Discriminative feature mining with relation regularization for person re-identification     

《Information processing & management》2023,60(3):103295

The appearance attribute and pose are two important and complementary features, so integrating them can effectively alleviate the impact of misalignment and occlusion on re-identification. In this paper, we deeply investigate the inner relation between attribute features and the spatial semantic relation between key-point region features of the pose in a person image and propose a person re-identification method based on discriminative feature mining with relation regularization. Firstly, an attribute relation detector based on nonlinear graph convolution is built on mining the inner correlation between attribute features of a person, providing relational attribute features for more effectively distinguishing persons with a similar appearance. Then, we construct a hierarchical pose pyramid to model the multi-grained semantic features of key-point regions of the pose and propose intra-graph and cross-graph node relation information propagation structures to infer the spatial semantic relation between node features within-graph and between-graph. This module is robust to complex pose changes and can suppress noise background redundancy caused by inaccurate key point detection and occlusion. Finally, a refined feature model is proposed to effectively fuse the global appearance feature with the relational attribute and multi-grained pose features, thus providing a more discriminative fusion feature for person re-identification. Many experiments on three large-scale datasets verify the effectiveness and state-of-the-art performance of the proposed method.  相似文献   

RGAN-EL: A GAN and ensemble learning-based hybrid approach for imbalanced data classification     

《Information processing & management》2023,60(2):103235

Imbalanced sample distribution is usually the main reason for the performance degradation of machine learning algorithms. Based on this, this study proposes a hybrid framework (RGAN-EL) combining generative adversarial networks and ensemble learning method to improve the classification performance of imbalanced data. Firstly, we propose a training sample selection strategy based on roulette wheel selection method to make GAN pay more attention to the class overlapping area when fitting the sample distribution. Secondly, we design two kinds of generator training loss, and propose a noise sample filtering method to improve the quality of generated samples. Then, minority class samples are oversampled using the improved RGAN to obtain a balanced training sample set. Finally, combined with the ensemble learning strategy, the final training and prediction are carried out. We conducted experiments on 41 real imbalanced data sets using two evaluation indexes: F1-score and AUC. Specifically, we compare RGAN-EL with six typical ensemble learning; RGAN is compared with three typical GAN models. The experimental results show that RGAN-EL is significantly better than the other six ensemble learning methods, and RGAN is greatly improved compared with three classical GAN models.  相似文献   

Creating value from Social Big Data: Implications for Smart Tourism Destinations     

Pasquale Del Vecchio  Gioconda Mele  Valentina Ndou  Giustina Secundo 《Information processing & management》2018,54(5):847-860

This paper aims to demonstrate how the huge amount of Social Big Data available from tourists can nurture the value creation process for a Smart Tourism Destination. Applying a multiple-case study analysis, the paper explores a set of regional tourist experiences related to a Southern European region and destination, to derive patterns and opportunities of value creation generated by Big Data in tourism. Findings present and discuss evidence in terms of improving decision-making, creating marketing strategies with more personalized offerings, transparency and trust in dialogue with customers and stakeholders, and emergence of new business models. Finally, implications are presented for researchers and practitioners interested in the managerial exploitation of Big Data in the context of information-intensive industries and mainly in Tourism.  相似文献   

Remaining in Control with an Illusion of Interactivity: The Paternalistic Side of Data Journalism*     

Ester Appelgren 《Journalism Practice》2019,13(8):956-960

  相似文献   

Structural Analysis for Replicating Japanese Cultural Property Braids     

Masumi Tada  Kazuaki Yamashiro  Iwataro Oka  Akihiko Goto  Noriaki Kuwahara 《文物保护研究》2020,65(7):399-410

ABSTRACT

Historic Japanese textiles from over 1000 years ago generally show marked deterioration and only very rare examples show their original forms and much information about textile reproduction has been lost. The replication of textile braids lacks systematic methodology and is still being practiced by only few individual braiding experts. The recreation of original braids as close as possible to original braids is a part of Japan’s intangible cultural heritage. The aim of this study is to clarify the decision-making procedure through which the braiding experts can decipher the original braiding structures. As a preliminary study of this project, interviews of a braid researcher and a replicating expert, Makiko Tada were performed regarding her working practices. It is important to clarify the braiding parameters for structural analysis such as the number of transits and the balance of ridges, and it became clear that the orientation of multiple colored threads plays an important role. The expert’s replicate works were also analyzed using a text-mining statistical technique to clarify the relationship of braiding parameters. The relationship between each braiding parameter and production method such as loop manipulation and Taka-dai became clear. As a result, the process of deciphering the original braid structure has been compiled in simplified workflows, which could contribute to the standardization and improvement in efficiency of replication of cultural property braids.  相似文献   

