The manually selected results for the queries described in our paper are here listed, along with the synonyms found on the ledger (schema transactions) and their respective IDs (seqNo).
| Synonym | Exact Field | seqNo |
|---|---|---|
| first name | first_name, First name, First name, DEMO-First Name, First name, first_name, DEMO-First Name, firstname, DEMO-First Name, firstName, C19EOLegalFirstName, First Name, firstName, firstName, userFirstName, firstName, First name, first_name, first_name, first_name, first_name, First Name, farmer_first_name, farmer_first_name, farmer_first_name, firstName, farmer_first_name, farmer_first_name, farmer_first_name, First Name, First Name, First Name, First Name, First Name, Buyer First Name, First Name |
14632, 31583, 31622, 33629, 35079, 43843, 47768, 54022, 54331, 54389, 54402, 54517, 54538, 54542, 54556, 54572, 54605, 54761, 54773, 54788, 54802, 55727, 56307, 56308, 56309, 56389, 57498, 58175, 58693, 59004, 59551, 59552, 59553, 59555, 59556, 59557 |
| name | DEMO-Name, name, name, name, DEMO-Name, name, name, name, name, name, name, name, name, name |
47766, 54002, 54003, 54170, 54171, 54439, 54782, 54796, 56136, 56365, 57614, 57630, 58041, 58042 |
| given name | Given Name, givenNames, Given Name, Given Name, Given Name, Given Names |
54679, 58011, 58608, 58699, 59232, 59554 |
| patient first name | Patient First Name, Patient First Name, Patient First Name, Patient First Name, Patient First Name |
57735, 57765, 57873, 57900, 58081 |
| member name | Member Name, Member Name |
15227, 15463 |
| full name | Full Name, Full Name |
54564, 54602 |
| inspector name | inspector_name |
58160 |
| student name | DEMO-Student Name |
33627 |
| Synonym | Exact Field | seqNo |
|---|---|---|
| organization | organization, organization, organization, organization, organization_name, organization_name, |
54782, 54796, 57614, 57630, 58692, 59050 |
| company | company_name, companyGroup, company, Company, Company, company, Company |
54439, 54533, 54556, 54602, 55727, 56389, 59004 |
| employer | employer, Employer, Employer |
54019, 59552, 59553 |
| department | department, department, department |
54020, 54876, 55429 |
| role | role, role |
56138, 56181 |
| contract | Contract Signed Date |
59556 |
| payment | payment_indicator |
54767 |
All schemas with employee, income, job, position, work were added by other synonyms.
| Synonym | Exact Field | seqNo |
|---|---|---|
| country | country, country, country, country, DEMO-Country of Residence,country, country, country, country, country,country_of_origin,issuingCountry, personIdentifierCountry, otherDocCountry, Country Code |
99, 188, 1050, 12027, 33629, 41055, 50799, 54000, 54001, 54004, 56308, 58011, 58039, 59554 |
| address | address, address, address, address, address,address, address, supplier_address, address, Street Address |
14635, 43843, 54393, 56310, 57498, 58043, 58044, 58462, 58693, 59556 |
| State | state, institution_state, institution_state, state |
54767, 54792, 54806, 56516 |
| Street | DEMO-Street Name, street |
47768, 54023 |
| Zip | DEMO-Zip |
54331 |
| city | city |
54170 |
All schemas with postal code, province,municipality were added by other synonyms.
Do not run the following commands, unless you have manually updated the list above with the new schemas.
We need to store all schemas to calculate statistics.
