Systems and methods for characterizing linked documents using a latent topic model
AssigneeNEC Laboratories America, Inc.
InventorsZhen Guo, Shenghuo Zhu, Yun Chi
Life of this patent
Estimating the term…
Abstract
Systems and methods are disclosed for extracting characteristics from a corpus of linked documents by deriving a content link model that explicitly captures direct and indirect relations represented by the links, and extracting document topics and the topic distributions for all the documents in the corpus using the content-link model.
Claims20
A method for characterizing a corpus of documents each having one or more links, comprising:
- a. forming a Bayesian network using the documents;
- b. determining a Bayesian network structure using the one or more links;
- c. generating a content link model;
- and
- d. determining one or more topics in the corpus and topic distribution for each document.
A method for extracting characteristics from a corpus of linked documents, comprising:
- a. deriving a content link model that explicitly captures direct and indirect relations represented by the links, and
- b. extracting document topics and the topic distributions for all the documents in the corpus using the content-link model.
A system for extracting characteristics from a corpus of linked documents, comprising:
- a. computer readable code to derive a content link model that explicitly captures direct and indirect relations represented by the links, and
- b. computer readable code to extract document topics and the topic distributions for all the documents in the corpus using the content-link model.
A system for characterizing a corpus of documents each having one or more links, comprising:
- a. means for forming a Bayesian network using the documents;
- b. means for determining Bayesian network structure using the one or more links;
- c. means for generating a content link model;
- and
- d. means for applying the content link model to determine one or more topics in the corpus and topic distribution for each document.
Dependencies are inferred from the claim text.