前言

​ 3.14 听完师兄师姐汇报导师觉得我们听得一头雾水，于是给了我们看期刊整理论文的任务😐……

​ 要求我们四个人整理完《中国计算机学会推荐国际学术会议和期刊目录》人工智能部分期刊和会议近三年的目录并从中总结出研究趋势。

​ 这篇博客就胡扯点完成论文干的各种事吧。

正文

##《中国计算机学会推荐国际学术会议和期刊目录》

{% pdf pdf.pdf%}

爬虫

​ 分完工后还是觉得要整理的论文特别多，于是整了个爬虫代码帮忙爬目录。

• zotero_dblp_bibtex: 通过 python 抓取 dblp 中某一作者、某年会议或期刊的论文 (gitee.com)

​ 然后估计爬的太多了，爬久了就爬不动了 orz 忙活了几乎一天才爬到了一半内容 orz

​ 然后就发现 dblp 这个网站本身就可以批量下载目录 orz 除了超过 1000 的和期刊数太多的期刊用爬虫可能更快些。

导出 Excel

​ 将得到的 bibtex 文件一个一个地导入进 Zotero，然后一个一个地导出成 csv 文件。

​ 接下来就是合并表格了，手动操作了几次后觉得太麻烦，整了个 python 代码帮我自动整理。

import os
import pandas as pd
 
path = r'D:\Study\1st-year-master\论文目录\Conf\A\NeurlPS'
save_file = path.split('\\')[-1]
 
data = pd.DataFrame(columns=['Publication Year', 'Author', 'Title'])
 
for index, file in enumerate(os.listdir(path)):
    if('.csv' in file):
        df = pd.read_csv(os.path.join(path,file))
        data = data.append(df[['Publication Year', 'Author', 'Title']])
 
data.head
write = pd.ExcelWriter(os.path.join(path,save_file) + '.xlsx')   # 新建 xlsx 文件。
data.to_excel(write, sheet_name='Sheet', index=False)  # 写入文件的 Sheet1
write.save()  # 这里一定要保存

​ 这样每个期刊/会议近三年的目录就被整成一个 Excel 表格了😎！

统计词频

​ 想用 python 把这些目录的词频统计出来，这样就分析当作研究趋势了😈！于是又一阵操作猛如虎：

import os
import pandas as pd
import re
 
path = r'D:\Study\1st-year-master\论文目录\xlsx'
save_path = r'D:\Study\1st-year-master\论文目录\freq'
 
for index, file in enumerate(os.listdir(path)):
    if '.xlsx' in file:
        df = pd.read_excel(os.path.join(path,file))
        series = df['Title'].str.split(' ').explode().value_counts()
        data = pd.DataFrame({'Word': series.index, 'Frequent': series.values})
        write = pd.ExcelWriter(os.path.join(save_path, file) + '_frequent.xlsx')   # 新建 xlsx 文件。
        data.to_excel(write, sheet_name='Sheet' + str(index), index=False)  # 写入文件的 Sheet1
        write.save()  # 这里一定要保存

​ 结果统计出来的全是 “a”、“of”、“from” 之类的，一点参考意义也没有😅……

​ 杰杰还搞了一个词频可视化，果然也都是一些没啥用的单词……不过突然感慨他的技术真的好全面……几乎什么都会一点。

翻译

​ 统计词频的计划失败后，想想还是手动检索论文吧。英文看的实在太难受，想了想有没有什么可以翻译 Excel 表格的方法……一开始想的很复杂，想先转成 markdown 文件然后传到博客上用浏览器帮忙翻译，在 Excel 表格调用有道的 API 直接翻译结果发现请求次数一上来就不让翻译了 balababala……最后伟哥说谷歌翻译可以直接上转 Excel 表格返回中文文档 orz

• Google 翻译

​ 最后就得到了中文的论文标题目录！不知道为什么有些英文没有进行翻译。想着文件不多就手动把中英文标题整成一个表格了。

总结

​ 最后大概看了一个晚上 + 一个上午吧……几万篇论文眼睛都快看花了💨，由于是主观判断研究趋势，感觉总结得也不是特别准，然后筛了点自己跟自己研究方向有点关系的论文。

