Matching offers to known products
AssigneeMicrosoft Corporation
InventorsAnitha Kannan, Inmar-Ella Givoni
Life of this patent
Estimating the term…
Abstract
A method and apparatus for electronically matching an electronic offer to structured data for a product offering is disclosed. The structure data is reviewed and a dictionary of terms for each attribute from the structure data is created. Attributes in unstructured text may be determined. Each pair of the attributes (name and value) from the unstructured data and the structured data are obtained, the attribute pairs of the structured data and the unstructured data and compared and a similarity level is calculated for the matching the attribute pairs. The structured data pair that has the highest similarity score to the unstructured data pair is selected and returned.
Claims20
A computer based method of matching a text description to a structured record of a product, the method comprising
- obtaining a text description and a structured record each containing one or more attributes of a product, the text description having one or more text segments individually associated with one or more attributes of the structured record;
- selectively assigning a similarity level between the attributes of the one or more text segments and the structured record;
- and
- deriving a weighting factor for the one or more attributes based on the selectively assigned similarity levels such that a product described by the text description is deemed to match that defined by the structured record.
A computer system of matching a text description to one of a set of structured records of products in a database, the structured records individually having a plurality of attributes with attribute values, the computer system having a processor coupled to a memory containing instructions, when executed by the processor, causing the processor to perform a method comprising:
- dividing a text description describing a product into a plurality of text segments each associated with one of the attributes in one of the structured record;
- generating a similarity vector having one or more elements individually representing a similarity of the attribute values in the text segments and those in the structured record;
- calculating a probability of match between the text description and the structured record based on the generated similarity vector;
- and
- selecting one of the structured record as a match for the text description based on the calculated probabilities.
A computer system having a processor coupled to a memory containing instructions, when executed by the processor, causing the processor to perform a method comprising:
- obtaining a set of text descriptions and a set of structured records individually containing a plurality of attributes of a product;
- parsing the text descriptions to form one or more text segments individually associated with one or more attributes of the structured records;
- and
- deriving a weight factor for at least some of the associated attributes such that a product described by the individual text descriptions matches that defined by one of the structured records.
Dependencies are inferred from the claim text.