Wrong Calculations When Computing Bsas Algorithm in the Library

[Ahmed-Nezar]

Consider this example:

data_points = [
    (2, 5),  
    (6, 4),  
    (5, 3),  
    (2, 2),  
    (1, 4),  
    (5, 4),  
    (3, 3),  
    (2, 3),  
    (2, 4),  
    (8, 2),  
    (9, 2),  
    (10, 2), 
    (11, 2), 
    (10, 3), 
    (9, 1)   
]
metric = distance_metric(type_metric.MANHATTAN)
bsas_instance = bsas(data_points, maximum_clusters=3, threshold=1.5, metric=metric)
bsas_instance.process()
clusters = bsas_instance.get_clusters()
representatives = bsas_instance.get_representatives()

Output of clusters is [[0, 3, 4, 7, 8], [1, 5, 9, 10, 11, 12, 13, 14], [2, 6]] the 6th point (3,3) should be assigned to cluster 1 not cluster 3 or I am getting something wrong here in the Algorithm implementation

[Seif-Yasser-Ahmed]

You are right, it appears that the Basic Sequential Algorithmic Scheme (BSAS) implementation assigns new data points to clusters with a Manhattan distance greater than the specified threshold. This happens even when there is a different cluster with a Manhattan distance lower than the threshold, leading to incorrect clustering results.

Steps to Reproduce:

Use the BSAS algorithm with the Manhattan distance metric and set a specific threshold.
Provide a dataset where a new data point should clearly belong to an existing cluster based on its distance (lower than the threshold).
Observe that the point is incorrectly assigned to a cluster with a larger distance.
Expected Behavior:
New data points should be assigned to the closest cluster (with a distance below the threshold after calculating the new mean of the cluster) or used to form a new cluster if no suitable cluster exists.

Observed Behavior:
New data points are assigned to clusters with distances exceeding the threshold, ignoring closer clusters with distances within the threshold.