基于深度学习的SSD模型尾矿库自动提取     

闫凯  沈汀  陈正超  闫弘轩 《中国科学院大学学报》2020,37(3):360-367

针对华北地区尾矿库自动提取问题，将基于深度学习的SSD目标检测模型应用于遥感图像尾矿库提取。首先标记华北地区2 000个样本，随机挑选1 500个作为训练样本，剩余样本作为测试样本，验证模型的检测精度。分析卷积层对应感受野与图像中尾矿库尺寸关系，发现原始SSD模型漏检误检大型尾矿库。改进SSD模型结构，提出增加额外卷积层的策略，提高对大型尾矿库目标的检测精度。实验表明，在置信度阈值为0.3时，改进的SSD模型相比原始模型，检测精确率提高10.0%，召回率提高14.4%，提高了大型尾矿库检测精度。验证了基于深度学习的SSD目标检测模型自动提取尾矿库的可行性以及改进算法的有效性。  相似文献   

MARC to BIBFRAME: Converting the PCC to Linked Data     

Jacquie Samples  Ian Bigelow 《Cataloging & classification quarterly》2020,58(3-4):403-417

Abstract

The Program for Cooperative Cataloging (PCC) has formal relationships with the Library of Congress (LC), Share-VDE, and Linked Data for Production Phase 2 (LD4P2) for work on Bibliographic Framework (BIBFRAME), and PCC institutions have been very active in the exploration of MARC to BIBFRAME conversion processes. This article will review the involvement of PCC in the development of BIBFRAME and examine the work of LC, Share-VDE, and LD4P2 on MARC to BIBFRAME conversion. It will conclude with a discussion of areas for further exploration by the PCC leading up to the creation of PCC conversion specifications and PCC BIBFRAME data.  相似文献   

Using an interactive platform to recognize the intersection of social and spatial inequalities     

Lisa Poling  Travis Weiland 《Teaching Statistics》2020,42(3):108-116

With the creation of interactive tasks that allow students to explore spatial ways of knowing in conjunction with their other ways of knowing the world, we create a space where students can make sense of information as they organize these new ideas into their already existing schema. Through the use of a Common Online Data Analysis Platform (CODAP) and data from Public Use Microdata Areas (PUMA), students can explore the communities in which they live and work, critically examining opportunities and challenges within a defined space.  相似文献   

Cross-company customer churn prediction in telecommunication: A comparison of data transformation methods     

《International Journal of Information Management》2019

Cross-Company Churn Prediction (CCCP) is a domain of research where one company (target) is lacking enough data and can use data from another company (source) to predict customer churn successfully. To support CCCP, the cross-company data is usually transformed to a set of similar normal distribution of target company data prior to building a CCCP model. However, it is still unclear which data transformation method is most effective in CCCP. Also, the impact of data transformation methods on CCCP model performance using different classifiers have not been comprehensively explored in the telecommunication sector. In this study, we devised a model for CCCP using data transformation methods (i.e., log, z-score, rank and box-cox) and presented not only an extensive comparison to validate the impact of these transformation methods in CCCP, but also evaluated the performance of underlying baseline classifiers (i.e., Naive Bayes (NB), K-Nearest Neighbour (KNN), Gradient Boosted Tree (GBT), Single Rule Induction (SRI) and Deep learner Neural net (DP)) for customer churn prediction in telecommunication sector using the above mentioned data transformation methods. We performed experiments on publicly available datasets related to the telecommunication sector. The results demonstrated that most of the data transformation methods (e.g., log, rank, and box-cox) improve the performance of CCCP significantly. However, the Z-Score data transformation method could not achieve better results as compared to the rest of the data transformation methods in this study. Moreover, it is also investigated that the CCCP model based on NB outperform on transformed data and DP, KNN and GBT performed on the average, while SRI classifier did not show significant results in term of the commonly used evaluation measures (i.e., probability of detection, probability of false alarm, area under the curve and g-mean).  相似文献   

基于深度学习的中文专利自动分类方法研究   总被引：2，自引：0，他引：2  

吕璐成  韩涛  周健  赵亚娟 《图书情报工作》2020,64(10):75-85

[目的/意义] 面向当前国内专利审查和专利情报分析工作中对于海量专利分类的客观需求，设计了7种基于深度学习的专利自动分类方法，对比各种方法的分类效果，从而助力专利分类效率和效果的提升。[方法/过程] 针对传统机器学习方法存在的缺陷，基于Word2Vec、CNN、RNN、Attention机制等深度学习技术，考虑专利文本语序特征、上下文特征以及分类关键特征，设计Word2Vec+TextCNN、Word2Vec+GRU、Word2Vec+BiGRU、Word2Vec+BiGRU+TextCNN等7种深度学习模型，以中国专利为例，选取IPC主分类号的"部"作为分类依据，对比这7种模型与3种传统分类模型在中文专利分类任务中的效果。[结果/结论] 实证研究效果显示，采用考虑语序特征、上下文特征及强化关键特征的深度学习方法进行中文专利分类具有更优的分类效果。  相似文献