检索结果-内蒙古大学图书馆

arXiv 2020年

作者： Haonan, Wang Yang, Gao Yu, Bai Lapata, Mirella Heyan, Huang School of Computer Science and Technology Beijing Institute of Technology Beijing100081 China Institute for Language Cognition and Computation School of Informatics University of Edinburgh 10 Crichton Street EdinburghEH8 9AB United Kingdom Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications

It is a big challenge to model long-range input for document summarization. In this paper, we target using a select and generate paradigm to enhance the capability of selecting explainable contents (i.e., interpret the selection given its semantics, novelty, relevance) and then guiding to control the abstract generation. Specifically, a new designed pair-wise extractor is proposed to capture the sentence pair interactions and its centrality. Furthermore, the generator is hybrid with the selected content and is jointly integrated with a pointer distribution that is derived from a sentence deployment's attention. The abstract generation can be controlled by an explainable mask matrix that determines to what extent the content can be included in the summary. Encoders are adaptable with both Transformerbased and BERT-based configurations. Overall, both results based on ROUGE metrics and human evaluation gain outperformance over several state-of-The-Art models on two benchmark CNN/DailyMail and NYT datasets. Copyright © 2020, The Authors. All rights reserved.

关键词： Abstracting

来源：评论

学校读者我要写书评

暂无评论

Chinese Parsing Model Based on Constraint Dependency Grammar

Chinese Parsing Model Based on Constraint Dependency Grammar

引用

International Conference on Consumer Electronics, Communications and Networks

作者： Peng Li Lejian Liao Beijing Engineering Research Centre of High Volume Language Information Processing & Cloud Computing Applications School of Computer Science Beijing Institute of Technology

ISBN: (纸本)9781457714146

Constraint Dependency Grammar (CDG) is a formalism representing grammatical rules as constraints on dependency relations. It reduces the structural ambiguity via constraint-propagation without generating individual parse trees. In this paper, we propose a high quality parsing model for Chinese, which defines the phrase molecule lexicalizing the CDG at the word-level. Ternary constraints are used to improve the accuracy and efficiency of parsing.

关键词： Constraint Dependency Grammar Parsing Chinese

来源：评论

学校读者我要写书评

暂无评论

Hierarchical attention network for visually-aware food recommendation

arXiv

引用

arXiv 2018年

作者： Gao, Xiaoyan Feng, Fuli He, Xiangnan Huang, Heyan Guan, Xinyu Feng, Chong Ming, Zhaoyan Chua, Tat-Seng Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications School of Computer Beijing Institute of Technology Beijing100081 China School of Computing National University of Singapore Singapore117417 Singapore University of Science and Technology of China Hefei Anhui230031 China Systems Engineering Institute Xi'an Jiaotong University Xi'an710049 China

Food recommender systems play an important role in assisting users to identify the desired food to eat. Deciding what food to eat is a complex and multi-faceted process, which is influenced by many factors such as the ingredients, appearance of the recipe, the user's personal preference on food, and various contexts like what had been eaten in the past meals. In this work, we formulate the food recommendation problem as predicting user preference on recipes based on three key factors that determine a user's choice on food, namely, 1) the user's (and other users') history;2) the ingredients of a recipe;and 3) the descriptive image of a recipe. To address this challenging problem, we develop a dedicated neural network-based solution Hierarchical Attention based Food Recommendation (HAFR) which is capable of: 1) capturing the collaborative filtering effect like what similar users tend to eat;2) inferring a user's preference at the ingredient level;and 3) learning user preference from the recipe's visual images. To evaluate our proposed method, we construct a large-scale dataset consisting of millions of ratings from ***. Extensive experiments show that our method outperforms several competing recommender solutions like Factorization Machine and Visual Bayesian Personalized Ranking with an average improvement of 12%, offering promising results in predicting user preference on food. Codes and dataset will be released upon acceptance. Copyright © 2018, The Authors. All rights reserved.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Latency-aware routing with bandwidth assignment for Software Defined Networks

Latency-aware routing with bandwidth assignment for Software...

引用

IEEE International Conference on Performance, computing and Communications (IPCCC)

作者： Qiongyu Zhang Liehuang Zhu Meng Shen Mingzhong Wang Fan Li Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications Beijing Institute of Technology P. R. China Key Laboratory of Computer Network and Information Integration (Southeast University) Ministry of Education P. R. China Faculty of Arts and Business University of the Sunshine Coast Queensland Australia

Reducing the flow latency is of great importance in traffic management, which benefits both service providers and end users. Routing design and flow scheduling are typical ways to improve the flow transmission efficiency. However, existing studies usually consider them separately, due to the complexity of joint consideration. Here, we propose a latency-aware routing scheme with bandwidth assignment in the Software Defined Networks, which can efficiently reduce the flow latency with a moderate complexity, to combine the routing and scheduling together.