echo "-" | docker run -i sovrin-schema-matching:cli --hits 0 --print-seq-only | jq -c '. | sort' > analysis/raw_results/all_schemas.json
echo "first name" | docker run -i sovrin-schema-matching:cli --hits 200 --min-score 0.6 --print-seq-only | jq -c '. | sort' > analysis/raw_results/spacy_first_name.json
echo "company job" | docker run -i sovrin-schema-matching:cli --hits 200 --min-score 0.6 --print-seq-only | jq -c '. | sort' > analysis/raw_results/spacy_company_job.json
echo "address" | docker run -i sovrin-schema-matching:cli --hits 200 --min-score 0.6 --print-seq-only | jq -c '. | sort' > analysis/raw_results/spacy_address.json
curl 'https://sovrin-mainnet-browser.vonx.io/ledger/domain?page=1&page_size=200&query=first%20name&type=101' | jq -c '[.results[] | .txnMetadata.seqNo] | sort' > analysis/raw_results/vonx_first_name.json
curl 'https://sovrin-mainnet-browser.vonx.io/ledger/domain?page=1&page_size=200&query=company%20job&type=101' | jq -c '[.results[] | .txnMetadata.seqNo] | sort' > analysis/raw_results/vonx_company_job.json
curl 'https://sovrin-mainnet-browser.vonx.io/ledger/domain?page=1&page_size=200&query=address&type=101' | jq -c '[.results[] | .txnMetadata.seqNo] | sort' > analysis/raw_results/vonx_address.json
curl 'https://indyscan.io/api/networks/SOVRIN_MAINNET/ledgers/domain/txs?filterTxNames=%5B%22SCHEMA%22%5D&format=expansion&search=first%20name&size=500&skip=0&sortFromRecent=false' | jq -c '[.[] | .imeta.seqNo] | sort' > analysis/raw_results/indyscan_first_name.json
curl 'https://indyscan.io/api/networks/SOVRIN_MAINNET/ledgers/domain/txs?filterTxNames=%5B%22SCHEMA%22%5D&format=expansion&search=company%20job&size=500&skip=0&sortFromRecent=false' | jq -c '[.[] | .imeta.seqNo] | sort' > analysis/raw_results/indyscan_company_job.json
curl 'https://indyscan.io/api/networks/SOVRIN_MAINNET/ledgers/domain/txs?filterTxNames=%5B%22SCHEMA%22%5D&format=expansion&search=address&size=500&skip=0&sortFromRecent=false' | jq -c '[.[] | .imeta.seqNo] | sort' > analysis/raw_results/indyscan_address.json
$ docker build -t sovrin-schema-matching:cli .
$ docker run -it --entrypoint "/usr/local/bin/python" sovrin-schema-matching:cli /app/analysis
Query address
Results for spacy
precision recall f1-score support
False 0.81 0.87 0.84 116
True 0.35 0.25 0.29 32
accuracy 0.74 148
macro avg 0.58 0.56 0.56 148
weighted avg 0.71 0.74 0.72 148
Results for indyscan
precision recall f1-score support
False 0.84 0.96 0.90 116
True 0.69 0.34 0.46 32
accuracy 0.82 148
macro avg 0.76 0.65 0.68 148
weighted avg 0.81 0.82 0.80 148
Results for vonx
precision recall f1-score support
False 0.89 0.96 0.92 116
True 0.78 0.56 0.65 32
accuracy 0.87 148
macro avg 0.84 0.76 0.79 148
weighted avg 0.87 0.87 0.86 148
Query first_name
Results for spacy
precision recall f1-score support
False 0.77 0.93 0.84 81
True 0.88 0.67 0.76 67
accuracy 0.81 148
macro avg 0.83 0.80 0.80 148
weighted avg 0.82 0.81 0.81 148
Results for indyscan
precision recall f1-score support
False 0.64 1.00 0.78 81
True 1.00 0.31 0.48 67
accuracy 0.69 148
macro avg 0.82 0.66 0.63 148
weighted avg 0.80 0.69 0.64 148
Results for vonx
precision recall f1-score support
False 0.70 1.00 0.83 81
True 1.00 0.49 0.66 67
accuracy 0.77 148
macro avg 0.85 0.75 0.74 148
weighted avg 0.84 0.77 0.75 148
Query company_job
Results for spacy
precision recall f1-score support
False 0.90 0.99 0.94 125
True 0.90 0.39 0.55 23
accuracy 0.90 148
macro avg 0.90 0.69 0.74 148
weighted avg 0.90 0.90 0.88 148
Results for indyscan
precision recall f1-score support
False 0.84 1.00 0.92 125
True 0.00 0.00 0.00 23
accuracy 0.84 148
macro avg 0.42 0.50 0.46 148
weighted avg 0.71 0.84 0.77 148
Results for vonx
precision recall f1-score support
False 0.84 1.00 0.92 125
True 0.00 0.00 0.00 23
accuracy 0.84 148
macro avg 0.42 0.50 0.46 148
weighted avg 0.71 0.84 0.77 148
See the README.md on the scalability folder for more details.