Commit 050c8495 by Paktalin

SMS spam detector

parent f2c9faea
This source diff could not be displayed because it is too large. You can view the blob instead.
......@@ -3,7 +3,7 @@ from sklearn.ensemble import AdaBoostClassifier
import pandas as pd
import numpy as np
data = pd.read_csv('./spambase/spambase.data').values
data = pd.read_csv('./files/spambase.data').values
np.random.shuffle(data)
X = data[:, :48]
Y = data[:, -1]
......
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.feature_extraction.text import TfidfVectorizer, CountVectorizer
from sklearn.naive_bayes import MultinomialNB
from wordcloud import WordCloud
def train_test_split(X, Y, test_size):
test_size = int(X.shape[0]*0.33)
Xtrain = X[:-test_size]
Xtest = X[-test_size:]
Ytrain = Y[:-test_size]
Ytest = Y[-test_size:]
return Xtrain, Xtest, Ytrain, Ytest
def visualize(label):
words = ''
for msg in df[df['labels'] == label]['data']:
msg = msg.lower()
words += msg + ' '
word_cloud = WordCloud(width=600, height=400).generate(words)
plt.imshow(word_cloud)
plt.axis('off')
plt.show()
df = pd.read_csv('./files/sms_spam.csv', encoding='ISO-8859-1')
df = df.drop(['Unnamed: 2', 'Unnamed: 3', 'Unnamed: 4'], axis=1)
df.columns= ['labels', 'data']
df['b_labels'] = df['labels'].map({'ham': 0, 'spam': 1})
Y = df['b_labels'].values
count_vectorizer = CountVectorizer(decode_error='ignore')
X = count_vectorizer.fit_transform(df['data'])
Xtrain, Xtest, Ytrain, Ytest = train_test_split(X, Y, test_size=0.33)
model = MultinomialNB()
model.fit(Xtrain, Ytrain)
print("Train score is", model.score(Xtrain, Ytrain))
print("Test score is", model.score(Xtest, Ytest))
# visualize('spam')
# visualize('ham')
df['predictions'] = model.predict(X)
sneaky_spam = df[(df['predictions'] == 0) & (df['b_labels'] == 1)]['data']
for msg in sneaky_spam:
print(msg)
print('\n\n')
not_actually_spam = df[(df['predictions'] == 1) & (df['b_labels'] == 0)]['data']
for msg in not_actually_spam:
print(msg)
\ No newline at end of file
1. Title: SPAM E-mail Database
2. Sources:
(a) Creators: Mark Hopkins, Erik Reeber, George Forman, Jaap Suermondt
Hewlett-Packard Labs, 1501 Page Mill Rd., Palo Alto, CA 94304
(b) Donor: George Forman (gforman at nospam hpl.hp.com) 650-857-7835
(c) Generated: June-July 1999
3. Past Usage:
(a) Hewlett-Packard Internal-only Technical Report. External forthcoming.
(b) Determine whether a given email is spam or not.
(c) ~7% misclassification error.
False positives (marking good mail as spam) are very undesirable.
If we insist on zero false positives in the training/testing set,
20-25% of the spam passed through the filter.
4. Relevant Information:
The "spam" concept is diverse: advertisements for products/web
sites, make money fast schemes, chain letters, pornography...
Our collection of spam e-mails came from our postmaster and
individuals who had filed spam. Our collection of non-spam
e-mails came from filed work and personal e-mails, and hence
the word 'george' and the area code '650' are indicators of
non-spam. These are useful when constructing a personalized
spam filter. One would either have to blind such non-spam
indicators or get a very wide collection of non-spam to
generate a general purpose spam filter.
For background on spam:
Cranor, Lorrie F., LaMacchia, Brian A. Spam!
Communications of the ACM, 41(8):74-83, 1998.
5. Number of Instances: 4601 (1813 Spam = 39.4%)
6. Number of Attributes: 58 (57 continuous, 1 nominal class label)
7. Attribute Information:
The last column of 'spambase.data' denotes whether the e-mail was
considered spam (1) or not (0), i.e. unsolicited commercial e-mail.
