electronics from the University ofParis XI
France. He has worked in a variety of research areas such as data mining web mining multimedia databases information retrieval web technology natural language processing distance education and collab
Recent data mining algorithms have been designed for application domains that involve several types of objects stored in multiple relations in relational databases. This fact has motivated the increasing number of suc...
详细信息
ISBN:
(纸本)9780909925925
Recent data mining algorithms have been designed for application domains that involve several types of objects stored in multiple relations in relational databases. This fact has motivated the increasing number of successful applications of relational data mining over recent years. On the other hand, such applications have introduced a new threat to privacy and information security since from non-sensitive data one is able to infer sensitive information, including personal information, facts or even patterns that are not supposed to be disclosed. The existing access control models adopted to successfully manage the access of information in complex systems present some limitations in the context of data mining tasks. The main reason is that such models were designed to protect the access to explicit data (e.g. tables, attributes, views, etc), whereas data mining tasks deal with the discovery of implicit data (e.g. patterns). In this paper, we take a first step toward an access control model for ensuring privacy in relational data mining, notably in multi-relational association rules (MRAR). In this model, users associated with different mining access levels, even using the same algorithm, are allowed to mine different sets of association rules. We provide the groundwork to build our access control model over existing technologies and discuss some directions for future work.
A high-speed, floating-point array processor 150AP is described in this paper. It is designed to interface to the host computer 150, a general-purpose computer made in China. The internal organization of 150AP is well...
详细信息
A Workshop in Two Year College Programs in Computer Science was held August 11-13, 1975 in Gloucester Point, Virginia. Sponsored by the ACM's Special Interest Group in Computer Science Education (SIGCSE), it broug...
详细信息
ISBN:
(纸本)9781450374125
A Workshop in Two Year College Programs in Computer Science was held August 11-13, 1975 in Gloucester Point, Virginia. Sponsored by the ACM's Special Interest Group in Computer Science Education (SIGCSE), it brought thirteen community and junior college participants together at the Virginia Institute for Marine Science (VIMS) to work toward recommendations for a two year program.
Named Entity Recognition (NER) is an important task in knowledge extraction, which targets extracting structural information from unstructured text. To fully employ the prior-knowledge of the pre-trained language mode...
详细信息
Named Entity Recognition (NER) is an important task in knowledge extraction, which targets extracting structural information from unstructured text. To fully employ the prior-knowledge of the pre-trained language models, some research works formulate the NER task into the machine reading comprehension form (MRC-form) to enhance their model generalization capability of commonsense knowledge. However, this transformation still faces the data-hungry issue with limited training data for the specific NER tasks. To address the low-resource issue in NER, we introduce a method named active multi-task-based NER (AMT-NER), which is a two-stage multi-task active learning training model. Specifically, A multi-task learning module is first introduced into AMT-NER to improve its representation capability in low-resource NER tasks. Then, a two-stage training strategy is proposed to optimize AMT-NER multi-task learning. An associated task of Natural Language Inference (NLI) is also employed to enhance its commonsense knowledge further. More importantly, AMT-NER introduces an active learning module, uncertainty selective, to actively filter training data to help the NER model learn efficiently. Besides, we also find different external supportive data under different pipelines improves model performance differently in the NER tasks. Extensive experiments are performed to show the superiority of our method, which also proves our findings that the introduction of external knowledge is significant and effective in the MRC-form NER tasks.
暂无评论