Abstract: A system for generating structured queries and carrying out searches using the structured queries may result in high quality queries and may adequately simulate natural language for suggested search queries. In embodiments, attributes and matching product/services may be extracted from various data sources. Structured queries that include the attributes, products/services, and/or additional terms may be combined to form a structured query. In various embodiments, each structured query may be structured according to a pre-defined pattern that is selected based on the type of relevant product and/or service.
Abstract: Techniques for search with autosuggest and refinements are disclosed. In some embodiments, search with autosuggest includes determining a plurality of potential query suggestions for a partially entered query string; and automatically suggesting a plurality of queries based on a query count for each of the queries. For example, the query count can correspond to a popularity of the query. In some implementations, the query count can be determined based on a number of times that the query was received, and the plurality of queries can be listed based on the popularity of each of the plurality of queries (e.g., to facilitate display of more popular queries higher in the list of suggested queries).
Abstract: A system for synonym generation and/or identification can be utilized to make a search engine more effective in finding relevant search results. In embodiments, a synonym generation system includes a phrase vector module, a vector similarity module, and a vector filter module. Candidate synonym phrase pairs are selected from data sources for analysis. Data sources may include a log of search queries, a corpus of web text, and a set of merchant descriptions of products. The data sources may be analyzed with respect to the phrase pairs to generate vector representations of the phrase pairs. The vector representations may then be analyzed to determine a similarity vector. The similarity vector allows the synonym generation system to filter synonyms from the candidate phrase pairs.
Abstract: Techniques for performing query-dependent and content-class (e.g., with product class as a subset, in which content class can be based on a product class in an example implementation) based ranking are disclosed in accordance with some embodiments. In some embodiments, a process for performing query-dependent and content-class based ranking includes receiving a query for a search for content on a web site; performing a query-dependent and content-class based ranking of content available on the web site (e.g., using a processor); and returning a ranked list of content based at least in part on the query-dependent and content-class based ranking of content available on the web site in response to the query.
Abstract: Techniques for thematic landing pages are disclosed. In some embodiments, a process for providing thematic landing pages includes receiving a user query for a theme; determining products (e.g., using a processor) that are relevant to the theme (e.g., based on a content relevancy); and generating a thematic web page for the theme based on the relevant products. For example, the thematic landing page can be associated with a merchant web site, and the relevant products can be products that are available via the merchant web site.
Abstract: Techniques for enhancing content are disclosed. In some embodiments, enhancing content includes automatically identifying a set of modifications to the existing content of a page that would result in the page satisfying a set of search properties and applying at least a subset of the set of modifications to the page. Furthermore, techniques for enhancing link structure are disclosed. In some embodiments, enhancing link structure includes automatically determining a constrained link structure for a set of pages that satisfies a set of one or more constraints and that is at least in part different from an existing link structure of the set of pages and making one or more modifications to the existing link structure to at least in part realize the constrained link structure for the set of pages.
Type:
Grant
Filed:
October 29, 2010
Date of Patent:
February 3, 2015
Assignee:
BloomReach Inc.
Inventors:
Raj K. De Datta, Ashutosh Garg, Joshua Levy, Praveen K. Patnala, Srinath Sridhar, William Uppington, Abhay Vardhan
Abstract: Techniques for performing query-dependent and content-class (e.g., with product class as a subset, in which content class can be based on a product class in an example implementation) based ranking are disclosed in accordance with some embodiments. In some embodiments, a process for performing query-dependent and content-class based ranking includes receiving a query for a search for content on a web site; performing a query-dependent and content-class based ranking of content available on the web site (e.g., using a processor); and returning a ranked list of content based at least in part on the query-dependent and content-class based ranking of content available on the web site in response to the query.
Abstract: A system for extracting attributes can analyze text from data sources, extract n-grams from the text as candidate attribute and service/product pairs, prompt a human operator to rate the suitability of the candidate attribute and service/product pairs, and, based on the ratings, add the candidate attribute and service/product pairs to an attribute dictionary. In embodiments, an attribute extraction system includes an n-gram extractor, an attribute selector interface, and a dictionary builder. Data sources may include product titles, category descriptions, product descriptions, and like data from one or more product databases. In embodiments, the attribute dictionary is analyzed to determine canonical names for products or services and name variants for the products or services.
Type:
Grant
Filed:
September 9, 2016
Date of Patent:
October 15, 2019
Assignee:
BloomReach, Inc.
Inventors:
Mohammad Salim Ahmed, Antariksh Bothale, Ailian Gan, Warren Mar, Ramkumar Rajendran, Sneha Venkatesh Yelimeli
Abstract: Techniques for providing control selection and analysis of Search Engine Optimization (SEO) activities for web sites for web service providers are disclosed. In some embodiments, a system, process, and/or computer program product for providing control selection and analysis of SEO activities for web sites includes selecting a first plurality of web pages associated with a web site for use in a control group of web pages for an experiment; selecting a second plurality of web pages associated with the web site for use in a test group of web pages for the experiment; canonicalizing the control group of web pages and the test group of web pages; and performing the experiment to determine a performance of the test group of web pages relative to the control group of web pages. For example, the test group of web pages can be configured with an SEO content improvement activity.
Type:
Grant
Filed:
June 29, 2015
Date of Patent:
August 27, 2019
Assignee:
BloomReach Inc.
Inventors:
Hai Dang, Rong Pan, Niran Kundapur, Pradeep Muthukrishnan
Abstract: A system for evaluating web pages for searchable content can be utilized to make an e-commerce search engine more effective by identifying pages with searchable value. In embodiments, a web page exhibiting “searchable value” is a page that provides useful information responsive to a user's query on an e-commerce search engine. One embodiment of a page evaluation system includes a web crawler module, a page analysis module, and a page index module. Web pages may be fetched and evaluated for various signals that indicate whether the web page exhibits searchable value. Examples of signals that may indicate that a web page lacks searchable value include a product page that does not have a product name nor a product image and a category page that does not have products listed in the page content. Web pages that lack searchable value may be excluded from subsequent e-commerce search results.