关键词： Routing Bandwidth Processor scheduling Complexity theory Bismuth Software Mathematical model

来源：评论

学校读者我要写书评

暂无评论

Self-adaptive anonymous communication scheme under SDN architecture

Self-adaptive anonymous communication scheme under SDN archi...

引用

IEEE International Conference on Performance, computing and Communications (IPCCC)

作者： Tingting Zeng Meng Shen Mingzhong Wang Liehuang Zhu Fan Li Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications Beijing Institute of Technology P. R. China Key Laboratory of Computer Network and Information Integration (Southeast University) Ministry of Education P. R. China Faculty of Arts and Business University of the Sunshine Coast Queensland Australia

Communication privacy and latency perceived by users have become great concerns for delay-sensitive Internet services. Existing anonymous communication systems either provide high anonymity at an expense of prolonged latency (e.g., mix-net), or offer better real-time performance by sacrificing the ability against traffic analysis attacks (e.g., Onion Routing). The emerging Software-Defined Networking (SDN) introduces additional challenges to communication anonymity, due to the existence of a centralized controller that has a global view of the entire network traffic. In this paper, we propose a new anonymous communication scheme for delay-sensitive services under SDN scenarios, which can simultaneously protect communication privacy and reduce the end-to-end latency. A self-adaptive method based on the mix-net framework is designed to dynamically modify the waiting threshold of mix nodes, which helps to reduce the communication latency. In order to preserve the degree of anonymity, the self-adaptive method is incorporated with a random walking strategy for packets forwarding. Both theoretical analysis and experimental results prove that our scheme provides a moderate degree of anonymity and effectively reduces the latency derived from mix-net by up to 50%.

关键词： Real-time systems Relays Entropy Lead Distance measurement Annealing Probabilistic logic

来源：评论

学校读者我要写书评

暂无评论

Group Daily Arrival Prediction Based on A Hybrid Model Combing Autoregressive Integrated Moving Average Model and Back Propagation Neural Network

Group Daily Arrival Prediction Based on A Hybrid Model Combi...

引用

IEEE International Conference on Unmanned Systems (ICUS)

作者： Xiaoling Zhu Xi Zhang Ge Shi Yashen Wang National Engineering Laboratory for Risk Perception and Prevention (RPP) China Academy of Electronics and Information Technology of CETC Beijing China School of Cyberspace Security Beijing University of Posts and Telecommunications Beijing China Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications School of Computer Beijing Institute of Technology Beijing China

ISBN: (数字)9781728180250

ISBN: (纸本)9781728180267

To solve the problem of predicting daily arrival number of people belonging to a specific group, this paper proposes a hybrid model which integrates linear component (from ARIMA model) with non-linear component (from BP neural network). The experimental results indicate that, hybrid methodology that has both linear and nonlinear modeling capabilities is a good strategy for practical use.

关键词： Predictive models Neurons Data models Analytical models Urban areas Solid modeling Backpropagation

来源：评论

学校读者我要写书评

暂无评论

From qestion to text: Qestion-oriented feature atention for answer selection

引用

ACM Transactions on information Systems 2018年第1期37卷

作者： Huang, Heyan Wei, Xiaochi Nie, Liqiang Mao, Xianling Xu, Xin-Shun Beijing Eng. Res. Ctr. of High Vol. Language Information Processing and Cloud Computing Applications School of Computer Beijing Institute of Technology No.5 Zhongguancun South Street Haidian Beijing100081 China School of Computer Science Shandong University 72 Binhai Road Jimo Qingdao Shandong Province266237 China

Understanding unstructured texts is an essential skill for human beings as it enables knowledge acquisition. Although understanding unstructured texts is easy for we human beings with good education, it is a great challenge for machines. Recently, with the rapid development of artifcial intelligence techniques, researchers put efforts to teach machines to understand texts and justify the educated machines by letting them solve the questions upon the given unstructured texts, inspired by the reading comprehension test as we humans do. However, feature effectiveness with respect to different questions signifcantly hinders the performance of answer selection, because different questions may focus on various aspects of the given text and answer candidates. To solve this problem, we propose a question-oriented feature attention (QFA) mechanism, which learns to weight different engineering features according to the given question, so that important features with respect to the specifc question is emphasized accordingly. Experiments on MCTest dataset have wellvalidated the effectiveness of the proposed method. Additionally, the proposed QFA is applicable to various IR tasks, such as question answering and answer selection. We have verifed the applicability on a crawled community-based question-answering dataset. © 2018 ACM.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Improving Naturalness and Controllability of Sequence-to-Sequence Speech Synthesis by Learning Local Prosody Representations

Improving Naturalness and Controllability of Sequence-to-Seq...

