sniffle

v0.0.1

Published

2 years ago

CLI app that learns and identifies user agent strings using a Redis-backed naive Bayes classifier. Because why the hell not.

Downloads

0High
0Medium
0Low

rgrove

user agent useragent ua naive Bayesian Bayes classifier parser browser

Sniffle

A CLI app (and someday a web app and REST API) that learns and identifies user agent strings using a Redis-backed naive Bayes classifier. Because why the hell not.

Currently incomplete, a bit rough, and not yet terribly useful (will it ever be?), but with a small training set it's pretty damn accurate (upwards of 90%) at identifying user agents it's never seen before. I'm working on improving that accuracy even more.

Why?

It seemed like a fun problem to solve, I guess.

Browser user agents are notoriously hard to parse accurately, and browsers love pretending to be other browsers. A parser that works okay today could break without warning when a new browser appears that pretends to be every other browser.

A naive Bayes classifier isn't a parser, though. It doesn't know crap about user agent strings, but if you show it enough of them and tell it what they are, it can learn to make pretty accurate guesses about new user agents it's never seen before. When a new user agent comes along that parsers don't know how to handle, a well-trained Bayes classifier might just be able to figure it out without needing any changes.

Is this useful? I don't know. But it was fun.

Install

git clone git://github.com/rgrove/sniffle.git
cd sniffle && npm install

You'll need to have a Redis server handy. Right now Sniffle only tries to connect to localhost. I'll make it more configurable eventually.

Usage

bin/sniffle.js [OPTIONS] <command> [ARGS]

Training

Out of the box, Sniffle knows nothing. You could train it yourself, one category per user agent at a time, but it'd be a lot easier to just import the sizable set of user agent data included in this repo:

$ bin/sniffle.js import data/useragents.json

Classifying a user agent

$ bin/sniffle.js classify 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_2) AppleWebKit/600.3.18 (KHTML, like Gecko) Version/8.0.3 Safari/600.3.18'
Name: Safari [38.1086919880846]
Engine: WebKit [40.60950658800003]
OS: Mac OS X [40.98912788439728]
Type: Browser [45.890633451452445]

Other things I'm too lazy to mention here

$ bin/sniffle.js --help

Plans

Eventually I'll wrap a pretty web app and REST API around this thing. Eventually.

License (MIT)

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

useragents.json

The file data/useragents.json includes data from the MAUL project at https://github.com/bholley/maul/tree/master/data, and is distributed under the following license: