Business Context
A large engineering firm has a knowledge base with thousands of technical documents. They need to stress-test the autocomplete generation to ensure it can handle a larger volume of similar technical terms (e.g., 'Hydraulic Pump Specification v1', 'Hydraulic Pump Specification v2') without performance degradation.
About the Set : searchAnalytics
Data indexing and search functionalities.
Discover all actions of searchAnalytics Data Preparation
Generating a larger synthetic dataset with repetitive technical headers to simulate volume.
Copied!
| 1 | |
| 2 | DATA mycas.tech_docs; |
| 3 | LENGTH doc_title $100; |
| 4 | DO i=1 to 10000; |
| 5 | doc_title = catx(' ', 'Technical Specification Document', 'Version', put(i, 5.), 'for Component', put(mod(i, 10), 2.)); |
| 6 | OUTPUT; |
| 7 | END; |
| 8 | |
| 9 | RUN; |
| 10 | PROC CAS; |
| 11 | search.buildTermIndex / TABLE={name='tech_docs'} docId='doc_title' casOut={name='tech_terms', replace=true}; |
| 12 | |
| 13 | RUN; |
| 14 |
Étapes de réalisation
1
Building the auto-complete index on the larger technical term dataset.
Copied!
| 1 | |
| 2 | PROC CAS; |
| 3 | searchAnalytics.buildAutoComplete / index={name='tech_terms'} casOut={name='kb_autocomplete', replace=true}; |
| 4 | |
| 5 | RUN; |
| 6 |
2
Validating table info to check row count and size.
Copied!
| 1 | |
| 2 | PROC CAS; |
| 3 | TABLE.tableInfo / TABLE={name='kb_autocomplete'}; |
| 4 | |
| 5 | RUN; |
| 6 |
Expected Result
The system creates the 'kb_autocomplete' table efficiently even with a higher cardinality of terms. The tableInfo action confirms the table exists and has a row count consistent with the unique terms generated from the 10,000 documents.