Most of the attributes indicate whether a particular word or
character was frequently occuring in the e-mail. The run-length
attributes (55-57) measure the length of sequences of consecutive
capital letters. For the statistical measures of each attribute,
see the end of this file. Here are the definitions of the attributes:
48 continuous real [0,100] attributes of type word_freq_WORD
= percentage of words in the e-mail that match WORD,
i.e. 100 * (number of times the WORD appears in the e-mail) /
total number of words in e-mail. A "word" in this case is any
string of alphanumeric characters bounded by non-alphanumeric
characters or end-of-string.
6 continuous real [0,100] attributes of type char_freq_CHAR
= percentage of characters in the e-mail that match CHAR,
i.e. 100 * (number of CHAR occurences) / total characters in e-mail
1 continuous real [1,...] attribute of type capital_run_length_average
= average length of uninterrupted sequences of capital letters
1 continuous integer [1,...] attribute of type capital_run_length_longest
= length of longest uninterrupted sequence of capital letters
1 continuous integer [1,...] attribute of type capital_run_length_total
= sum of length of uninterrupted sequences of capital letters
= total number of capital letters in the e-mail
1 nominal {0,1} class attribute of type spam
= denotes whether the e-mail was considered spam (1) or not (0),
i.e. unsolicited commercial e-mail.
8. Missing Attribute Values: None
9. Class Distribution:
Spam 1813 (39.4%)
Non-Spam 2788 (60.6%)
Attribute Statistics:
Min: Max: Average: Std.Dev: Coeff.Var_%:
1 0 4.54 0.10455 0.30536 292
2 0 14.28 0.21301 1.2906 606
3 0 5.1 0.28066 0.50414 180
4 0 42.81 0.065425 1.3952 2130
5 0 10 0.31222 0.67251 215
6 0 5.88 0.095901 0.27382 286
7 0 7.27 0.11421 0.39144 343
8 0 11.11 0.10529 0.40107 381
9 0 5.26 0.090067 0.27862 309
10 0 18.18 0.23941 0.64476 269
11 0 2.61 0.059824 0.20154 337
12 0 9.67 0.5417 0.8617 159
13 0 5.55 0.09393 0.30104 320
14 0 10 0.058626 0.33518 572
15 0 4.41 0.049205 0.25884 526
16 0 20 0.24885 0.82579 332
17 0 7.14 0.14259 0.44406 311
18 0 9.09 0.18474 0.53112 287
19 0 18.75 1.6621 1.7755 107
20 0 18.18 0.085577 0.50977 596
21 0 11.11 0.80976 1.2008 148
22 0 17.1 0.1212 1.0258 846
23 0 5.45 0.10165 0.35029 345
24 0 12.5 0.094269 0.44264 470
25 0 20.83 0.5495 1.6713 304
26 0 16.66 0.26538 0.88696 334
27 0 33.33 0.7673 3.3673 439
28 0 9.09 0.12484 0.53858 431
29 0 14.28 0.098915 0.59333 600
30 0 5.88 0.10285 0.45668 444
31 0 12.5 0.064753 0.40339 623
32 0 4.76 0.047048 0.32856 698
33 0 18.18 0.097229 0.55591 572
34 0 4.76 0.047835 0.32945 689
35 0 20 0.10541 0.53226 505
36 0 7.69 0.097477 0.40262 413
37 0 6.89 0.13695 0.42345 309
38 0 8.33 0.013201 0.22065 1670
39 0 11.11 0.078629 0.43467 553
40 0 4.76 0.064834 0.34992 540
41 0 7.14 0.043667 0.3612 827
42 0 14.28 0.13234 0.76682 579
43 0 3.57 0.046099 0.22381 486
44 0 20 0.079196 0.62198 785
45 0 21.42 0.30122 1.0117 336
46 0 22.05 0.17982 0.91112 507
47 0 2.17 0.0054445 0.076274 1400
48 0 10 0.031869 0.28573 897
49 0 4.385 0.038575 0.24347 631
50 0 9.752 0.13903 0.27036 194
51 0 4.081 0.016976 0.10939 644
52 0 32.478 0.26907 0.81567 303
53 0 6.003 0.075811 0.24588 324
54 0 19.829 0.044238 0.42934 971
55 1 1102.5 5.1915 31.729 611
56 1 9989 52.173 194.89 374
57 1 15841 283.29 606.35 214
58 0 1 0.39404 0.4887 124
This file: 'spambase.DOCUMENTATION' at the UCI Machine Learning Repository
http://www.ics.uci.edu/~mlearn/MLRepository.html
| SPAM E-MAIL DATABASE ATTRIBUTES (in .names format)
|
| 48 continuous real [0,100] attributes of type word_freq_WORD
| = percentage of words in the e-mail that match WORD,
| i.e. 100 * (number of times the WORD appears in the e-mail) /
| total number of words in e-mail. A "word" in this case is any
| string of alphanumeric characters bounded by non-alphanumeric
| characters or end-of-string.
