cpragadeesh

local lua_util = require "lua_util"
local lua_settings = require "lua_settings"
local rspamd_kann = require "rspamd_kann"
local ucl = require "ucl"
local argparse = require "argparse"
local rspamd_logger = require "rspamd_logger"
local rspamd_task = require "rspamd_task"
local tensor = require "rspamd_tensor"

local SPAM_LABEL = -1

cpragadeesh

/**
 * Definition for a binary tree node.
 * struct TreeNode {
 *     int val;
 *     TreeNode *left;
 *     TreeNode *right;
 *     TreeNode(int x) : val(x), left(NULL), right(NULL) {}
 * };
 */
class Solution {

cpragadeesh

Pre-rescore test stats


Statistics at threshold: 15
F-score: 0.09
False positive rate: 0.00 %
False negative rate: 95.51 %
Overall accuracy: 43.33 %

cpragadeesh

test23123

cpragadeesh

Corpus testing and Automatic Symbol score generation

Link to repository

Introduction

Emails are scanned by rspamd to produce a list of symbols associated with them (such as MISSING_SUBJECT, SPF_FAIL). Each symbol has a score associated with it. An email's score is the sum total of the scores of the symbols associated with it. This total sum scores determines the action taken on an email. Symbol scores were set manually by us so far. This project aims to generate an optimal set of symbols scores to improve email classification accuracy using Neural Networks.

Project

cpragadeesh

SYMBOL                              OLD_SCORE  NEW_SCORE 
R_MIXED_CHARSET                     5          3.69      
FORGED_MUA_THEBAT_BOUN              2          0.43      
FORGED_MUA_THEBAT_MSGID_UNKNOWN     3          1.52      
MID_BARE_IP                         2          0.64      
FROM_EXCESS_BASE64                  1.5        1.71      
SUBJ_ALL_CAPS                       3          1.56      
FAKE_REPLY_C                        6          2.69      
TO_DOM_EQ_FROM_DOM                  0          3.69      
R_BAD_CTE_7BIT                      4          5.56      

cpragadeesh

SYMBOL                              OLD_SCORE  NEW_SCORE 
R_MIXED_CHARSET                     5          5.43      
FORGED_MUA_THEBAT_BOUN              2          0.44      
FORGED_MUA_THEBAT_MSGID_UNKNOWN     3          17.22     
MID_BARE_IP                         2          -5.16     
FROM_EXCESS_BASE64                  1.5        3.21      
SUBJ_ALL_CAPS                       3          20.95     
FAKE_REPLY_C                        6          2.69      
TO_DOM_EQ_FROM_DOM                  0          9.70      
R_BAD_CTE_7BIT                      4          22.36     

cpragadeesh

# current error = 0.5285708144281
# current error = 0.52857081431571
    .
    .
    .

# StochasticGradient: you have reached the maximum number of iterations
# training error = 0.52857081431571
SYMBOL                              OLD_SCORE  NEW_SCORE 
R_MIXED_CHARSET                     5          4.65      

cpragadeesh

SYMBOL                              OLD SCORE     NEW SCORE 
FAKE_REPLY                          1.0           0.69      
URI_COUNT_ODD                       1.0           1.77      
INVALID_FROM_8BIT                   6.0           6.82      
SUSPICIOUS_RECIPS                   1.5           2.06      
CT_EXTRA_SEMI                       1.0           1.65      
BROKEN_HEADERS                      10.0          11.6      
HAS_X_PRIO_FIVE                     0.0           0.22      
TO_EXCESS_BASE64                    1.5           1.26      
MID_RHS_IP_LITERAL                  0.5           1.14      

cpragadeesh

#include "LinkedList.cpp"
#include <utility>
#include <typeinfo>

using namespace std;

template<class key_type, class value_type>
class HashTable {

	const int size;