U tweetove putem weba ili aplikacija drugih proizvođača možete dodati podatke o lokaciji, kao što su grad ili točna lokacija. Povijest lokacija tweetova uvijek možete izbrisati. Saznajte više
Facial Recognition meets malware clustering: training on family names plus some embedding tricks stolen from the FR literature plus TSNE leads to super sharp clusters, with a few cases of potential mislabeling to dig into (check out the potential FNs southeast of the origin)!pic.twitter.com/HEE6MltlBr
Random side note -- the benign samples (which were not in the training data) are clustered around the origin. I almost always see this with embedding models: out of distribution stuff lands near (0,0) when you visualize it in 2D. Anyone have any insight into this phenomenon?
V interesting. There are three shades of orange, which specific malware family of the three includes the FNs?
The three maybe-FNs that jump out at me (black 'x' symbols in the malware cluster just below and a bit right of center) are in a cluster with Troj/Agent-AVYL. (Also: this is an old-ish data set, and we do catch those samples now :) )
My clusters are never this pretty!
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.