最终结果

Jounrnals

A

IEEE Transactions on Pattern Analysis and Machine Intelligence

人体姿态估计、人体检测、3D 视觉、CNN、图像修复、点云、GAN、超分辨率、图像分割、视频

• Baselines extraction from curved document images via slope fields recovery
• Mask TextSpotter: An end-to-end trainable neural network for spotting text with arbitrary shapes
• Shape-matching GAN++: Scale controllable dynamic artistic text style transfer
• Unambiguous text localization, retrieval, and recognition for cluttered scenes
• Content and style aware generation of text-line images for handwriting recognition
• Towards end-to-end text spotting in natural scenes
• Deep generative modelling: A comparative review of VAEs, GANs, normalizing flows, energy-based and autoregressive models
• A geometrical perspective on image style transfer with adversarial learning
• GAN inversion: A survey
• Arbitrary shape text detection via segmentation with probability maps
• End-to-end handwritten paragraph text recognition using a vertical attention network
• Real-time scene text detection with differentiable binarization and adaptive scale fusion
• A survey on vision transformer

B

Data & Knowledge Engineering

建模、数据、解决某些实际问题

IEEE Transactions on Cybernetics

算法、信息安全、传统机器学习方法（马尔科夫链、支持向量机）

International Journal of Approximate Reasoning

模糊逻辑

Machine Learning

传统机器学习、强化学习

Transactions of the Association for Computational Linguistics

自然语言处理、语言学、BERT、Transformer

C

Applied Intelligence

解决某些实际问题、超分辨率、模式识别、医学、COVID-19（预测、检测）、3D 视觉、CNN、图神经网络、聚类、分类、情感识别

• Anchor-free multi-orientation text detection in natural scene images

Computer Speech & Language

语音识别、自然语言处理

Expert Systems

解决某些实际问题、医学、COVID-19、推荐系统

IET Computer Vision

人体姿态估计、人脸识别、目标检测、CNN、3D 视觉

• Stroke controllable style transfer based on dilated convolutions
• KText: Arbitrary shape text detection using modified K-Means

International Journal of Computational Intelligence and Applications

解决某些实际问题

International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems

模糊逻辑

Machine Translation

机器翻译

Neural Computing and Applications

解决某些实际问题

• Hybrid HMM/BLSTM system for multi-script keyword spotting in printed and handwritten documents with identification stage
• SPN: short path network for scene text detection
• Robustly detect different types of text in videos
• Effective offline handwritten text recognition model based on a sequence-to-sequence approach with CNN-RNN networks
• Historical document image binarization via style augmentation and atrous convolutions
• Character-based handwritten text transcription with attention networks

Pattern Recognition Letters

模式识别（文字、医学、人体）、特征提取、GAN

• Compressing the CNN architecture for in-air handwritten Chinese character recognition
• LPR-Net: Recognizing Chinese license plate in complex environments
• An attention-based row-column encoder-decoder model for text recognition in Japanese historical documents
• Assessing similarity in handwritten texts
• Clustering online handwritten mathematical expressions
• Beyond visual semantics: Exploring the role of scene text in image understanding
• PMMN: Pre-trained multi-Modal network for scene text recognition
• RectiNet-v2: A stacked network architecture for document image dewarping
• Transformer-based approach for joint handwriting and named entity recognition in historical document
• Cross lingual handwritten character recognition using long short term memory network with aid of elephant herding optimization algorithm
• CE-text: A context-Aware and embedded text detector in natural scene images

Conf

A

Conference on Neural Information Processing Systems

数据集、图神经网络、算法、GAN、元学习、对比学习、强化学习、3D、Transformer

• OmniPrint: A configurable printed character synthesizer
• Are transformers more robust than CNNs?
• Diffusion models beat GANs on image synthesis

International Conference on Machine Learning

模型攻击、算法、游戏、生成模型、元学习、强化学习、Transformer、Wasserstein 距离、强化学习

B

Conference on Empirical Methods in Natural Language Processing

NLP、BERT、强化学习、文章生成摘要、文本生成、迁移学习、Transformer、数据集、COVID-19

• Cleaning dirty books: Post-OCR processing for previously scanned texts

International Conference on Automated Planning and Scheduling

路径规划、强化学习

Conference on Uncertainty in Artificial Intelligence

马尔科夫链、贝叶斯、无监督学习、蒙特卡洛、强化学习

C

Asian Conference on Computer Vision

贝叶斯、马尔可夫、3D 视觉、人体识别、GAN

• Accurate arbitrary-shaped scene text detection via iterative polynomial parameter regression

Conference on Computational Natural Language Learning

NLP

International Conference on Algorithmic Learning Theory

在线学习、强化学习

International Conference on Inductive Logic Programming

太杂太少太偏了...

International Joint Conference on Biometrics

安全、攻击、人脸识别

领域有关的论文