示例#1
0
def getSummarizeFeed():
    threadId = request.args['key']
    cur = mysql.connection.cursor()
    query = "SELECT * FROM mails1 WHERE thread_no = " + threadId + ";"
    result1 = cur.execute(query)
    mails = cur.fetchall()
    # print("query : " + query)
    cur.close()
    subject = ""
    sents = []
    subject = mails[0]['subject']
    for mail in mails:
        sents.append(mail['content'])

    # print(sents)

    ob = summarizer.Summarizer(subject, sents)
    data = ob.generate_summary()
    return data
示例#2
0
import summarizer

summary = summarizer.Summarizer()
summary.generate_summaries()
示例#3
0
import summarizer
import summary_generator as sg
import rouge

# Fissare le variabili
body = "rouge-data/body/"
len = "rouge-data/len.txt"
destination = "rouge-data/all_systems_new/"
result = "lda_result-bis.txt"
rouge_script = "rouge-data/ROUGE-1.5.5.pl"
data_rouge = "rouge-data/data/"
summary_dest = "rouge-data/all_systems_n_bis/"
gold = "rouge-data/models/"

# Preparare il summarizer
s = summarizer.Summarizer(model_path="C:/enwiki_20161220_skip_300.bin")

# Preparare il loop
num_topic = [2, 3, 4, 5, 6, 7, 8, 9]
num_words = [2, 3, 4, 5, 6, 7, 8, 9]

print "Start summarizing..."
for topic in num_topic:
    for word in num_words:
        new_dir = 'topic_' + str(topic) + "_word_" + str(word)
        print new_dir
        destination_path = destination + new_dir

        generator = sg.SummaryGenerator(body_dir_path=body,
                                        target_length_path=len,
                                        destination_path=destination_path)
示例#4
0
model_path = 'C:/Users/Peppo/Desktop/w2vm/enwiki_20161220_skip_300.bin'
summary_destination_root = 'C:/grid-search/'
script_path = 'C:/Users/Peppo/Desktop/w2vm/rouge4MultiLing/rouge/ROUGE-1.5.5.pl'
data_path = 'C:/Users/Peppo/Desktop/w2vm/rouge4MultiLing/rouge/data'
gold_standard_path = 'C:/training/summary/'
results_path = 'grid-search-results.txt'
training_body_path = 'C:/training/body/'
training_length_path = 'C:/training/length.txt'

# Set your ranges
tfidf_values = [0.2, 0.25, 0.3, 0.35]
redundancy_values = [0.8, 0.85, 0.9, 0.95]

# Init my summarizer
print 'Loading model...'
s = s.Summarizer(model_path=model_path)
print 'done\n'

# Generate summaries for all parameters combinations
for tfidf_value in tfidf_values:
    print 'Summarization with tdidf ' + str(tfidf_value) + '...'

    for redundancy_value in redundancy_values:
        # Update summarizer's configuration
        tfidf_str = "".join(str(tfidf_value).split("."))
        redundancy_str = "".join(str(redundancy_value).split("."))

        # Generate new name for the destination directory
        new_dir = 'tfidf_' + str(tfidf_str) + "_redundancy_" + str(
            redundancy_str)
        destination_path = summary_destination_root + new_dir
示例#5
0
# coding=utf-8

from rest_framework import generics
from rest_framework import mixins
import json
from django.http import JsonResponse

import summarizer

s = summarizer.Summarizer(
    model_path=
    "C:/Users/Gianni Mastroscianni/Desktop/Magistrale/Accesso Intelligente all'Informazione ed Elaborazione del Linguaggio Naturale/Progetto/word2vec_models/enwiki_20161220_skip_300.bin"
)


class Summary(generics.GenericAPIView, mixins.CreateModelMixin):
    def post(self, request):
        # load json
        data = json.loads(request.body)

        list = []

        redundancy_threshold = data['redundancy_threshold']
        tfidf = data['tfidf_threshold']
        summary_length = data['summary_length']
        query_based_token = data['query_based_token']

        # print (query_based_token, "query token")

        s.set_tfidf_threshold(tfidf)
        s.set_redundancy_threshold(redundancy_threshold)
示例#6
0
import summarizer

SUMMARY_LENGTH = 7

input_file = "input_file.txt"
# "reason_deep_learning_csv.txt"
s = summarizer.Summarizer()
s.set_factors(10, 10, 10)
summary = s.generate_summary(input_file, SUMMARY_LENGTH)
print summary
示例#7
0
import newsProvider
import summarizer

if __name__ == '__main__':
    newsProvider = newsProvider.NewsProvider("Apple stock", 5)
    news = newsProvider.getNews()

    summary = {}
    for source, text in news.items():
        summarize = summarizer.Summarizer(text, source,5)
        print(source, summarize.summarize())
示例#8
0
                            roseType: 'radius',
                            animationType: 'scale',
                            animationEasing: 'elasticOut'
                        }
                    ]
                });
''')
    return indent(doc.getvalue())


while True:
    print('----------------------------------------')
    with open(input("please choose input file:")) as file:
        reviews = '\n\n'.join(item['review'] for item in json.load(file))
    print('parsing text ...')
    s = summarizer.Summarizer(reviews).summary()

    for i, f in enumerate(s):
        print(
            str(i + 1) + '.', '[' + f.lemma + ']',
            str(len(f.positive)) + '/' + str(len(f.negative)))

        print('\tpositive:')
        for n in f.positive[:3]:
            print('\t', n.sent)

        print('\tnegative:')
        for n in f.negative[:3]:
            print('\t', n.sent)

    with open('summary.html', 'w') as file:
示例#9
0
def generate_summary():
    summary = summarizer.Summarizer()
    summary.generate_summaries()
    return ""