Methods, systems and apparatus for assessing the likely status of an operator of a computing device interacting with a server as a human operator or an autonomic computer application, such as a “bot” are described herein. By monitoring at least some data, e.g., biometric data, generated at the client computing device, a comparison can be made between the monitored data and model data relating to human interaction with the computing device. The results of the comparison can lead to a value that represents the likelihood that the monitored data results from human interaction.

A technique includes modifying a first record based on a plurality of training records to provide a plurality of modified records. The plurality of training records are associated with a regression model, and a target outcome is associated with the first record. The technique includes applying the regression model to the plurality of modified records to provide outcomes for the modified records; and selecting a group of the modified records based at least in part on the outcomes for the modified records. The technique includes providing a recommendation for the first record to change an outcome predicted for the first record by the regression model based at least part on the target outcome and the outcomes for the modified records.

A technique includes modifying a first record based on a plurality of training records to provide a plurality of modified records. The plurality of training records are associated with a regression model, and a target outcome is associated with the first record. The technique includes applying the regression model to the plurality of modified records to provide outcomes for the modified records; and selecting a group of the modified records based at least in part on the outcomes for the modified records. The technique includes providing a recommendation for the first record to change an outcome predicted for the first record by the regression model based at least part on the target outcome and the outcomes for the modified records.

A symbol prediction method includes storing a statistic for each of a set of symbols w in at least one context, each context including a string of k preceding symbols and a string of l subsequent symbols, the statistic being based on observations of a string kwl in training data. For an input sequence of symbols, a prediction is computed for at least one symbol in the input sequence, based on the stored statistics. The computing includes, where the symbol is in a context in the sequence not having a stored statistic, computing the prediction for the symbol in that context based on a stored statistic for the symbol in a more general context.

Embodiments described herein are directed to providing contextually relevant cues to users and to providing cues based on predicted conditions. In one scenario, a computer system identifies a task that is to be performed by a user. The computer system accesses data structures to identify current conditions related to the identified task. The computer system then generates, based on the identified current conditions related to the task, contextually relevant cues for the task. The contextually relevant cues provide suggestive information associated with the task. The computer system further provides the generated cue to the user. In other scenarios, the computer system identifies anticipated conditions related to the task using accessed historical information, and generates contextually relevant cues based on the identified anticipated conditions.

A method for executing a query includes determining one or more nodes that are likely to have local content that matches a search query. The determination is based on a location profile for each of the one or more nodes and a conditional probabilistic model for each of a set of distinct locations. The search query is executed at the one or more nodes.

An online system identifies an impression opportunity for a target user of the online system. The online system accesses predictors for a third party system, each predictor determining a prediction value indicating a likelihood of users to provide a specified benefit to the third party system after a specified timeframe from the performance of a specified type of action by the users at the online system, each predictor trained using a training feature set extracted from an impressions log including metadata for past impression opportunities made to users. The online system determines a combined bid value for the third party system based on prediction values determined by the predictors trained for the third party system. In response to determining that the combined bid value for the third party system is a winning bid value, the online system presents a sponsored content from the third party system to the user.

In one embodiment, a method includes one or more computing devices detecting a triggering action by a user of a social-networking system, wherein the detecting includes receiving information about the triggering action from a client device associated with the user and accessing a queue including multiple notifications. The method also includes, for each of one or more of the notifications, calculating using a machine-learning model, a click-through probability that the user will interact with the notification upon display of the notification, wherein the machine-learning model is based at least in part on one or more features associated with the user or the notification, determining whether the click-through probability satisfies a threshold, and if the click-through probability satisfies the threshold, then sending the notification to the client device associated with the user for display, else, removing the notification from the queue.

A method using a fast algorithm automated analysis of time series sensor data that can find an optimal clustering value k for k-means analysis by using statistical analysis of the results of clustering for a stated maximal upper value of k.

Methods, systems, and apparatus for prioritizing communications are described. Metadata that characterizes an electronic communication is obtained and a machine learning algorithm is applied to the metadata to generate a scoring model. A score for the electronic communication is generated based on the scoring model.