Abstract: Techniques performed by a data processing system for analyzing the lexical difficulty of words of textual content include analyzing plurality of textual content sources to determine a first frequency at which each of a plurality of first words appears, analyzing search data to determine a second frequency at which each of the plurality of first words appear in searches for a definition, generating a lexical difficulty model based on the first frequency and the second frequency, the model is configured to receive a word as an input and to output a prediction for how difficult the word is likely to be for a user, receiving a request to analyze first textual content from a client device, analyzing the first textual content using the lexical difficulty model to generate lexical difficulty information, and sending a response to the client device that includes requested information.

Abstract: A method and system for providing a personalized user interface and personalized search suggestions when using an application is disclosed. The method includes receiving a request for providing a personalized user interface such as a personalized menu for an application and examining a history of the user's usage of the application, a history of other user's usage of the application, and/or the content of the current document in the application, before identifying one or more suggested options offered by the application based on one or more of these factors. A list may then be generated and provided for being displayed as part of the personalized user interface. Upon receiving an indication of a user's intent to perform a search within the application, the user's usage history, usage history of users with similar activities, and/or content of the current document may be examined to provide the user with a list of search suggestions.

Abstract: Non-limiting examples of the present disclosure describe surfacing a representation of digital artifacts for a user. An operation context associated with a user is identified. The operational context may correspond with an electronic document that is accessed through a productivity application. One or more of a plurality of digital artifacts may be identified based on the identified operational context. An exemplary representation of the one or more digital artifacts is surfaced through a user interface. An exemplary user interface may be a user interface of a productivity application, for example, where the electronic document is being accessed through the productivity application. In alternative examples, an exemplary user interface may be associated with a home service that is executed in association with an operating system (OS). In some instances, an operational context may change resulting in an update to an exemplary representation of digital artifacts. Other examples are also described.

Abstract: A method and system for providing a personalized user interface and personalized search suggestions when using an application is disclosed. The method includes receiving a request for providing a personalized user interface such as a personalized menu for an application and examining a history of the user's usage of the application, a history of other user's usage of the application, and/or the content of the current document in the application, before identifying one or more suggested options offered by the application based on one or more of these factors. A list may then be generated and provided for being displayed as part of the personalized user interface. Upon receiving an indication of a user's intent to perform a search within the application, the user's usage history, usage history of users with similar activities, and/or content of the current document may be examined to provide the user with a list of search suggestions.

Abstract: Non-limiting examples of the present disclosure describe surfacing a representation of digital artifacts for a user. An operation context associated with a user is identified. The operational context may correspond with an electronic document that is accessed through a productivity application. One or more of a plurality of digital artifacts may be identified based on the identified operational context. An exemplary representation of the one or more digital artifacts is surfaced through a user interface. An exemplary user interface may be a user interface of a productivity application, for example, where the electronic document is being accessed through the productivity application. In alternative examples, an exemplary user interface may be associated with a home service that is executed in association with an operating system (OS). In some instances, an operational context may change resulting in an update to an exemplary representation of digital artifacts. Other examples are also described.

Abstract: Topical search engines can add contextual keywords to an input query to bias results toward a particular topic or domain. In one instance, query context models can be constructed to facilitate topical search. Upon receipt of one or more topic-relevant sites, a plurality of topical queries can identified automatically. Contextual keywords can be identified with respect to the plurality of topical queries as a function of lexical generality, among other things. Subsequently, a query context model, comprising the identified topical queries and related contextual keywords, can be employed to restrict query results to a particular topic.

Abstract: Described is the presenting of suggested queries for web pages that are not search engine results pages, based upon the URL and/or content of a currently displayed page. The suggested query set may be dynamically extracted (locally or remotely) based upon the content of the web page, and/or obtained from a data store of per-URL suggested query sets, e.g., generated from historical logs. Also described are various techniques for generating suggested queries, and user interface mechanisms that display and allow interaction with suggested queries.

Abstract: Described is a technology by which phishing-related data sources are processed into aggregated data and a given site evaluated the aggregated data using a predictive model to automatically determine whether the given site is likely to be a phishing site. The predictive model may be built using machine learning based on training data, e.g., including known phishing sites and/or known non-phishing sites. To determine whether an object corresponding to a site is likely a phishing-related object are described, various criteria are evaluated, including one or more features of the object when evaluated. The determination is output in some way, e.g., made available to a reputation service, used to block access to a site or warn a user before allowing access, and/or used to assist a hand grader in being more efficient in evaluating sites.

Abstract: Described is a search technology in which a search engine constructs a results page for a query that integrates suggested queries with the individual query results (e.g., displayed URLs). When rendered, the proximity of the suggested queries to their corresponding individual query result provides context as to the specific URL to which the suggested query is related. Suggested queries may appear alongside their associated search result, e.g., a displayed URL, and/or in an expandable panel proximate that individual search result. Suggested queries may appear within text accompanying a URL, and/or in a drop down menu following interaction with such text or the like. Related queries may be found by using a search result URL to find a query, by analyzing a search result's text snippet, by accessing historical data, and/or by accessing current user session data.

Abstract: Topical search engines can add contextual keywords to an input query to bias results toward a particular topic or domain. In one instance, query context models can be constructed to facilitate topical search. Upon receipt of one or more topic-relevant sites, a plurality of topical queries can identified automatically. Contextual keywords can be identified with respect to the plurality of topical queries as a function of lexical generality, among other things. Subsequently, a query context model, comprising the identified topical queries and related contextual keywords, can be employed to restrict query results to a particular topic.

