YELP-DS/feature_extraction.py at master · DistributedSystemsGroup/YELP-DS · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
from sklearn.feature_extraction.text import CountVectorizer
from numpy.linalg import norm

from sklearn.preprocessing import normalize, scale
from sklearn import preprocessing
import numpy as np
import re, io, json
from string import punctuation

from time import gmtime, strftime
from datetime import datetime

def words(text):
    """An iterator over tokens (words) in text. Replace this with a
    stemmer or other smarter logic.
    """

    for word in text.split():
        # normalize words by lowercasing and dropping non-alpha characters
        normed = re.sub('[^a-z]', '', word.lower())
        if normed:
            yield normed

def buildingHistogram6Bins(keyword1, keyword2, result):
    if keyword1 == 'strongsubj' and keyword2 == 'positive':
        result[0]+=1
    elif keyword1 == 'weaksubj' and keyword2 == 'positive':
        result[1]+=1
    elif keyword1 == 'strongsubj' and keyword2 == 'neutral':
        result[2]+=1
    elif keyword1 == 'weaksubj' and keyword2 == 'neutral':
        result[3]+=1
    elif keyword1 == 'strongsubj' and keyword2 == 'negative':
        result[4]+=1
    elif keyword1 == 'weaksubj' and keyword2 == 'negative':
        result[5]+=1
    elif keyword1 == 'weaksubj' and keyword2 == 'both':
        result[1]+=1
        result[5]+=1
    elif keyword1 == 'strongsubj' and keyword2 == 'both':
        result[0]+=1
        result[4]+=1

def buildingHistogram3Bins(keyword, result):
    if keyword == 'positive':
        result[6]+=1
    elif keyword == 'negative':
        result[7]+=1
    elif keyword == 'negation':
        result[8]+=1

def buildingHistogram2Bins(ps, ns, result):
    result[12]+=float(ps)
    result[13]+=float(ns)

def buildingHistogramDay(date, result):
    date_object = datetime.strptime(date, '%Y-%m-%d')
    day = date_object.strftime("%A")
    if day == 'Sunday' or day == 'Saturday':
        result[10] = 1
    else:
        result[10] = 0

def buildingHistogramVote(votes, result):
    v_result = votes["funny"] + votes["useful"] + votes["cool"]
    result[11] = v_result

def buildingHistogramLength(length, result):
    result [9] = length

def main(inputFile, outputFile):

    filenames = ['data/input/bow/1StarsSamples.json', 'data/input/bow/2StarsSamples.json', 'data/input/bow/3StarsSamples.json', 'data/input/bow/4StarsSamples.json', 'data/input/bow/5StarsSamples.json']

    vectorizer = CountVectorizer(input='filename', ngram_range=(1,3), stop_words='english', strip_accents='unicode', token_pattern=ur'\b\w+\b')

    dtm = vectorizer.fit_transform(filenames).toarray()
    dtm = scale(dtm)
    vocab = np.array(vectorizer.get_feature_names())

    _vectorizer = CountVectorizer(input='content', ngram_range=(1,3), stop_words='english', strip_accents='unicode', token_pattern=ur'\b\w+\b')
    analyze = _vectorizer.build_analyzer()


    with open(inputFile) as fileobject, open("data/dictionary/mydictionary_2bins.json") as dicObject2Bins, open("data/dictionary/mydictionary_3bins.json") as dicObject3Bins, open("data/dictionary/mydictionary_6bins.json") as dicObject6Bins:
        dicData2Bins = json.load(dicObject2Bins)
        dicData3Bins = json.load(dicObject3Bins)
        dicData6Bins = json.load(dicObject6Bins)
        listOfHistogramAndRating = []

        for i, line in enumerate(fileobject):

            print(i)

            if line == '\n':
                break
            data = json.loads(line)
            #Descripton [0-spos, 1-wpos, 2-sneu, 3-wneu, 4-sneg, 5-wneg, 6-pos, 7-neg, 8-negation, 9-len, 10-day, 11-vote, 12-ps, 13-ns]
            result =    [0     , 0     , 0     , 0     , 0     , 0     , 0    , 0    , 0         , 0    , 0     , 0      , 0.0  , 0.0  ]
            length = 0
            text = data["text"]
            for word in words(text):
                length += 1

                for item in dicData2Bins:
                    if word == item["word"]:
                        buildingHistogram2Bins(item["pos"], item["neg"], result)
                for item in dicData3Bins:
                    if word == item["word"]:
                        buildingHistogram3Bins(item["priorpolarity"], result)
                for item in dicData6Bins:
                    if word == item["word"]:
                        buildingHistogram6Bins(item["type"], item["priorpolarity"], result)

            buildingHistogramDay(data["date"], result)

            buildingHistogramVote(data["votes"], result)

            buildingHistogramLength(length, result)

            character_count = len(text)
            uppercase_count = sum(1 for c in text if c.isupper())
            lowercase_count = sum(1 for c in text if c.islower())
            punctuation_count = sum(1 for c in text if c in punctuation)
            alphabetic_count = sum(1 for c in text if c.isalpha())
            numeric_count = sum(1 for c in text if c.isnumeric())


            #Fearures lesection
            final_result = []


            final_result.append(result[0])
            final_result.append(result[1])
            final_result.append(result[2])
            final_result.append(result[3])
            final_result.append(result[4])
            final_result.append(result[5])
            final_result.append(result[6])
            final_result.append(result[7])
            final_result.append(result[8])
            final_result.append(result[9])
            final_result.append(result[10])
            final_result.append(result[11])
            final_result.append(result[12])
            final_result.append(result[13])

            final_result.append(character_count) #14
            final_result.append(uppercase_count) #15
            final_result.append(lowercase_count) #16
            final_result.append(punctuation_count) #17
            final_result.append(alphabetic_count) #18
            final_result.append(numeric_count) #19

            analyzed = analyze(text)
            #Descripton [20-1star, 21-2star, 22-3star, 23-4star, 24-5star]
            resultBOW = [0       , 0       , 0       , 0       , 0       ]

            for item in analyzed:
                feature_index = vectorizer.vocabulary_.get(item)
                if feature_index:
                    item_histogram = dtm[:, feature_index]

                    resultBOW = [x + y for x, y in zip(resultBOW, item_histogram)]

            for item in resultBOW:
                if sum(resultBOW) != 0:
                    item = item*1.0/sum(resultBOW)
                final_result.append(item)


            listOfHistogramAndRating.append({'rating': data["stars"], 'histogram': final_result})
        with io.open(outputFile, 'w', encoding='utf-8') as outfile:
            outfile.write(unicode(json.dumps(listOfHistogramAndRating, ensure_ascii=False)))

    print "FINISHED"


if __name__  == "__main__":

    starttime = strftime("%Y-%m-%d %H:%M:%S", gmtime())

    config = {}
    execfile("params.conf", config)
    #Change input and output file here
    inputFile = 'data/input/' + str(config["dataset_Samples"]) + 'Samples.json'
    outputFile = 'data/output/histogram_' + str(config["dataset_Samples"]) + 'Samples.json'

    main(inputFile, outputFile)

    endtime = strftime("%Y-%m-%d %H:%M:%S", gmtime())

    mailmessage = 'This is a notification email to show that the task is completed\n' + "Start time: " + starttime + " End time: " + endtime + "\n"
    print(mailmessage)