Microsoft Technology Licensing, LLC

Karan Khanna

Edward A Martinez

Triptpal Singh Lamba

Javier Nisim Flores Assad

Saumaya Sharma

Darya Mazandarany

Scott Michael Bragg

Vacha Rajendra Dave

Scott A Inglis

Rajvardhan Virendra Oak

Fatima Taher Calcuttawala

A technique uses a graph neural network (GNN) to determine whether a particular entity under consideration is engaging in abusive network-related activity over a computing network in collaboration with other entities. In some applications, the particular entity is part of a bot attack aimed at fraudulently engaging with advertisements. The technique trains the GNN by performing machine learning on a training set that includes a plurality of nodes, edges, and node labels. In forming the training set, the technique associates a feature set with each node in the training set that describes the network activity exhibited by that node's entity. The technique then connects each pair of nodes in the training set with an edge if the feature sets of the pair satisfy a prescribed test for similarity. The technique assigns labels to at least some nodes to convey whether the nodes are associated abusive network-related activity.

Detecting abusive collaborative activity using a graph neural network

Various technologies described herein pertain to delivering shared WiFi credentials. According to various embodiments, at least one server computing device is configured to transmit shared WiFi credentials to a mobile device. The at least one server computing device receives a request, where the request includes data that identifies the mobile device. Moreover, the request indicates that the mobile device desirably receives credentials for a shared WiFi network. Responsive to receiving the request, the at least one server computing device determines that the mobile device is authorized to receive the credentials for the shared WiFi network based at least in part upon the data that identifies the mobile device. Moreover, the at least one server computing device transmits the credentials for the shared WiFi network to the mobile device responsive to determining that the mobile device is authorized to receive the credentials for the shared WiFi network.

Delivery of shared WiFi credentials

A method and system for assisting a user in constructing a formula in a cell of a spreadsheet operating on a computer system. A function dialog box is displayed that lists various functions that may be selected by the user to be entered into the formula. By selecting a button in the function dialog box, an insert function dialog box is displayed that lists additional functions by category. When the user selects a function, a parameter dialog box associated with the selected function is displayed. Parameters may include cell addresses as well as numerical or other data entered directly by the user. The parameter dialog box displays the formula as it is being constructed. After the parameters have been entered, a result dialog box is displayed that allows the user to select the result cell in which to place the formula. The user selects the result cell and the result dialog box is removed from the display screen, the formula is entered in the result cell, and the result of the formula is displayed in the result cell.

Method and system for constructing a formula in a spreadsheet

In association with a computer system, a method and system for detecting and correcting errors in a spreadsheet formula. A token is created corresponding to each entry in the formula. As the formula is constructed, each token is added to the top of a formula token stack. When the formula is completed, a parser parses the formula token stack starting at the bottom of the stack and parsing each subsequent token. By examining the tokens in the formula token stack and their relationship to other tokens, the system determines whether an error exists in the formula. If an error is found to exist in the formula, an error class code is assigned to the error corresponding to the type of error. An error corrector rule corresponding to the error class code is applied and identifies a correction for the error. After a correction for the error is identified, the remainder of the formula token stack is parsed to detect and correct any other errors. When the entire formula token stack has been parsed and all detected errors have been corrected, a dialog box is displayed containing the completely corrected formula.

Method and system for detecting and correcting errors in a spreadsheet

Growing community of inventors

Clyde Hill, WA, United States of America

Karan Khanna