Brute Force Algorithm in Routine Detection

This project is an implementation of the brute force algorithm in routine detection with our customized synthetic dataset and dictionary.

Run main code

python -m main --data_dir path-to-activities-simulation.csv 
               --dictionary_dir path-to-dictionary_rooms.json
               --param_m int
               --param_R int 
               --param_C int
               --param_G int
               --epsilon float

Where:

--data_dir: path to the synthetic activity dataset 
--dictionary_dir: path to the dictionary_rooms.json with the correspondencies
--param_m: length of each subsequence
--param_R: least maximum distance 
--param_C: minimum number of matches required
--param_G: least magnitude required
--epsilon: minimum overlap percentage

You can modify in main.py the method plot_results to visualize the results of the algorithm.

    routine_detector.plot_results(title_fontsize=40, labels_fontsize=35,
                                  xlim=(0, 100), xticks_fontsize=18,
                                  yticks_fontsize=20, figsize=(40, 20),
                                  linewidth_bars=5, save_dir="figs/routines_detected.png")

The synthetic activity dataset and the JSON with the correspondencies can be obtained on this website implemented by us Synthetic Activity Dataset

Run tests

    python test/test.py