[{"data":1,"prerenderedAt":245},["ShallowReactive",2],{"project-22268436":3},{"id":4,"name":5,"line":6,"line_source":7,"title":8,"url":9,"domain":10,"maker":11,"posted_at":12,"year":13,"points":14,"comments":15,"rank_in_year":16,"shown_in_year":17,"top":18,"tier":19,"category":20,"subcategory":21,"format":22,"keywords":23,"reception":28,"sentiment":29,"stance_mix":30,"scored_comments":34,"aspects":35,"strip":41,"post_text":43,"what":44,"said":58,"maker_in_thread":134,"shape":145,"grid":149,"attempts":193,"same_week":208,"maker_shows":37,"maker_hidden":237,"reader_version":238,"card":239},"22268436","MIDAS","Fast Real-Time Anomaly Detection in Dynamic Graphs","title","Show HN: Fast Real-Time Anomaly Detection in Dynamic Graphs","https:\u002F\u002Fgithub.com\u002Fbhatiasiddharth\u002FMIDAS\u002F","github.com","siddharthbhatia",1581096777,2020,118,22,550,3028,19,"notable","ai","graph anomaly detection","library",[24,25,26,27],"anomaly detection","dynamic graphs","edge streams","microcluster detection","loved",0.455,{"positive":31,"neutral":32,"mixed":33,"critical":33},5,6,0,11,{"idea":36,"performance":38,"docs":39},{"positive":37,"critical":33},2,{"positive":37,"critical":33},{"positive":40,"critical":33},1,[33,33,33,33,31,42,33,42,33,42,33,42,33,33,31,42,33,42,33,31,31,31],3,null,{"problem":45,"audience":43,"pricing":43,"motivation":52,"time_to_build":43,"stack":56,"openness":57,"stage":57,"team_size":57},{"value":46,"source":47,"evidence":48},"Detecting suddenly arriving groups of suspiciously similar edges in graph streams in real time","maker",[49],{"id":50,"by":11,"text":51},"22269494","Anomaly detection in graphs is a critical problem for finding suspicious behavior in innumerable systems, such as intrusion detection, fake ratings, and financial fraud. But most…",{"value":53,"source":47,"evidence":54},"Intrusion detection, fake ratings and fraud need real-time detection in dynamic graphs",[55],{"id":50,"by":11,"text":51},[],"unknown",{"praise":59,"objections":90,"requests":98,"questions":115,"alternatives":127},[60,72,83],{"value":61,"source":62,"evidence":63},"Impressive speed and results over prior methods","commenter",[64,68],{"id":65,"by":66,"text":67},"22272942","shivin9","Impressive results! Few questions though:- 1) Do you use sliding window approach or exponential decay? 2) When you say groups of similar edges, do they have to be spatially close…",{"id":69,"by":70,"text":71},"22287088","raghavjajodia","Great to see that! The results look really impressive!",{"value":73,"source":62,"evidence":74},"Real-time detection with theoretical bounds is valuable",[75,79],{"id":76,"by":77,"text":78},"22342330","ZhuoCai","Before I read your paper, I didn't realize the power of hypothesis tests. I was surprised to see theoretical bounds on anomaly detection with such weak constraint through simple…",{"id":80,"by":81,"text":82},"22273788","riteshkumar99","I found the paper to be quite exciting, and it is also solving one of the main tasks in anomaly detection, that is real-time anomaly detection. With other machine\u002Fdeep learning…",{"value":84,"source":62,"evidence":85},"Paper is clear and easy to follow",[86],{"id":87,"by":88,"text":89},"22273642","ramarohit","Interesting work and great paper as anomaly detection in real time is very important now a days. Easy to understand as well.",[91],{"value":92,"source":62,"evidence":93},"Thread has suspected astroturfing from new accounts",[94],{"id":95,"by":96,"text":97},"22274589","ugh123","The comments in this thread exhibit some serious astroturfing. Also the 6 unique commenters accounts all created in the last 12 days. Just sayin'",[99,108],{"value":100,"source":62,"evidence":101},"Handle DDoS-like attacks spread across destinations and multiple servers",[102,105],{"id":103,"by":66,"text":104},"22273490","Thanks. It would be nice to have similarity methods that are not dependent on spatial locality to detect DDoS like attacks.",{"id":106,"by":66,"text":107},"22273662","Ah