Abstract: A structural analysis system analyzes the pages of a web site, including the content of the pages and their relationships to each other. Based on the analysis, the structural analysis system additionally modifies (or suggests modification of) the web pages of the web site, such as by altering the inter-page linking structure of the web site. This improves the structure of the web site, leading search engines to be able to more quickly find and index its pages, as well as increasing the expected ranking that a search engine would provide to pages of the web site within the set of search results provided by the search engine in response to a user query.

Abstract: Summary generation functionality generates summaries (“content briefs”) with respect to topics from an existing topic graph that organizes a set of keywords into related groups over a set of electronic documents, such as web pages. Each content brief provides a summary of a corresponding topic from the topic graph. Content authors can consult the content brief for a given topic to gain insight into how to revise or draft a web page or other electronic document to improve its ranking in a search engine with respect to which the topic graph was generated.

Abstract: A structural analysis system analyzes the pages of a web site, including the content of the pages and their relationships to each other. Based on the analysis, the structural analysis system additionally modifies (or suggests modification of) the web pages of the web site, such as by altering the inter-page linking structure of the web site. This improves the structure of the web site, leading search engines to be able to more quickly find and index its pages, as well as increasing the expected ranking that a search engine would provide to pages of the web site within the set of search results provided by the search engine in response to a user query.

Abstract: A system generates a topic graph based on the SERP data for high-ranking keywords in a search engine. Clustering may be based on (for example) degrees of intersection between links in search results of keywords from the SERP data, or keyword embeddings on the SERP data. The topic graph loosely clusters the keywords, such that the keywords have at least a threshold degree of similarity to their clusters, but not necessarily to all the other keywords in the cluster. As a consequence of the loose clustering, a given topic contains keywords that represent different aspects of the same concept, such that a content viewer would likely be interested in a piece of content that addresses the different aspects, and a search engine would be more likely to highly rank the content within its search results for one of the keywords.

Abstract: A technique for increasing efficiency of inference of structure variables (e.g., an inference problem) using a priority-driven algorithm rather than conventional dynamic programming. The technique employs a probable approximate underestimate which can be used to compute a probable approximate solution to the inference problem when used as a priority function (“a probable approximate underestimate function”) for a more computationally complex classification function. The probable approximate underestimate function can have a functional form of a simpler, easier to decode model. The model can be learned from unlabeled data by solving a linear/quadratic optimization problem. The priority function can be computed quickly, and can result in solutions that are substantially optimal. Using the priority function, computation efficiency of a classification function (e.g., discriminative classifier) can be increased using a generalization of the A* algorithm.

Abstract: A technique for increasing efficiency of inference of structure variables (e.g., an inference problem) using a priority-driven algorithm rather than conventional dynamic programming. The technique employs a probable approximate underestimate which can be used to compute a probable approximate solution to the inference problem when used as a priority function (“a probable approximate underestimate function”) for a more computationally complex classification function. The probable approximate underestimate function can have a functional form of a simpler, easier to decode model. The model can be learned from unlabeled data by solving a linear/quadratic optimization problem. The priority function can be computed quickly, and can result in solutions that are substantially optimal. Using the priority function, computation efficiency of a classification function (e.g., discriminative classifier) can be increased using a generalization of the A* algorithm.