def center(category):
database = DataGetter.get__db()
appIds = database.get_app_ids()
c = center.Center()
for appId in appIds:
c.set_property(category, appId)
data = c.calculate()
c.record_data(data)
def cluster_data(category, Minpts=2, e=0.6):
# 获取应用id
database = DataGetter.get__db()
appIds = database.get_app_ids()
d = dbscan.DBSCAN()
for appId in appIds:
# dbscan算法聚簇
d.set_property(category, appId)
clusters = d.dbscan(Minpts, e)
# 持久化数据
d.record_cluster(clusters)
def classify_data(lower, quantity):
database = DataGetter.get__db()
# 从原始数据中读取指定数量数据
data = database.load_specified_quantity_data(lower, quantity)
# 初始化预处理器和分类器
p = preprocess.Preprocess()
b = bayes.Bayes()
for i in range(len(data)):
# 将数据进行预处理
p.set_sentence(data[i][1])
sentence_processed = p.preprocess()
if sentence_processed == '':
category = con.USELESS
else:
# 朴素贝叶斯算法分类器对文本进行分类
# print sentence_processed, data[i][2]
b.set_property(sentence_processed, data[i][2])
category = b.bayes_classifier_improved2()
data[i] = [data[i][0], sentence_processed, data[i][2], category]
# print data[i]
# 持久化
database.save_data(data)
def __init__(self):
self.__clusterHelper = DataGetter.get__clusterHelper()
self.__db = DataGetter.get__db()
def __init__(self):
self.__database = DataGetter.get__db()
def recover(id, category, appId):
c = DataGetter.get__clusterHelper()
c.recover(id, category, appId)
# @staticmethod
# # 比较1
# def compare(version):
# list = []
#
# p = preprocess.Preprocess()
# b = bayes.Bayes()
# database = DataGetter.get__db()
# data = database.load_specified_quantity_data(1500, 500)
# for d in data:
# p.set_sentence(d[1])
# sentence_processed = p.preprocess()
# b.set_property(sentence_processed, d[2])
# if version == 1:
# res = b.bayes_classifier()
# elif version == 2:
# res = b.bayes_classifier_improved()
# else:
# res = b.bayes_classifier_improved2()
# list.append([d[0], res])
#
# @staticmethod
# def compare2():
# list3 = []
#
# p = preprocess.Preprocess()
# b = bayes.Bayes()
# database = DataGetter.get__db()
# data = database.load_specified_quantity_data(1500, 500)
# bug, feature = 0, 0
# for d in data:
# p.set_sentence(d[1])
# sentence_processed = p.preprocess()
# b.set_property(sentence_processed, d[2])
# res = b.bayes_classifier_improved2()
# list3.append([d[0], res])
# if res == con.BUG:
# bug += 1
# elif res == con.FEATURE:
# feature += 1
#
# print bug, feature, bug + feature
#
# # list2 = compare1()
#
# list1 = file.test()
#
# for i in range(100):
# if not (int(list1[i]) == list2[i][1] and int(list1[i]) == list3[i][1]):
# print list2[i][0], ':', list1[i], list2[i][1], list3[i][1]
#
# for i in range(400):
# if not (list2[i][1] == list3[i][1]):
# print list2[i][0], ':', list2[i][1], list3[i][1]
# print 'end...'
# def compare():
# list1, list2, list3 = [], [], []
# p = preprocess.Preprocess()
# b = bayes.Bayes()
# database = DataGetter.get__db()
# data = database.load_specified_quantity_data(1500, 500)
# for d in data:
# p.set_sentence(d[1])
# sentence_processed = p.preprocess()
# b.set_property(sentence_processed, d[2])
# # 版本1
# list1.append([d[0], b.bayes_classifier()])
# # 版本2
# list2.append([d[0], b.bayes_classifier_improved()])
# # 版本3
# list3.append([d[0], b.bayes_classifier_improved2()])
# return list1, list2, list3
#
# list1, list2, list3 = compare()
# for i in range(500):
# if not (list1[i][1] == list2[i][1] == list3[i][1]):
# print 'id' + str(list2[i][0]), ':', list1[i][1], list2[i][1], list3[i][1]
#
# if __name__ == "__main__":
#
# pass
# 数据分类
# 聚簇
# Controller.classify_data(4909, 3553)
# Controller.cluster_data(con.FEATURE, 2, 0.5)
# Controller.center(1)
# d = dbscan.DBSCAN()
# d.set_property(2, 481)
# da = d.dbscan(2, 0.75)
# for a in da:
# print a
# d.record_cluster(da)
# # prioritize_data(con.BUG)
# # compare1()
# # compare2()
# with open('/Users/dongyibo/Desktop/script/nla_businessReview.sql', 'r') as fr:
# lines = fr.readlines()
# with open('/Users/dongyibo/Desktop/script/nla_businessReview2.sql', 'w+') as fw:
# i = 1
# for line in lines:
# token = line.split('VALUES (')
# sql = 'UPDATE raw_review set appId = ' + token[1][:3] + ' WHERE id = ' + str(i)
# fw.write(sql + ';\n')
# i += 1
# d = dbscan.DBSCAN()
# d.set_property(2, 482)
# clusters = d.dbscan(2, 0.6)
# list = []
# for cluster in clusters:
# l = []
# for c in cluster:
# l.append(c[0])
# list.append(l)
# for l in list:
# print l
def abort(id, category, appId):
# print 'abort'
c = DataGetter.get__clusterHelper()
c.abort(id, category, appId)
def __init__(self, category, appId):
self.__category = category
self.__appId = appId
self.__database = DataGetter.get__db()
self.__clusterHelper = DataGetter.get__clusterHelper()
self.__isAborted = False