IKNN Explicit

This page analyzes the hyperparameter tuning results for the ItemKNN scorer in explicit-feedback mode.

Parameter Search Space

Parameter Type Distribution Values Selected
max_nbrs Integer Uniform 2 ≤ \(x\) ≤ 50 33
min_nbrs Integer Uniform 1 ≤ \(x\) ≤ 5 2
min_sim Float LogUniform 1e-06 ≤ \(x\) ≤ 0.1 2.42e-06

Final Result

Searching selected the following configuration:

{'max_nbrs': 33, 'min_nbrs': 2, 'min_sim': 2.4241169726298714e-06}

With these metrics:

{
    'RBP': 0.03763695085296711,
    'LogRBP': 0.5144709940683754,
    'NDCG': 0.31592988623365825,
    'RecipRank': 0.08791259351908244,
    'RMSE': 0.7588452558611104,
    'TrainTask': '7b3f70b7-b82e-4f07-b4e9-1e8aa24556db',
    'TrainTime': 1.482392541016452,
    'TrainCPU': 1.4771879999999997,
    'TestTask': '9c791a55-6058-43e3-ba51-a8d87c56cf95',
    'TestTime': 119.39707937597996,
    'TestCPU': 119.835566,
    'timestamp': 1746386046,
    'checkpoint_dir_name': None,
    'done': True,
    'training_iteration': 1,
    'trial_id': '43fb673c',
    'date': '2025-05-04_15-14-06',
    'time_this_iter_s': 123.28853225708008,
    'time_total_s': 123.28853225708008,
    'pid': 2419446,
    'hostname': 'CCI-ws21',
    'node_ip': '10.248.127.152',
    'config': {'max_nbrs': 33, 'min_nbrs': 2, 'min_sim': 2.4241169726298714e-06},
    'time_since_restore': 123.28853225708008,
    'iterations_since_restore': 1,
    'experiment_tag': '98_max_nbrs=33,min_nbrs=2,min_sim=0.0000'
}

Parameter Behavior

Neighborhood Size

The neighborhood size is key argument to consider, let’s look at RMSE with repspect to that:

Filtering

There are two filtering parameters we also test — minimum neighbors and mininmum simialrity.