引用

IEEE International Conference on Acoustics, Speech and Signal processing

作者： Cheng Gong Longbiao Wang Zhenhua Ling Shaotong Guo Ju Zhang Jianwu Dang Tianjin Key Laboratory of Cognitive Computing and Application College of Intelligence and Computing Tianjin University Tianjin P.R.China National Engineering Laboratory for Speech and Language Information Processing University of Science and Technology of China Hefei P.R.China Huiyan Technology (Tianjin) Co. Ltd Japan Advanced Institute of Science and Technology Ishikawa Japan

ISBN: (纸本)9781728176055;9781728176062

State-of-the-art neural text-to-speech (TTS) networks are trained with a large amount of speech data, which significantly improves the quality of synthetic speech compared with traditional approaches. However, the prosody and controllability of the generated speech is still insufficient, especially in tonal languages. Moreover, the generated prosody is solely defined by the input text, which does not allow for different styles for the same sentence or words. In this study, we extended Tacotron2 with a pitch prediction task to capture discrete pitch-related representations. Specifically, the learned pitch-related suprasegmental information is fed simultaneously with traditional character features into the decoder to generate final Mel spectrogram. Experiments show that the proposed method can improve the quality of the generated speech (mean opinion score of 4.37 vs. 4.22). Moreover, we demonstrated that we can easily achieve word-level pitch control during generation by changing local pitch-related representations before passing them to the decoder network.

关键词： Conferences Neural networks Controllability Acoustics Decoding Speech synthesis Task analysis

来源：评论

学校读者我要写书评

暂无评论

Unsupervised word sense disambiguation using neighborhood knowledge

PACLIC 25 - Proceedings of the 25th Pacific Asia Conference ...

引用

PACLIC 25 - Proceedings of the 25th Pacific Asia Conference on language, information and Computation 2011年 333-342页

作者： Heyan, Huang Zhizhuo, Yang Ping, Jian Beijing Engineering Applications Research Center of High Volume Language Information Processing and Cloud Computing Beijing Institute of Technology No.5 Yard Zhong Guan Cun South Street Haidian District Beijing 100081 China Department of Computer Science Beijing Institute of Technology No.5 Yard Zhong Guan Cun South Street Haidian District Beijing 100081 China

ISBN: (纸本)9784905166023

Usually ambiguous words contained in article appear several times. Almost all existing methods for unsupervised word sense disambiguation make use of information contained only in ambiguous sentence. This paper presents a novel approach by considering neighborhood knowledge. The approach can naturally make full use of the within-sentence relationship from the ambiguous sentence and cross-sentence relationship from the neighborhood knowledge. Experimental results indicate the proposed method can significantly outperform the baseline method. © 2011 by Huang Heyan, Yang Zhizhuo, Jian Ping.

关键词： information use

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Hashing with Semantic Concept Mining

arXiv

引用

arXiv 2022年

作者： Tu, Rong-Cheng Mao, Xian-Ling Lin, Kevin Qinghong Cai, Chengfei Qin, Weize Wang, Hongfa Wei, Wei Huang, Heyan School of Computer Science and Technology Beijing Institute of Technology Beijing China Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications Beijing China Beijing Institute of Technology Southeast Academy of Information Technology Fujian China National University of Singapore Singapore School of Computer Science Zhejiang University Zhejiang China Institute of Computing Technology Chinese Academy of Sciences Beijing China School of Computer Science Huazhong University of Science and Technology Wuhan China

Recently, to improve the unsupervised image retrieval performance, plenty of unsupervised hashing methods have been proposed by designing a semantic similarity matrix, which is based on the similarities between image features extracted by a pre-trained CNN model. However, most of these methods tend to ignore high-level abstract semantic concepts contained in images. Intuitively, concepts play an important role in calculating the similarity among images. In real-world scenarios, each image is associated with some concepts, and the similarity between two images will be larger if they share more identical concepts. Inspired by the above intuition, in this work, we propose a novel Unsupervised Hashing with Semantic Concept Mining, called UHSCM, which leverages a VLP model to construct a high-quality similarity matrix. Specifically, a set of randomly chosen concepts is first collected. Then, by employing a vision-language pretraining (VLP) model with the prompt engineering which has shown strong power in visual representation learning, the set of concepts is denoised according to the training images. Next, the proposed method UHSCM applies the VLP model with prompting again to mine the concept distribution of each image and construct a high-quality semantic similarity matrix based on the mined concept distributions. Finally, with the semantic similarity matrix as guiding information, a novel hashing loss with a modified contrastive loss based regularization item is proposed to optimize the hashing network. Extensive experiments on three benchmark datasets show that the proposed method outperforms the state-of-the-art baselines in the image retrieval task. The source codes are available 1 Copyright © 2022, The Authors. All rights reserved.

关键词： Image retrieval

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：