Logic for vessel validation scores

vessel-scoring-logic.py

def jaro_cost(inputs, weight=1):
    min_score = 1.0
    for a,b in list(itertools.combinations(inputs, 2)):
        jaro_score = jellyfish.jaro_distance(a, b)
        if jaro_score < min_score:
            min_score = jaro_score

    return (1 - min_score)*weight

def attr_score(attribute):
    score = 0.0
    if len(attribute) > 1:
        attr_set= list_to_set(attribute)
        if len(attr_set) == 1:
            score += 1
        else:
            score -= jaro_cost(attr_set)
    return score

# some logic here to pull the data from the right database, then we start weighting based on the attributes...
for vessel in the_vessel_list:
    score += attr_score(names)
    score += attr_score(callsigns)

    if len(mmsis) > 2:
        score -= (len(mmsis) - 2)

    if input_count > 5:
        score -= (input_count - 5)

    for label in ship_class:
        if label in reverse_classified_ships.keys():
            val = reverse_classified_ships[label]
            observed_types[val] += 1
    if len(observed_types) == 1:
        score += 1