当前位置>主页 > 期刊在线 > 计算机技术 >

计算机技术21年19期

基于机器学习的馆内阅读行为模式分析
许智敏,施玲玲,肖珊,郭雁瑶,邹嘉书
(广州航海学院,广东 广州 510725)

摘  要:针对大量阅读者的阅读能力制约着其阅读效果的难题提出了一种基于机器学习的馆内行为模式分析方法。首先,研究了一种“宏 - 微”复合视觉成像方法及工作原理;其次,研究了基于图像特征的阅读行为模式识别算法;最后,将基于馆内目标图像获取的阅读行为模式与实际的阅读行为模式进行对比分析。并由实验验证该文提出的方法能够精准、稳定地识别馆内阅读者的阅读行为模式。


关键词:阅读行为模式;机器视觉;眼动跟踪;SVM



DOI:10.19850/j.cnki.2096-4706.2021.19.017


课题项目:广州市教育科学规划 2019 年度 课题(201911942)


中图分类号:TP18                                      文献标识码:A                                     文章编号:2096-4706(2021)19-0067-05


Analysis of Reading Behavior Patterns in the Library Based on Machine Learning

XU Zhimin, SHI Lingling, XIAO Shan, GUO Yanyao, ZOU Jiashu

(Guangzhou Maritime University, Guangzhou 510725, China)

Abstract: Aiming at the difficult problem that a large number of readers’ reading ability restricts their reading effect, a library behavior pattern analysis method based on machine learning is proposed. Firstly, a “macro-micro” composite vision imaging method and its working principle are studied; secondly, the reading behavior pattern recognition algorithm based on image features is studied; finally, the reading behavior patterns obtained based on the target images in the library are compared with the actual reading behavior patterns. Experiments show that the proposed method in this paper can accurately and stably identify the reading behavior patterns of readers in the library.

Keywords: reading behavior pattern; machine vision; eye tracking; SVM


参考文献:

[1] 董一凡 . 论当代读者的阅读方式与图书馆的对策 [J]. 农业图书情报学刊,2010,22(1):125-128.

[2] 张晓松、朱基钗.习近平:要提倡多读书,建设书香社会 [EB/ OL].(2019-08-22).https://baijiahao.baidu.com/s?id=16425306689806 71789&wfr=spider&for=pc.

[3] JOACHIMS T,GRANKA L,PAN B,et al. Accurately interpreting clickthrough data as implicit feedback [C]//SIGIR05:The 28th ACM/SIGIR International Symposium on Information Retrieval 2005. Salvador:Association for Computing Machinery,2005:154-161.

[4] JOACHIMS T,GRANKA L,PAN B,et al. Evaluating the accuracy of implicit feedback from clicks and query reformulations in Web search [J/OL].Acm Transactions on Information Systems,2007, 25(2):[2021-09-10].https://doi.org/10.1145/1229179.1229181.

[5] 闫国利,熊建萍,臧传丽,等 . 阅读研究中的主要眼动指 标评述 [J]. 心理科学进展,2013,21(4):589-605.

[6] 王先梅,杨萍,王志良 . 多姿态眼球中的瞳孔定位算法 [J]. 计算机辅助设计与图形学学报,2011,23(8):1427-1432.

[7] 刘晓龙 . 基于图像的行人检测算法研究 [D]. 长沙:国防科学技术大学,2017.

[8] 吴迪 . 基于改进卷积神经网络的行人检测及再识别方法研究 [D]. 秦皇岛:燕山大学,2019.

[9] 谢林江,季桂树,彭清,等 . 改进的卷积神经网络在行人检测中的应用 [J]. 计算机科学与探索,2018,12(5):708-718.

[10] RAYNER K.Eye movements and attention in reading,scene perception,and visual search [J].The Quarterly Journal of Experimental Psychology,2009,62(8):1457-1506.

[11] 许洁,王豪龙 . 阅读行为眼动跟踪研究综述 [J]. 出版科学,2020,28(2):52-66.

[12] BILAL M,HANIF M S. Benchmark Revision for HOGSVM Pedestrian Detector Through Reinvigorated Training and Evaluation Methodologies [J].IEEE Transactions on Intelligent Transportation Systems,2020,21(3):1277-1278.

[13] FANG F,LI L Y,ZHU H Y,et al. Combining Faster R-CNN and Model-Driven Clustering for Elongated Object Detection [J].IEEE Transactions on Image Processing,2019,29:2052-2065.

[14] DALAL N,TRIGGS B. Histograms of oriented gradients for human detection [C]//2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR’05).San Diego: IEEE,2005:886-893.

[15] ZEILER M D,FERGUS R. Visualizing and Understanding Convolutional Networks [J]//Computer Vision–ECCV 2014.Zurich: Springer,2014,8689:818-833.

[16] SIMONYAN K,ZISSERMAN A. Very Deep Convolutional Networks for Large-Scale Image Recognition [J/OL].arXiv:1409.1556 [cs.CV].[2021-09-11].https://arxiv.org/abs/1409.1556.


作者简介:许智敏(1988—),女,汉族,广东广州人,图书馆馆员,硕士研究生,主要研究方向:阅读素养分析、数据挖掘; 施玲玲(2000—),女,汉族,广东陆丰人,本科在读,主要研究方向: 机器视觉;肖珊(1999—),女,汉族,福建龙岩人,本科在读, 主要研究方向:机器人技术、机器视觉技术;郭雁瑶(2001—),女, 汉族,广东陆丰人,本科在读,主要研究方向:机器视觉、深度学习; 邹嘉书(2001—),男,汉族,广东梅州人,本科在读,主要研究方向:机械设计。