Microsoft Technology Licensing, LLC

Walter Sun

Junaid Ahmed

Sasi Parthasarathy

Xiao Lin Zhang

Jingfeng Li

Yipeng Li

Peter Bailey

Abhishek Kumar Singh

Nikhil Dandekar

Jay Goyal

Yinzhe Yu

Xin Chen

Yatharth Saraf

Lin Guo

Kumar Hemachandra Chellapilla

Pratibha Permandla

Vijay Krishna Narayanan

Mayank Shrivastava

Vinay Deshpande

Gagan Bansal

Paul Alexander Dow

Fabrice Canel

Thomas Francis McElroy

Vishnu Challam

Debraj GuhaThakurta

Yajie Siamwalla

Amita Surendra Gajewar

Konstantin Golyaev

Thomas Arthur Ledbetter

Brahm Kiran Singh

A multi-horizon predictor system that predicts a future parameter value for multiple horizons based on time-series data of the parameter, external data, and machine-learning. For a given time horizon, a time series data splitter splits the time into training data corresponding to a training time period, and a validation time period corresponding to a validation time period between the training time period and the given horizon. A model tuner tunes the prediction model of the given horizon fitting an initial prediction model to the parameter using the training data thereby using machine learning. The model tuner also tunes the initial prediction model by adjusting an effect of the external data on the prediction to generate a final prediction model for the given horizon using the validation data. A multi-horizon predictor causes the time series data splitter and the model tuner to operate for each of multiple horizons.

Predictive modeling across multiple horizons combining time series and external data

Methods, systems, and computer-readable media are provided for updating a search index with new uniform resource locators (URLs) and spiking URLs with increased user interest. History data, provided from browser applications residing on users' computers that indicate URLs accessed by the users, is parsed to identify new/previously unknown URLs. The history data also indicates URLs in which there is increased interest based on a number of recent hits as compared to an average number of hits determined over time. Author postings of new URLs to social networking sites and a quality rating of the authors may also be used to identify and filter new URLs. Search indexes are updated with the new and spiking URL data. As such, lag time between posting of new URLs and spiking of URL interest and inclusion of this data in a search index is greatly decreased.

Updating a search index using reported browser history data

Methods, systems, and media are provided for delivering clustered search results for recent and non-recent events by maintaining the identification (ID) numbers of the respective clustered documents beyond the 'fresh' life span of the clustered documents. When clusters are formed according to similar content, an ID number and associated attributes are assigned to each of the clusters. This provides a mechanism to track and retrieve the respective clusters for subsequent delivery of search results. The respective ID numbers of the clusters are maintained, even after the documents are no longer considered 'fresh.' These similar-content clusters are further subdivided according to publication date. This provides individual subdivided clusters for similar content events that occurred at different time spans, which are delivered along with individual non-clustered search results in a SERP.

Clustering web pages on a search engine results page

Methods, systems, and computer-readable media are provided for updating a search index with new uniform resource locators (URLs) and with metadata for new and known URLs. Data associated with communications made by users using a social network is received. The communications include a URL therein that a user has shared, posted, or otherwise communicated to one or more other users using the social network. When the URL is not found in a search index it is identified as a new URL and is added to the search index. A measure of a trending interest, or virality, of the URL is determined from the data. The determined virality is associated with the URL in a search index as metadata. The virality is useable to inform a ranking of the URL against a plurality of other URLs for identification and presentation as a search result in a search engine results page.

Using social-network data for identification and ranking of URLs

Methods, systems, and computer-storage media for improving the freshness, or the apparent freshness, of search results are described. In an embodiment, the first portion of search results presented on a search results page are based on responsiveness to the search query and a second portion of results describe only recently published documents that are responsive to the search query. In an embodiment, a more recent version of the document, which is not directly used to determine responsiveness, is used to build the caption for a search result. Another way to make search results appear fresh is to include a publication time within the search result caption. In one embodiment, the publication time is generated by calculating a point in time between when a document is first added to a search index and the previous time the search engine visited the site where the document was found.

Enhancing freshness of search results

A client application installed on end user computers generates metadata from the content of web pages visited by end users and provides the metadata to a search engine. When an end user visits a web page, the end user's computer downloads and displays the web page to the end user. The client application may simultaneously access the web page content and generate this metadata in the form of a content signature of the web page from the web page content. The client application then provides the content signature to a search engine. The search engine may employ content signatures to identify new web pages to crawl and index. Additionally, the search engine may employ content signatures to identify changes to web pages and determine the crawl frequency of web pages.

Content signature notification

Methods and systems for determining an adaptive crawl rate for a Web crawler based on historical publication data from a Web source are provided. A frequency of publication of the Web source is determined over a specified period of time, and an adaptive crawl rate is calculated using the frequency of publication. The Web crawler is then deployed at the calculated adaptive crawl rate.

Adaptive crawl rates based on publication frequency

Systems and methods for applying user behavior data to improve search query result ranking are provided. Upon receiving an update file indicating that recent, significant user behavior data is available for a document associated with an inverted index, the update file is published periodically and frequently to an index server. After filtering out the relevant update information from the update file, the index server extracts identifiers of the documents having the associated user behavior data. The update file and the identifier of the documents are utilized to update an in-memory index containing representations of metadata indicative of the user behavior. The in-memory index is continuously updated and utilized to serve search query results in response to user search queries. Search query results from the in-memory index are ranked using the user behavior data prior to serving. Thus, results associated with recent, significant user-behavior metadata receive prominent placement on the search results page.

Using behavior data to quickly improve search ranking

Growing community of inventors

Bellevue, WA, United States of America

Walter Sun