Business Context
A telecom company receives thousands of support tickets daily. They want to automatically tag tickets related to 'Billing Disputes' and 'Technical Outages' to route them to the correct department efficiently. The rules have already been extracted from a training set.
About the Set : boolRule
Extraction of Boolean rules for classification.
Discover all actions of boolRule Data Preparation
Creation of a document-term matrix representing support tickets and a rule definition table.
Copied!
| 1 | |
| 2 | DATA casuser.ticket_terms; |
| 3 | INPUT ticket_id term_id; |
| 4 | DATALINES; |
| 5 | 101 1 101 2 102 3 103 1 103 4 ; |
| 6 | |
| 7 | RUN; |
| 8 | |
| 9 | DATA casuser.defined_rules; |
| 10 | INPUT ruleid termid; |
| 11 | DATALINES; |
| 12 | 1 1 1 2 2 3 ; |
| 13 | |
| 14 | RUN; |
| 15 |
Étapes de réalisation
1
Execution of the brScore action to classify the tickets against the defined rules.
Copied!
| 1 | |
| 2 | PROC CAS; |
| 3 | boolRule.brScore TABLE={name='ticket_terms', caslib='casuser'} ruleTerms={name='defined_rules', caslib='casuser'} docId='ticket_id' termId='term_id' casOut={name='scored_tickets', caslib='casuser', replace=true}; |
| 4 | |
| 5 | QUIT; |
| 6 |
Expected Result
The 'scored_tickets' table is created. Ticket 101 should match Rule 1 (terms 1 and 2 present). Ticket 102 should match Rule 2 (term 3 present). Ticket 103 has term 1 but missing term 2, so it should not match Rule 1.