Book content item search
AssigneeGoogle Inc.
InventorsShumeet Baluja, Yushi Jing
Life of this patent
Estimating the term…
Abstract
Methods, systems, and apparatus, including computer program products are provided for ranking distinct book content items based on implicit links to other distinct book content items. The implicit links are defined based on the identification of matching features in the distinct book content items. In some implementations, the matching features are uncommon phrases in textual content of the distinct book content items. Edges representing implicit links are generated between distinct nodes representing distinct book content items in a weighted graph. Search results for distinct book content items can be ordered based on the edges connected to the distinct nodes in the weighted graph that represent the distinct book content items.
Claims16
A system, comprising:
- a data store storing an index of distinct book content items and corresponding characteristics of the distinct book content items;
- a search system comprising one or more processors coupled to the data store to identify distinct book content items, the search system being configured to access an index in which the corresponding characteristics of the identified distinct book content items are stored;
- and
- a book search subsystem coupled to the search system, the book search subsystem configured to i) identify implicit textual links between the distinct book content items based on textual content of the distinct book content items ii) represent the distinct book content items as a weighted graph in computer memory, where each of the distinct book content items is represented as a distinct node in the weighted graph, where an edge exists in the weighted graph between each pair of distinct nodes that represent distinct book content items that both include a matching uncommon word string, and wherein each edge in the weighted graph is weighted based on a relative importance of the distinct node from which the edge originates and iii) generate a rank score for the identified distinct book content items based on the implicit textual links.
A computer-implemented method comprising:
- storing, in a data store, an index of distinct book content items and corresponding characteristics of the distinct book content items;
- accessing the index in which the corresponding characteristics of the identified distinct book content items are stored;
- identifying implicit textual links between the distinct book content items based on textual content of the distinct book content items;
- representing the distinct book content items as a weighted graph in computer memory, where each of the distinct book content items is represented as a distinct node in the weighted graph, where an edge exists in the weighted graph between each pair of distinct nodes that represent distinct book content items that both include a matching uncommon word string, and wherein each edge in the weighted graph is weighted based on a relative importance of the distinct node from which the edge originates;
- and
- generating a rank score for the identified distinct book content items based on the implicit textual links.
One or more non-transitory computer readable media comprising instructions that, when executed by data processing apparatus cause the data processing apparatus to perform operations comprising:
- storing an index of distinct book content items and corresponding characteristics of the distinct book content items;
- accessing the index in which the corresponding characteristics of the identified distinct book content items are stored;
- identifying implicit textual links between the distinct book content items based on textual content of the distinct book content items;
- representing the distinct book content items as a weighted graph in computer memory, where each of the distinct book content items is represented as a distinct node in the weighted graph, where an edge exists in the weighted graph between each pair of distinct nodes that represent distinct book content items that both include a matching uncommon word string, and wherein each edge in the weighted graph is weighted based on a relative importance of the distinct node from which the edge originates;
- and
- generating a rank score for the identified distinct book content items based on the implicit textual links.
Dependencies are inferred from the claim text.