Untitled
unknown
plain_text
2 years ago
5.5 kB
3
Indexable
pred_data: ticket_category ticket_type ticket_item ticket_summary \
0 Application HCM - OPM Administration Application Function error
ticket_severity resolution_sla_violated created_date
0 2 - Major\t False 2018-09-28 16:29:45
row: ticket_category ticket_type ticket_item ticket_summary ticket_severity \
0 0 0 0 0 0
resolution_sla_violated
0 0
New Data ticket_category ticket_type ticket_item ticket_summary ticket_severity \
0 0 0 0 0 0
resolution_sla_violated
0 0
Similarity Matrix: [[0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]]
Similar Indices (array([0, 1, 2, 3, 4, 5]), array([ 3, 4, 5, 6, 7, 16]))
Similar Row Indices [0 1 2 3 4 5]
Predicted Output person_who_resolved owner_user_id role_name
0 Tanvir Mirkar 104 L2 Support
1 Tanvir Mirkar 104 L2 Support
2 Tanvir Mirkar 104 L2 Support
3 Tanvir Mirkar 104 L2 Support
4 Tanvir Mirkar 104 L2 Support
5 Tanvir Mirkar 104 L2 Support
Recommended users for the ticket are-
Recommendation 1: User Tanvir Mirkar, Owner User ID 104, Role Name L2 Support
Recommendation 2: User Tanvir Mirkar, Owner User ID 104, Role Name L2 Support
Recommendation 3: User Tanvir Mirkar, Owner User ID 104, Role Name L2 Support
Recommendation 4: User Tanvir Mirkar, Owner User ID 104, Role Name L2 Support
Recommendation 5: User Tanvir Mirkar, Owner User ID 104, Role Name L2 Support
Recommendation 6: User Tanvir Mirkar, Owner User ID 104, Role Name L2 Support
If you see here, the original data was-
Application HCM - OPM Administration Application Function error 2 - Major False 0 Roshni Gangrade 983 L2 Support
So when we tested the response it gave the name as Tanvir Mirkar which seems to be totally incorrect .
The main problem is here -
row: ticket_category ticket_type ticket_item ticket_summary ticket_severity \
0 0 0 0 0 0
resolution_sla_violated
0 0
where the input data is encoded so every time it gives 0 0 for every field which eventually returns Tanvir Mirkar as the 1st entry.
I have shared the codes for get_top_persons_resolved and API code as well ,
The get_top_persons_resolved takes input , redirects to feature_engineering() function which is as this-
import pandas as pd
from sklearn.preprocessing import LabelEncoder
import numpy as np
import pdb
def feature_engineering(data):
## Creating a dictionary
#pdb.set_trace()
input_data= data
# Convert the 'creation_date' and 'resolution_date' columns to datetime
input_data['created_date']=pd.to_datetime(input_data['created_date'])
#input_data['ticket_resolution_date']=pd.to_datetime(input_data['ticket_resolution_date'])
#input_data['ticket_resolution_time']=input_data.apply(lambda row:row['ticket_resolution_date']-row['created_date'],axis=1)
#input_data['ticket_resolution_time'] =input_data['ticket_resolution_time'].apply(lambda x: x.total_seconds() / 3600)
## removing three columns and getting a final dataframe for building
input_data.drop(columns=['created_date'],axis=1,inplace=True)
## Performing Encoding for Categorical Columns-
label_enc= LabelEncoder()
input_data['ticket_category']=label_enc.fit_transform(input_data['ticket_category'])
input_data['ticket_type']=label_enc.fit_transform(input_data['ticket_type'])
input_data['ticket_item']=label_enc.fit_transform(input_data['ticket_item'])
input_data['ticket_severity']=label_enc.fit_transform(input_data['ticket_severity'])
input_data['ticket_summary']=label_enc.fit_transform(input_data['ticket_summary'])
input_data['resolution_sla_violated'] = label_enc.fit_transform(input_data['resolution_sla_violated'])
return input_data
Now can you tell what changes can be done in order to avoid this issue after encoding and get correct prediction.Editor is loading...