# graduation_design
毕业设计——基于监督学习的web入侵检测系统（0day收集器）
##机器学习方面
用的技术比较low，只是scikit-learn里面的svm用了一下，开始的时候用knn来着，当现在样本数量过万的时候发现knn的确吃力。
##数据方面
发现陈讲的“在大公司里”百分之八十的时间都在洗数据，卧槽我现在简直是花了百分之九十的时间收集数据，数据清洗的脚本已经写好，数据收集的确麻烦。
SQL注入样本还好，正常请求是火狐插件代理日志中正则匹配出来的。昨晚各种网页逛了搞了很久，去重后搞出来一千多条……
基于监督学习的web入侵检测系统源码.zip （60个子文件） 
code 
reptile.py 678B justTest.ipynb 1KB makeChar.py 195B wordProcess.py 4KB removeRepeat.py 748B src 
ff_11281116.html 303KB
ff_11261946.txt 33KB access_log_payload2.txt 172KB ff_temp.txt 14KB ff_12010007.html 345KB key_list.txt 977B ff_11281116.txt 26.95MB ff_12052221.html 428KB payload.txt 768KB ff_11302308.html 341KB normal_require.txt 372KB access_log_payload0.txt 10KB access_log_payload4.txt 1.43MB ff_1201625.html 320KB access_log_payload1.txt 26.69MB ff_11302337.html 558KB ff_12052212.html 343KB ff_11281135.html 383KB safe_key.txt 585B ff_11302203.html 392KB ff_11281135.txt 21KB normal-use1.txt 0B ff_11302226.html 339KB ff_11261946.html 415KB ff_12052236.html 355KB ff_12011805.html 108KB replace1.txt 104KB ff_11302249.html 443KB save_with_pickle.ipynb 1KB save test.pickle 37B jpblib.pkl 14B joblib.pkl 14B getPayloadPrarmaters.ipynb 2KB core.ipynb 38KB try.py 61B getNormalPara.py 934B core.py 4KB getNormalParamaters.ipynb 3KB some test.ipynb 3KB firstTry.ipynb 3KB wordProcess.pyc 3KB removeRepeat.ipynb 3KB save_with_joblib.ipynb 2KB removeRepeat.pyc 751B .ipynb_checkpoints removeRepeat-checkpoint.ipynb 3KB core-checkpoint.ipynb 37KB save_with_pickle-checkpoint.ipynb 1KB justTest-checkpoint.ipynb 1KB getNormalParamaters-checkpoint.ipynb 6KB some test-checkpoint.ipynb 3KB save_with_joblib-checkpoint.ipynb 2KB getPayloadPrarmaters-checkpoint.ipynb 29.42MB firstTry-checkpoint.ipynb 3KB README.md 680B count.py 215B资源评论