alright, that makes sense. But it works only when the destination is the same. In a setting where there are multiple web-hosting servers, you would need to treat a group of…",{"value":109,"source":62,"evidence":110},"Extend to non-graph data such as video streaming",[111],{"id":112,"by":113,"text":114},"22273649","sinhayash","Hmm... Interesting work! Can we extend this to find anomalies in other kinds of data like video streaming?",[116,120],{"value":117,"source":62,"evidence":118},"Can attackers evade detection by timing attacks",[119],{"id":65,"by":66,"text":67},{"value":121,"source":62,"evidence":122},"Difference between microcluster detection and subgraph detection",[123],{"id":124,"by":125,"text":126},"22271767","arjitjain","Great work! However, from what I understand, microcluster detection seems to be very similar to subgraph detection. Is there a difference between the two problems? And if yes,…",[128],{"value":129,"source":47,"evidence":130},"Subgraph detection methods",[131],{"id":132,"by":11,"text":133},"22272446","Thanks! There is a subtle difference. We define microclusters within the category of anomalous edge detection as 'suddenly arriving groups of suspiciously similar edges' e.g.…",{"comments":135,"answered":136,"key":137},7,4,[138,139,142],{"id":50,"text":51},{"id":140,"text":141},"22273119","Thank you :) 1. We use a temporal decay (alpha). 2. Good question! We consider similar edges as those having at least one of source and destination node as the same. 3. Very…",{"id":143,"text":144},"22273608","We handle spatial locality in terms of not just the source but also the destination, therefore we should be able to handle DDoS like attacks when simultaneous edges come from…",{"commenters":146,"max_depth":32,"maker_replies":135,"first_comment_mins":147,"lifespan_hours":148},12,92,271,[150,152,156,158,160,162,164,166,167,168,169,171,172,173,176,177,178,180,182,184,188,191],[95,33,33,151,96,33],"other\u002Fneutral\u002Fnone",[153,33,33,154,155,33],"22274670","commenter\u002Fneutral\u002Fnone","adevx",[157,33,33,154,96,33],"22361844",[50,40,33,159,11,40],"project\u002Fpositive\u002Fperformance",[76,37,31,161,77,33],"project\u002Fpositive\u002Fidea",[65,42,42,163,66,33],"project\u002Fneutral\u002Fidea",[140,42,33,165,11,40],"maker\u002Fneutral\u002Fidea",[103,42,42,163,66,33],[143,42,33,165,11,40],[106,42,42,163,66,33],[170,42,33,165,11,40],"22273810",[124,136,42,163,125,33],[132,136,33,165,11,40],[174,31,33,151,175,33],"22275465","shingeki",[69,32,31,159,70,33],[112,135,42,163,113,33],[179,135,33,165,11,40],"22273668",[181,135,42,163,66,33],"22273722",[183,135,33,154,11,40],"22273749",[185,186,31,161,187,33],"22274294",8,"geetika92",[87,189,31,190,88,33],9,"project\u002Fpositive\u002Fdocs",[80,192,31,159,81,33],10,[194,198,201,204,207],{"id":195,"t":196,"points":197,"comments":37},"21504089",1573473897,15,{"id":199,"t":200,"points":34,"comments":33},"22157248",1580106738,{"id":202,"t":203,"points":40,"comments":33},"22236384",1580829194,{"id":205,"t":206,"points":136,"comments":33},"22257005",1581004779,{"id":4,"t":12,"points":14,"comments":15},[209,214,218,223,228,233],{"id":210,"name":211,"points":212,"category":213},"22306996","Autocode",247,"dev_tools",{"id":215,"name":216,"points":217,"category":213},"22288276","geoman.io",214,{"id":219,"name":220,"points":221,"category":222},"22279308","HDMI",195,"hardware",{"id":224,"name":225,"points":226,"category":227},"22263758","LibreLingo",159,"learning_science",{"id":229,"name":230,"points":231,"category":232},"22287174","ssh-over-ssm",131,"infrastructure",{"id":234,"name":235,"points":236,"category":227},"22290812","Mathpendium",108,false,"v2",{"id":240,"name":5,"line":6,"ai":237,"cat":20,"year":13,"points":14,"comments":15,"rec":28,"s":241,"top":18,"kw":242,"maker":11,"domain":10,"tier":19,"attempts":31,"strip":243,"url":9,"mix":244,"praise":61,"objection":92,"sub":21},22268436,0.46,[24,25,26,27],"0000530303030053030555",[31,32,33,33],1794036586298]