|
| 6 continuous real [0,100] attributes of type char_freq_CHAR
| = percentage of characters in the e-mail that match CHAR,
| i.e. 100 * (number of CHAR occurences) / total characters in e-mail
|
| 1 continuous real [1,...] attribute of type capital_run_length_average
| = average length of uninterrupted sequences of capital letters
|
| 1 continuous integer [1,...] attribute of type capital_run_length_longest
| = length of longest uninterrupted sequence of capital letters
|
| 1 continuous integer [1,...] attribute of type capital_run_length_total
| = sum of length of uninterrupted sequences of capital letters
| = total number of capital letters in the e-mail
|
| 1 nominal {0,1} class attribute of type spam
| = denotes whether the e-mail was considered spam (1) or not (0),
| i.e. unsolicited commercial e-mail.
|
| For more information, see file 'spambase.DOCUMENTATION' at the
| UCI Machine Learning Repository: http://www.ics.uci.edu/~mlearn/MLRepository.html
1, 0. | spam, non-spam classes
word_freq_make: continuous.
word_freq_address: continuous.
word_freq_all: continuous.
word_freq_3d: continuous.
word_freq_our: continuous.
word_freq_over: continuous.
word_freq_remove: continuous.
word_freq_internet: continuous.
word_freq_order: continuous.
word_freq_mail: continuous.
word_freq_receive: continuous.
word_freq_will: continuous.
word_freq_people: continuous.
word_freq_report: continuous.
word_freq_addresses: continuous.
word_freq_free: continuous.
word_freq_business: continuous.
word_freq_email: continuous.
word_freq_you: continuous.
word_freq_credit: continuous.
word_freq_your: continuous.
word_freq_font: continuous.
word_freq_000: continuous.
word_freq_money: continuous.
word_freq_hp: continuous.
word_freq_hpl: continuous.
word_freq_george: continuous.
word_freq_650: continuous.
word_freq_lab: continuous.
word_freq_labs: continuous.
word_freq_telnet: continuous.
word_freq_857: continuous.
word_freq_data: continuous.
word_freq_415: continuous.
word_freq_85: continuous.
word_freq_technology: continuous.
word_freq_1999: continuous.
word_freq_parts: continuous.
word_freq_pm: continuous.
word_freq_direct: continuous.
word_freq_cs: continuous.
word_freq_meeting: continuous.
word_freq_original: continuous.
word_freq_project: continuous.
word_freq_re: continuous.
word_freq_edu: continuous.
word_freq_table: continuous.
word_freq_conference: continuous.
char_freq_;: continuous.
char_freq_(: continuous.
char_freq_[: continuous.
char_freq_!: continuous.
char_freq_$: continuous.
char_freq_#: continuous.
capital_run_length_average: continuous.
capital_run_length_longest: continuous.
capital_run_length_total: continuous.
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or sign in to comment