feat: added algorithm updates

.gitignore

@@ -73,3 +73,6 @@ __pycache__/
 
 # Databases
 *.sqlite3
+
+# Test result files
+output.csv

tests/algorithm/Dockerfile.algo

@@ -9,4 +9,4 @@ COPY scripts /app/scripts
 COPY data /app/data
 
 # Install Python dependencies
-RUN pip install --no-cache-dir pandas requests
+RUN pip install --no-cache-dir requests

tests/algorithm/README.md

@@ -1,6 +1,6 @@
 # Record Linkage Algorithm Testing
 
-This repository contains a project to test the effectiveness of the RecordLinker algorithm.
+This repository contains a project to evaluate the match accuracy performance of the RecordLinker algorithm.
 
 ## Prerequisites
 
@@ -12,48 +12,75 @@ Before getting started, ensure you have the following installed:
 ## Directory Structure
 
 - `/`: Contains the `.env` file and `Dockerfile` to build
-- `configurations/`: Contains the configuration file for the algorithm tests
-- `data/`: Contains the data `.csv` files used for the algorithm tests (seed file and test file)
-- `results/`: Contains the results of the algorithm tests
-- `scripts/`: Contains the scripts to run the algorithm tests
+- `configurations/`: Contains the configuration `.json` file that will be used for the test
+- `data/`: Contains the data `.csv` files used for the algorithm test (seed file and test file)
+- `results/`: Contains the results `.csv` file after running the test
+- `scripts/`: Contains the scripts to run the test
 
-## Steup
+## Setup
 
 1. Build the Docker images:
 
     ```bash
     docker compose --profile algo-test build
     ```
 
-2. Configure environment variables
+2. Add seed and test data files
+    You can use the sample data files provided in the `data` directory or add your own data files.
+    The format of the input files should be a CSV file with the same column headers as shown in the sample files.
+
+        `/data/sample_seed_data.csv`
+
+        `/data/sample_test_data.csv`
+
+
+3. Configure environment variables
 
     `/algo.env`
 
     Edit the environment variables in the file
 
-3. Edit the algorithm configuration file
+4. Edit the algorithm configuration file
 
     `/configurations/algorithm_configuration.json`
 
     Edit the configuration file to tune the algorithm parameters
 
 ## Running Algorithm Tests
 
-1. Run the tests
+1. Run the test
 
     ```bash
-    docker compose --profile algo-test run --rm algo-test-runner python scripts/run_test.py
+    docker compose run --rm algo-test-runner scripts/run_test.py
     ```
 
 2. Analyze the results
 
     The results of the algorithm tests will be available in the `results/output.csv` file.
 
-    The results will be in a csv formatted file with each test case number, the expected result, and the actual response from the algorithm.
+    The results will be in a CSV formatted file with the following columns: 
+    `Test Case #`, `Expected Result`, `Match Result`, `Details`
+
+## Rerunning Algorithm Tests
+
+After you've run the algorithm tests, you may want to rerun the tests with different seed data, test data, or configurations.
+
+Edit the csv files and/or the configuration file as needed and then run the following commands to rerun the tests.
+
+1. Reset the mpi database
+    
+    ```bash
+    docker compose run --rm algo-test-runner python scripts/reset_db.py
+    ```
+2. Run the tests
+    
+    ```bash
+    docker compose run --rm algo-test-runner scripts/run_test.py
+    ```
 
 ## Environment Variables
 
-1. `env_file`: The attributes that should be tuned for your particular algorithm test,
+1. `env file`: The attributes that should be tuned for your particular algorithm test,
     are located in the `algo_test.env` file.
 
 2. `environment`: The attributes that should likely remain static for all algorithm tests are located directly in the `compose.yml` file.
@@ -74,4 +101,4 @@ After you've finished running algorithm tests and analyzing the results, you can
 
 ```bash
 docker compose --profile algo-test down
-```
+```

tests/algorithm/algo.env

@@ -1,4 +1,4 @@
-SEED_FILE="data/seed_data.csv"
-TEST_FILE="data/test_data.csv"
+SEED_FILE="data/sample_seed_data.csv"
+TEST_FILE="data/sample_test_data.csv"
 ALGORITHM_CONFIGURATION="configurations/algorithm_configuration.json"
 ALGORITHM_NAME="test-config"

tests/algorithm/configurations/algorithm_configuration.json

@@ -7,8 +7,7 @@
     "passes": [
         {
             "blocking_keys": [
-                "BIRTHDATE",
-                "SEX"
+                "BIRTHDATE"
             ],
             "evaluators": [
                     {

tests/algorithm/data/sample_seed_data.csv

@@ -0,0 +1,6 @@
+Match Id,ID,BIRTHDATE,FIRST,LAST,SUFFIX,MAIDEN,RACE,ETHNICITY,GENDER,ADDRESS,CITY,STATE,COUNTY,ZIP,SSN,
+1,3020167,1951-06-02,Linda,Nash,Sr,Gutierrez,Asian,Hispanic,F,968 Gonzalez Mount,South Emilybury,GU,North Kennethburgh County,93236,675-79-1449,
+2,9488697,1942-08-03,Jose,Singleton,Sr,Ingram,Asian,Hispanic,M,631 Fowler Causeway,Port Williamfurt,IN,Wardburgh County,90637,587-60-3668,
+3,1805504,1963-01-29,Ryan,Lawrence,IV,Armstrong,Black,Non-Hispanic,M,5256 Lisa Light,Port Monica,GA,South Christine County,51813,371-33-0433,
+4,1792678,1950-08-10,Thomas,Brady,II,Cobb,White,Unknown,M,944 Hayes Port,Jonesville,FM,Jonesview County,6015,272-78-9905,
+5,1332302,1972-08-26,Angie,Murphy,Sr,Mcmahon,Black,Non-Hispanic,F,60015 Edward Vista Suite 518,Lake Andreaview,UT,North Rodney County,46540,740-16-5170,

tests/algorithm/data/sample_test_data.csv

@@ -0,0 +1,7 @@
+Test Case #,Match Id,ID,BIRTHDATE,FIRST,LAST,SUFFIX,MAIDEN,RACE,ETHNICITY,GENDER,ADDRESS,CITY,STATE,COUNTY,ZIP,SSN,Expected Result
+1,1,3020167,1951-06-02,Linda,Nash,Jr,Gutierrez,Asian,Hispanic,F,968 Gonzalez Mount,South Emilybury,GU,North Kennethburgh County,93236,675-79-1449,Should be a Match
+2,2,9488697,1942-08-03,Singleton,Jose,Sr,Ingram,Asian,Hispanic,M,631 Fowler Causeway,Port Williamfurt,IN,Wardburgh County,90637,587-60-3668,Should be a Match
+3,3,1805504,1963-01-29,Ryan,Law-rence,IV,Armstrong,Black,Non-Hispanic,M,5256 Lisa Light,Port Monica,GA,South Christine County,51813,371-33-0433,Should be a Match
+4,4,1792678,1950-08-10,Tho-mas,Brady,II,Cobb,White,Unknown,M,944 Hayes Port,Jonesville,FM,Jonesview County,6015,272-78-9905,Should be a Match
+5,4,1792678,1950-08-10,ThoMas,Brady,II,Cobb,White,Unknown,M,944 Hayes Port,Jonesville,FM,Jonesview County,6015,272-78-9905,Should be a Match
+6,0,1792679,1950-18-10,ThoMas,Brady,II,Cobb,White,Unknown,M,944 Hayes Port,Jonesville,FM,Jonesview County,6015,272-78-9905,Should fail