Methods and apparatus for matching entities and for predicting an attribute of an entity based on an attribute frequency value

A method for operating a machine to generate a prediction of an attribute value of at least one attribute of a new entity based on at least one attribute value of at least one attribute of a group of entities, the method comprising steps of:

• a) accepting, by the machine, information relating to the new entity, the group of entities and the attributes and attribute values of the new entity and the group of entities;
• b) for each of the at least one attribute of each entity of the group of entities, adjusting, with the machine, the attribute value associated therewith based on a frequency of the at least one attribute over all entities of the group of entities, to generate an adjusted attribute value;
• c) for each entity of the group of entities, determining, with the machine, a degree to which (i) the entity and (ii) the new entity match based on the adjusted attribute value of each of the at least one attribute of the entity and based on the attribute value associated with each of the at least one attribute of the new entity;
• and
• d) predicting, with the machine, an attribute value of at least one of the at least one attribute of the new entity based on the degree to which (i) the entities of the group of entities and (ii) the new entity match and based on the at least one adjusted attribute value associated with the at least one attribute of the entities of the group of entities.

A method for operating a machine to generate a prediction of an attribute value of at least one attribute of a new entity based on at least one attribute value of at least one attribute of a group of entities, the method comprising steps of:

• a) accepting, by the machine, information relating to the new entity, the group of entities, and the attributes and attribute values of the new entity and the group of entities;
• b) for each of the at least one attribute of each entity of the group of entities, adjusting, with the machine, the attribute value associated therewith based on a frequency of the at least one attribute throughout the entities of the group of entities to thereby generate an adjusted attribute value;
• c) for each of the at least one attribute of the new entity, determining, with the machine, a normalized attribute value associated with the at least one attribute value;
• d) determining, with the machine, a similarity between the new entity and the one entity based on each of the at least one normalized attribute value associated with each of the at least one attribute of the new entity and each of the at least one adjusted attribute value of the at least one attribute of the one entity;
• e) determining, with the machine, a case weight for each entity of the group of entities based on the similarity and a square root of a sum, over all of the at least one attribute, of the adjusted attribute value squared;
• f) summing, with the machine, the case weights of each entity of the group of entities to generate a total case weight;
• g) for each of the at least one attribute of each of the entities of the group of entities, determining, with the machine, a new attribute value based on the attribute value of the at least one attribute of the entity and the case weight of the entity;
• h) for each of the at least one attribute, summing, with the machine, over all of the entities of the group of entities, the new attribute values to generate a new attribute value sum;
• and
• i) predicting, with the machine, an attribute value of at least one of the at least one attribute of the new entity based on the new attribute value sum and the total case weight.

A machine-readable medium having machine-executable instructions which, when executed by the machine, perform a method for operating a machine to generate a determination of a degree to which (i) a new entity having at least one attribute with an associated attribute value, and (ii) one entity selected from a group of entities, each entity of the group of entities having at least one attribute with an associated attribute value, match, the method comprising steps of:

• a) accepting, by the machine, information relating to the new entity, the entities of the group of entities, and the attributes and attribute values of the new entity and the entities of the group of entities;
• b) for each of the at least one attribute of each entity of the group of entities, adjusting, with the machine, the attribute value associated therewith based on a frequency of the at least one attribute throughout the entities of the group of entities to thereby generate an adjusted attribute value;
• and
• c) determining, with the machine, a degree to which (i) the new entity and (ii) the one entity match based on the attribute value associated with each of the at least one attribute of the new entity and the adjusted attribute value of each of the at least one attribute of the one entity.