Update README.md with reproducibility instructions

Author: david-chapela

README.md

@@ -8,8 +8,8 @@ executed activities in the index.
 For example, for an ongoing case `A-B-F-T-W-S-G-T-D`, after building the 5-gram index, the state would be computed
 by searching in the index with the sequence `[W, S, G, T, D]`.
 
-This approach has been submitted as a publication to ICPM 2024 under the title "Efficient State Computation for Log
-Animation and Short-Term Simulation Using N-Gram Indexing", by David Chapela-Campa and Marlon Dumas.
+This approach has been submitted as a publication to VLDB 2025 under the title "Efficient Online Computation of Business
+Process State From Trace Prefixes via N-Gram Indexing", by David Chapela-Campa and Marlon Dumas.
 
 ## Requirements
 
@@ -47,7 +47,7 @@ ongoing_state = n_gram_index.get_best_marking_state_for(n_gram)
 
 ## Evaluation Reproducibility
 
-The scripts with a name starting with `icpm24_` under folder `tests/` contain the necessary code to reproduce the
+The scripts with a name starting with `vldb25_` under folder `tests/` contain the necessary code to reproduce the
 evaluation presented in the publication.
 Most of them are only necessary to preprocess the original datasets.
 This data is already available in this [Zenodo repository](doi.org/10.5281/zenodo.11409897).
@@ -56,32 +56,35 @@ there.
 
 ### Dependencies
 
-The evaluation scripts depend on two versions of PM4PY.
+The evaluation scripts depend on two versions of PM4PY:
 
-1. To discover the process models and measure their fitness, uncomment the line `pm4py = "2.7.11.9"` in the
-   file `pyproject.toml` and run `poetry install`. This is unnecessary if you downloaded the input files from Zenodo (
-   see above).
-2. For the other scripts where the prefix-alignment technique is used, the requirement is a package with a PM4PY fork
-   implemented by Daniel
-   Schuster ([repo](https://github.com/fit-daniel-schuster/online_process_monitoring_using_incremental_state-space_expansion_an_exact_algorithm/)).
+1. To run the script `vldb25_compute_states.py` where the prefix-alignment technique is used, the requirement is a
+   package with a PM4PY fork implemented by Daniel Schuster
+   ([repo](https://github.com/fit-daniel-schuster/online_process_monitoring_using_incremental_state-space_expansion_an_exact_algorithm/)).
    Download the project from the corresponding repository and specify its path in the `pyproject.toml` file in the
    line `pm4py = {path = "../schuster-prefix-alignments"}`, then, run `poetry install`.
+2. For all the other scripts, the used PM4PY version is 2.7.11.9. Uncomment the line `pm4py = "2.7.11.9"` in the file
+   `pyproject.toml` and run `poetry install`.
 
 ### Synthetic Evaluation
 
-1. Adapt the log routes in the file `icpm24_compute_states.py` as said in the comments, and the IDs passed by the `main`
-   function to `compute_current_states()` so the executed datasets are the synthetic ones.
-2. Run the script, obtaining the results with the computed states and runtimes for each proposal in the
-   folder `outputs`.
-3. Move this files to the folder `results`.
-4. Run the script `icpm24_exact_state_accuracy.py`, obtaining the accuracy results in the folder `outputs`.
+1. Install the project with the PM4PY version specified in point 1 (see above).
+2. Comment the lines in the `main()` function in `vldb25_compute_states.py` that run the state computation for real-life
+   logs, leaving only the calls to function `compute_current_states()` for the synthetic datasets.
+3. Run the script, obtaining the results with the computed states and runtimes (also the reachability graphs) for each  
+   proposal in the folder `outputs`.
+4. Reinstall the project with the PM4PY version specified in point 2 (see above).
+5. Run the script `vldb25_compute_states_token_replay.py`, adding the token-based replay results to the previous result
+   files.
+6. Move these files to the folder `results`.
+7. Run the script `vldb25_exact_state_accuracy.py`, obtaining the accuracy results in the folder `outputs`.
 
 ### Real-life Evaluation
 
-1. Adapt the log routes in the file `icpm24_compute_states.py` as said in the comments, and the IDs passed by the `main`
-   function to `compute_current_states()` so the executed datasets are the real-life ones.
-2. Run the script, obtaining the results with the computed states and runtimes for each proposal in the
-   folder `outputs`.
-3. Move this files to the folder `results`.
-4. Run the script `icpm24_next_activity_accuracy.py`, obtaining the accuracy results in the folder `outputs`.
-
+1. Install the project with the PM4PY version specified in point 1 (see above).
+2. Comment the lines in the `main()` function in `vldb25_compute_states.py` that run the state computation for synthetic
+   logs, leaving only the calls to function `compute_current_states()` for the real-life datasets.
+3. Run the script, obtaining the results with the computed states and runtimes for each proposal in the folder
+   `outputs`.
+4. Move this files to the folder `results`.
+5. Run the script `vldb25_next_activity_accuracy.py`, obtaining the accuracy results in the folder `outputs`.