Classification of items with manual confirmation

A system, comprising:

• at least one hardware computing device;
• and
• an item classification application executable in the at least one hardware computing device, the item classification application comprising:
• logic that identifies one of a plurality of classifications based at least in part on data associated with an item and a data dictionary, thereby producing an identified classification that is associated with a confidence score;
• logic that automatically assigns the identified classification to the item when the confidence score meets a threshold;
• logic that obtains a manual confirmation of the identified classification when the confidence score is below the threshold, the manual confirmation indicating whether the identified classification is correct;
• logic that builds the data dictionary based at least in part on data describing a plurality of preclassified items, each one of the preclassified items being assigned a manually determined one of the classifications;
• and
• logic that periodically rebuilds the data dictionary based at least in part on data associated with automatically classified items haying manually confirmed classifications.

A method, comprising:

• identifying, in at least one computing device, one of a plurality of classifications based at least in part on a plurality of tokens associated with an item and a data dictionary, thereby producing an identified classification that is associated with a confidence score;
• automatically assigning, in the at least one computing device, the identified classification to the item when the confidence score meets a threshold;
• and
• obtaining, in the at least one computing device, a manual confirmation of the identified classification when the confidence score is below the threshold, the manual confirmation indicating whether the identified classification is erroneous;
• building, in the at least one computing device, the data dictionary based at least in part on data describing a plurality of preclassified items, each one of the preclassified items being assigned a manually determined one of the classifications;
• and
• periodically rebuilding, in the at least one computing device, the data dictionary based at least in part on data associated with automatically classified items having manually confirmed classifications.