PEAK SALE AND ONE YEAR SALE PREDICTION FOR HARDCOVER FIRST RELEASES

Item request has been placed! ×
Item request cannot be made. ×
loading   Processing Request
  • Publication Date:
    January 3, 2019
  • Additional Information
    • Document Number:
      20190005519
    • Appl. No:
      16/012181
    • Application Filed:
      June 19, 2018
    • Abstract:
      Systems and methods are disclosed for predicting a product's (e.g., a book's) performance prior to its availability. An example embodiment is a system for machine learning classification that includes representations of characteristics of products, a pre-processor, and a machine learning classifier. The pre-processor can determine (i) representations of comparative intrinsic characteristics of the products based on the representations of characteristics of products and (ii) representations of corresponding comparative extrinsic characteristics of the products. The pre-processor can generate a data structure representing relationships between the comparative intrinsic characteristics and the comparative extrinsic characteristics. The machine learning classifier is trained with the data structure. The classifier can return representations of comparative extrinsic characteristics in response to given comparative intrinsic characteristics. A disambiguator can rank a plurality of intervals between the extrinsic characteristics for a plurality of other products and determine an extrinsic characteristic for the given product based on the ranking.
    • Claim:
      1. A system for machine learning classification, the system comprising: representations of characteristics of products; a pre-processor configured to: determine (i) representations of comparative intrinsic characteristics of the products based on the representations of characteristics of products, and (ii) representations of corresponding comparative extrinsic characteristics of the products; and generate a data structure representing relationships between the comparative intrinsic characteristics and the comparative extrinsic characteristics; and a machine learning classifier trained with the data structure, the classifier configured to return representations of comparative extrinsic characteristics in response to given comparative intrinsic characteristics.
    • Claim:
      2. A system as in claim 1 wherein the pre-processor is configured to filter the representations of characteristics of products before determining the representations of comparative intrinsic characteristics of the products and generating the data structure.
    • Claim:
      3. A system as in claim 1 wherein the products are books.
    • Claim:
      4. A system as in claim 3 wherein the intrinsic characteristics of the products are any of fame of an author of the book, previous cumulative sales for the author, genre and topic of the book, publisher value for the book, and seasonal fluctuations.
    • Claim:
      5. A system as in claim 3 wherein the pre-processor is configured to filter the representations of intrinsic characteristics of the books to filter-out books that do not fit in a general market.
    • Claim:
      6. A system as in claim 3 wherein the comparative intrinsic characteristics of the products are differences between a type of characteristic between two books.
    • Claim:
      7. A system as in claim 3 wherein the comparative intrinsic characteristics of the products are differences between multiple types of characteristics between two books.
    • Claim:
      8. A method of machine learning classification, the method comprising: determining representations of comparative intrinsic characteristics of products based on representations of characteristics of the products; determining representations of corresponding comparative extrinsic characteristics of the products; generating a data structure representing relationships between the comparative intrinsic characteristics and the comparative extrinsic characteristics; and training a machine learning classifier with the data structure to return representations of comparative extrinsic characteristics in response to given comparative intrinsic characteristics.
    • Claim:
      9. A method as in claim 8 further comprising filtering the representations of intrinsic characteristics of products before determining the representations of comparative intrinsic characteristics of the products and generating the data structure.
    • Claim:
      10. A method as in claim 8 wherein the products are books.
    • Claim:
      11. A method as in claim 10 wherein the intrinsic characteristics of the products are any of fame of an author of the book, previous cumulative sales for the author, genre and topic of the book, publisher value for the book, and seasonal fluctuations.
    • Claim:
      12. A method as in claim 10 wherein the comparative intrinsic characteristics of the products are differences between a type of characteristic between two books.
    • Claim:
      13. A method as in claim 10 wherein the comparative intrinsic characteristics of the products are differences between multiple types of characteristics between two books.
    • Claim:
      14. A system for machine learning classification, the system comprising: a machine learning classifier trained with data representing relationships between intrinsic characteristics of products and extrinsic characteristics of the products, the classifier configured to return a plurality of representations of comparative extrinsic characteristics for a given product in response to a plurality of comparative intrinsic characteristics between the given product and a plurality of other products; and a disambiguator configured to, based on the plurality of representations of comparative extrinsic characteristics for the given product, rank a plurality of intervals between the extrinsic characteristics for the plurality of other products and determine an extrinsic characteristic for the given product based on the ranking.
    • Claim:
      15. A system as in claim 14 wherein the disambiguator is configured to rank the plurality of intervals based on a tally of the intervals, an interval being tallied when it is in a range of a given comparative extrinsic characteristic for the given product when compared to a given other product.
    • Claim:
      16. A system as in claim 14 wherein the given product is an unpublished book and the other products are published books.
    • Claim:
      17. A system as in claim 16 the determined extrinsic characteristic for the given product is a peak sales value.
    • Claim:
      18. A method of machine learning classification, the method comprising: inputting, to a machine learning classifier trained with data representing relationships between intrinsic characteristics of products and extrinsic characteristics of the products, a plurality of representations of comparative intrinsic characteristics between a given product and a plurality of other products to obtain a plurality of representations of comparative extrinsic characteristics for the given product; ranking a plurality of intervals between the extrinsic characteristics for the plurality of other products based on the plurality of representations of comparative extrinsic characteristics for the given product; and determining an extrinsic characteristic for the given product based on the ranking.
    • Claim:
      19. A method as in claim 18 wherein ranking the plurality of intervals includes tallying the intervals, an interval being tallied when it is in a range of a given comparative extrinsic characteristic for the given product when compared to a given other product.
    • Claim:
      20. A method as in claim 18 wherein the given product is an unpublished book, the other products are published books, and the determined extrinsic characteristic for the given product is a peak sales value or cumulative sales value.
    • Current International Class:
      06; 06; 06
    • Accession Number:
      edspap.20190005519