/
final.py
120 lines (75 loc) · 3.72 KB
/
final.py
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
import sys
#sys.path.append('C:/Users/jay dev/Anaconda3/Lib/site-packages/numpy')
import pandas as pd;
import numpy as np;
from pandas import Series
from pandas import DataFrame
import sklearn.neighbors
from sklearn.neighbors import NearestNeighbors
import spacy
nlp = spacy.load('en_core_web_md')
path = sys.argv[1]
# "D:\\final_year_project\\ml\\final.csv"
dataset = pd.read_csv(path)
numaricSubset = dataset.select_dtypes(include = [np.number])
numaricSubset = numaricSubset.iloc[:,1:-2]
stringSubset = dataset.select_dtypes(include = [np.object])
maxVisitsRowIndex = dataset["visits"].idxmax()
maxVisitsRowNumaric = numaricSubset.iloc[maxVisitsRowIndex]
maxVisitsRowString = stringSubset.iloc[maxVisitsRowIndex]
def getNumaricDataDistance(numaricSubset:DataFrame , maxVisitsRowNumaric:Series)->Series:
neigh = NearestNeighbors()
neigh.fit(numaricSubset)
distanceDetails = neigh.kneighbors([maxVisitsRowNumaric])
distance = distanceDetails[0].flatten()
indexes = distanceDetails[1].flatten()
return Series(distance , index = indexes)
numaricDistanceSeries = getNumaricDataDistance(numaricSubset , maxVisitsRowNumaric)
def calculateSimilarityScore(word1:str , word2:str)->int:
concatinatedWords = word1+" "+word2
tokens = nlp(concatinatedWords)
return tokens[0].similarity(tokens[1])
def findSimilarity(stringSubset:DataFrame,maxVisitsRowString:Series)->DataFrame:
similarity = {}
for (columnName,columnValues) in stringSubset.iteritems():
columnSimilarity = [None]*len(columnValues)
for index,item in columnValues.items():
columnSimilarity[index] = calculateSimilarityScore(item, maxVisitsRowString[columnName])
similarity[columnName] = columnSimilarity
return pd.DataFrame(similarity)
def calcualteAverageSimilarity(similarityDataFrame:DataFrame)->Series:
averageSimilarityList = [None]*len(similarityDataFrame)
for index,row in similarityDataFrame.iterrows():
averageSimilarityList[index] = row.mean()
return Series(averageSimilarityList)
def getAverageSimilarity(stringSubset:DataFrame,maxVisitsRowString:Series)->Series:
similarityDataFrame = findSimilarity(stringSubset, maxVisitsRowString)
averageSimilarity = calcualteAverageSimilarity(similarityDataFrame)
return averageSimilarity
textSimilaritySeries = getAverageSimilarity(stringSubset,maxVisitsRowString)
def formatNumaricDistance(numaricDistanceSeries:Series)->Series:
subtractedList = [None]*len(numaricDistanceSeries)
maxValue = numaricDistanceSeries.max()
for index in numaricDistanceSeries.index:
subtractedList[index] = maxValue-numaricDistanceSeries[index]
subtractedSeries = Series(subtractedList)
subtractedMax = subtractedSeries.max()
dividedSeries = subtractedSeries.divide(subtractedMax)
return dividedSeries
numaricPrioritySeries = formatNumaricDistance(numaricDistanceSeries)
numaricColumnCount =len(numaricSubset.columns)
stringColumnCount = len(stringSubset.columns)
totalColumnCount = numaricColumnCount+stringColumnCount
numaricRatio = numaricColumnCount/totalColumnCount
stringRatio = stringColumnCount/totalColumnCount
updatedTextSimilarity = textSimilaritySeries.mul(stringRatio)
updatedNumaricSeries = numaricPrioritySeries.mul(numaricRatio)
columnPriority = updatedTextSimilarity.add(updatedNumaricSeries)
visitColumn = dataset["visits"]
visitMaxValue = visitColumn.max()
visitPriority = visitColumn.div(visitMaxValue)
finalPriority = columnPriority.add(visitPriority)
finalPriority = finalPriority.divide(2)
dataset["priority"] = finalPriority
dataset.to_csv(path,index = False)
print('written')