Method, apparatus, and system for data collection, transformation and extraction to support image and text search of antiques and collectables

An apparatus comprising

• one or more computer processors;
• and
• one or more computer memories;
• wherein the one or more computer processors are programmed by computer stored in the one or more computer memories to execute the steps of:
• generating a knowledge base in a database, by use of the one or more computer processors, the knowledge base comprised of a plurality of fields, wherein a first field of the plurality of fields of the knowledge base specifies a plurality of known brands of a plurality of known items, wherein a second field of the plurality of fields specifies a plurality of known categories corresponding to the plurality of known items, wherein a third field of the plurality of fields specifies a plurality of sets of known image-based parameters of the plurality of known items, wherein a fourth field of the plurality of fields specifies a plurality of sets of known text based description parameters of the plurality of known items;
• receiving in the one or more computer memories an inquiry image and textual descriptive parameters for the inquiry image;
• determining, by use of one or more computer processors, a plurality of matched item identifications which match the inquiry image with a first plurality of the plurality of known items;
• and further comprising sorting the matched item identifications, based on the following:
• for each matched item identification of the plurality of matched item identifications, retrieving a corresponding category identification from a table in computer memory and adding the corresponding category identification to a category list in computer memory;
• determining a category identification popularity score based on each category identification's occurrences in the category identification list,
• analyzing inputs, by one or more computer processors, to determine if the inputs satisfy predetermined pre-validation criteria for one or more of the plurality of known brands, one or more of the plurality of known categories, and one or more of the plurality of known image-based parameters thereby eliminating duplicates based on image identification with categorized items;
• sorting the category identification list in computer memory in descending order of popularity score;
• and
• sorting the plurality of item identifications based on popularity, and identifying in computer memory the most popular item identification of the plurality of item identifications as the most likely item identification match for the inquiry image;
• wherein the step of generating the knowledge base in the database includes receiving a plurality of inputs via one or more computer processors, having one or more experts in one or more fields related to the inputs, analyze the plurality of inputs through the internet and through one or more computer processors, and if the one or more experts determine that the inputs are acceptable, storing the plurality of inputs for one or more of the plurality of known brands, one or more the plurality of known categories, and one or more of the plurality of sets of known image-based parameters of the plurality of known items in the database, and if the one or more experts determine that the inputs are not acceptable, not storing the plurality of inputs in the database;
• and
• wherein a plurality of experts, which includes the one or more experts, are selected by a review manager computer software module which is programmed to search an experts table to find the plurality of experts in response to receiving the plurality of inputs;
• wherein the plurality of experts are assigned by the review manager computer module the task of determining whether the inputs are acceptable;
• and
• wherein the one or more experts, of the plurality of experts, accept the task of determining whether the inputs are acceptable, and wherein experts of the plurality of experts other than the one or more experts, decline the task of determining whether the inputs are acceptable;
• and further comprising:
• selecting one of more experts, by a review manager computer module which is programmed to search an experts table to find the plurality of experts in response to receiving the inputs;
• accepting a review task of determining whether the inputs are acceptable, wherein the plurality of experts are assigned by the review manager computer module to determine whether the inputs are acceptable;
• and
• storing, by one or more computer processors, results of review task and a status of the review in a review record table.