Abstract: Described is the presenting of suggested queries for web pages that are not search engine results pages, based upon the URL and/or content of a currently displayed page. The suggested query set may be dynamically extracted (locally or remotely) based upon the content of the web page, and/or obtained from a data store of per-URL suggested query sets, e.g., generated from historical logs. Also described are various techniques for generating suggested queries, and user interface mechanisms that display and allow interaction with suggested queries.

Abstract: Described is a search technology in which a search engine constructs a results page for a query that integrates suggested queries with the individual query results (e.g., displayed URLs). When rendered, the proximity of the suggested queries to their corresponding individual query result provides context as to the specific URL to which the suggested query is related. Suggested queries may appear alongside their associated search result, e.g., a displayed URL, and/or in an expandable panel proximate that individual search result. Suggested queries may appear within text accompanying a URL, and/or in a drop down menu following interaction with such text or the like. Related queries may be found by using a search result URL to find a query, by analyzing a search result's text snippet, by accessing historical data, and/or by accessing current user session data.

Abstract: The system and method of the present invention automatically extracts the top k recommendations of objects, such as topics, items, products, books, movies, food, drinks, etc., from a local probabilistic recommendation system. Unlike prior systems, the present invention accomplishes the extraction of the top k recommendations of objects without examining a probability for every object that can be recommended. Further, the system and method of the present invention is capable of being implemented using probabilistic recommendation systems based on any conventional type of probabilistic distribution or machine learning technique, including, for example, decision trees and Bayesian networks.

Abstract: The system and method of the present invention automatically assigns “scores” to the predictor/variable value pairs of a conventional probabilistic model to measure the relative impact or influence of particular elements of a set of topics, items, products, etc. in making specific predictions using the probabilistic model. In particular, these scores measure the relative impact, either positive or negative, that the value of each individual predictor variable has on the posterior distribution of the target topic, item, product, etc., for which a probability is being determined. These scores are useful for understanding why each prediction is made, and how much impact each predictor has on the prediction. Consequently, such scores are useful for explaining why a particular prediction or recommendation was made.

Abstract: The invention relates to a system for filtering messages—the system includes a seed filter having associated therewith a false positive rate and a false negative rate. A new filter is also provided for filtering the messages, the new filter is evaluated according to the false positive rate and the false negative rate of the seed filter, the data used to determine the false positive rate and the false negative rate of the seed filter are utilized to determine a new false positive rate and a new false negative rate of the new filter as a function of threshold. The new filter is employed in lieu of the seed filter if a threshold exists for the new filter such that the new false positive rate and new false negative rate are together considered better than the false positive and the false negative rate of the seed filter.

Abstract: The present invention involves a system and method that facilitate identifying human interaction by utilizing HIPs such as order-based HIPs and determining a difficulty rating of any type of HIPs in an automated fashion. Order-based HIPs require a user to identify elements in the sequence as well as to identify a correct order of the elements in the sequence. The invention involves presenting a user with at least two HIPs such that the HIP can be of known and/or unknown difficulty. A user that correctly answers the HIP of known difficulty gains access to the HIP-controlled resource, action or application. The user's response to the HIP of unknown difficulty can then be examined and employed to determine whether that HIP is too difficult for humans to solve. Alternatively, at least one HIP can be presented. Difficulty of individual HIP parameters can also be determined.

Abstract: The subject invention provides for a feedback loop system and method that facilitate classifying items in connection with spam prevention in server and/or client-based architectures. The invention makes uses of a machine-learning approach as applied to spam filters, and in particular, randomly samples incoming email messages so that examples of both legitimate and junk/spam mail are obtained to generate sets of training data. Users which are identified as spam-fighters are asked to vote on whether a selection of their incoming email messages is individually either legitimate mail or junk mail. A database stores the properties for each mail and voting transaction such as user information, message properties and content summary, and polling results for each message to generate training data for machine learning systems. The machine learning systems facilitate creating improved spam filter(s) that are trained to recognize both legitimate mail and spam mail and to distinguish between them.

Abstract: A unique recording system and method that facilitates recording live meetings, discussions or conversations whereby such recordings are available for immediate or near immediate playback is provided. As a result, a user who has momentarily become distracted or inattentive during the meeting can quickly re-listen to what was missed or misunderstood in order to readily catch up to the current discussion. The current discussion can continue to be recorded during playback of any previously recorded data. User behavior can be monitored to estimate when the user has started to become inattentive and likely segments or time points of the recordings can be suggested for playback. One or more portions of the recordings can be filtered or selected for playback so that any desired content can be eliminated or skipped in the playback version.

Abstract: Described is a technology by which phishing-related data sources are processed into aggregated data and a given site evaluated the aggregated data using a predictive model to automatically determine whether the given site is likely to be a phishing site. The predictive model may be built using machine learning based on training data, e.g., including known phishing sites and/or known non-phishing sites. To determine whether an object corresponding to a site is likely a phishing-related object are described, various criteria are evaluated, including one or more features of the object when evaluated. The determination is output in some way, e.g., made available to a reputation service, used to block access to a site or warn a user before allowing access, and/or used to assist a hand grader in being more efficient in evaluating sites.

Abstract: The present invention involves a system and method that facilitate extracting data from messages for spam filtering. The extracted data can be in the form of features, which can be employed in connection with machine learning systems to build improved filters. Data associated with origination information as well as other information embedded in the body of the message that allows a recipient of the message to contact and/or respond to the sender of the message can be extracted as features. The features, or a subset thereof, can be normalized and/or deobfuscated prior to being employed as features of the machine learning systems. The (deobfuscated) features can be employed to populate a plurality of feature lists that facilitate spam detection and prevention. Exemplary features include an email address, an IP address, a URL, an embedded image pointing to a URL, and/or portions thereof.