Abstract: Various embodiments discussed herein relate to prompting a model, such as a Large Language Model (LLM), to ingest natural language clustering instructions and generate corresponding natural language clustering information, such as a cluster description and/or a cluster label without the need to generate any numeric text embeddings.

Abstract: Methods, computer systems, computer-storage media, and graphical user interfaces are provided for efficiently generating and using intent taxonomies. In embodiments, training data, including data requests for information, is obtained. Thereafter, a model prompt to be input into a large language model is generated. The model prompt includes an instruction to generate an intent taxonomy, an indication of the training data to use for generating the intent taxonomy, and a taxonomy attribute desired to be used as criteria to generate a quality intent taxonomy. An intent taxonomy that includes user intent classes is obtained as output from the large language model. The intent taxonomy is analyzed to determine whether the intent taxonomy is valid. When the intent taxonomy is determined as valid, the intent taxonomy is provided for use in identifying user intent, and when the intent taxonomy is determined as invalid, the intent taxonomy is refined.

Abstract: A system and method for automatically generating a workplan schedule for a team of employees in an organization includes receiving a request to create a workplan schedule for the team, the team including two or more team members who are employees of the organization and the workplan schedule identifying a work location for the team members. One or more collaborators for one or more of the team members is detected, where the collaborator is an employee of the organization who is not a member of the team and with whom at least one of the team members collaborate.

Abstract: The disclosure relates to a sensitivity detection system that accurately and efficiently determines when information based on a user's browsing activity unintentionally reveals private or other sensitive information about the user. For example, the sensitivity detection system generates and utilizes machine learning models for detecting sensitivity to accurately detect when sensitive user information is being leaked from a collection of user information, such as a user profile. Additionally, upon determining that sensitive user information is being revealed, in many instances, the sensitivity detection system performs mitigation actions to stop and/or reduce sensitive user information from being undesirably revealed.

Abstract: Methods, computer systems, computer-storage media, and graphical user interfaces are provided for enhancing remote work productivity data. In one embodiment, worker data is obtained for a set of target workers and a set of control workers. The target workers are generally identified as working in the office prior to a treatment date and working remote after a treatment date, and the control workers are generally identified as working remote before and after the treatment date. Thereafter, an effect of working remotely is determined using a difference-in-differences model that measures differences in observational changes between the set of target workers and the set of control workers measured before and after the treatment date. The effect of working remotely can be used to generate a generating a productivity score indicating an extent of productivity in relation to the set of target workers.

Abstract: A device may receive a code set that includes a set of objects. The set of objects may include one or more successor objects and one or more predecessor objects corresponding to the one or more successor objects. The device may populate a data structure based on the code set. The data structure may identify each object of the set of objects. The device may identify relationships between the one or more successor objects and the one or more predecessor objects. The device may add, to the data structure and in association with the one or more predecessor objects, dependency information that identifies the relationships between the one or more successor objects and the one or more predecessor objects. The dependency information may identify the one or more successor objects. The device may perform an action based on the dependency information.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for analyzing impact of a computer-readable code on a database are enclosed. In some implementations, actions include: receiving a computer-readable code from a computing device, determining one or more logical sets based on one or more database functions, identifying one or more table names in the one or more logical sets, the one or more table names being identified based on at least one database function that is set as a table identifier, the table names referring to respective tables in the database, providing a list of table names comprising one or more of the table names, and transmitting the list of table names to the computing device for display to a user.

Abstract: A device may include one or more processors. The device may obtain a first code set and a second code set. The first code set may have at least one difference from the second code set. The at least one difference may relate to addition, removal, or modification of code of the first code set in comparison to code of the second code set. The device may identify one or more first portions of the first code set and one or more second portions of the second code set. The device may concurrently compare the one or more first portions and the one or more corresponding second portions to identify the at least one difference. The device may generate comparison information identifying the at least one difference. The device may store or provide the comparison information.

Abstract: A device may receive a code set that includes a set of objects. The set of objects may include one or more successor objects and one or more predecessor objects corresponding to the one or more successor objects. The device may populate a data structure based on the code set. The data structure may identify each object of the set of objects. The device may identify relationships between the one or more successor objects and the one or more predecessor objects. The device may add, to the data structure and in association with the one or more predecessor objects, dependency information that identifies the relationships between the one or more successor objects and the one or more predecessor objects. The dependency information may identify the one or more successor objects. The device may perform an action based on the dependency information.

Abstract: A device may include one or more processors. The device may obtain a first code set and a second code set. The first code set may have at least one difference from the second code set. The at least one difference may relate to addition, removal, or modification of code of the first code set in comparison to code of the second code set. The device may identify one or more first portions of the first code set and one or more second portions of the second code set. The device may concurrently compare the one or more first portions and the one or more corresponding second portions to identify the at least one difference. The device may generate comparison information identifying the at least one difference. The device may store or provide the comparison information.

Abstract: The amount of time that a web page is displayed by a browser is tracked. The amount that is charged to a content provider, whose content item is contained within the web page, is determined based on this amount of time. One way of tracking this amount of time is by tracking the amount of time that passes in between the moment that the web page is loaded by the web browser and the moment in which the web page goes out of focus. If the web page is displayed for less than a specified minimum amount of time during a first session, then the content item is displayed again during a second session. Once the content item has been displayed for a specified maximum amount of time, the content item is replaced with another content item on the web page.

Abstract: The amount of time that an advertisement-presenting web page is displayed by a browser is tracked. The amount of money that is charged to an advertiser, whose advertisement is contained within the web page, is determined based on this amount of time. One way of tracking this amount of time is by tracking the amount of time that passes in between the moment that the web page is loaded by the web browser and the moment in which the web page goes out of focus. If the web page is displayed for less than a specified minimum amount of time during a first session, then the advertisement is displayed again during a second session. Once the advertisement has been displayed for a specified maximum amount of time, the advertisement is replaced with another advertisement on the web page.

Abstract: A system and method which allow estimation of the value of display advertising are disclosed. Access is provided to a simulated web browsing experience to a plurality of users on a browser on the users' computers. The users have agreed to act as test subjects in exchange for financial compensation. At least one web page comprising a randomly selected article and at least one advertisement is presented to the users. After a predetermined duration, a questionnaire including questions relating to the at least one advertisement is presented to the users. The users are provided with financial compensation for answering the questions.

Abstract: Collaborative-filtering software creates a neighborhood using a map-reduce architecture by pair-wise application of a similarity measure to a sparse matrix of users and items of media designated by the users. The collaborative-filtering software then generates recommendations for a particular user by rating items of media designated by other users in the neighborhood (but not the particular user). The collaborative-filtering software rates the item of media, using a weighted vote of the users in the neighborhood. The weighted vote depends at least in part on the mean similarity of the users in the neighborhood who have designated the item of media. Then the collaborative-filtering software records the item of media as a recommendation for subsequent presentation to the other user, if the rating of the item of media is among the highest in comparison to the ratings of other items of media designated by users in the neighborhood.

Abstract: Collaborative-filtering software creates a neighborhood using a map-reduce architecture by pair-wise application of a similarity measure to a sparse matrix of users and items of media designated by the users. The collaborative-filtering software then generates recommendations for a particular user by rating items of media designated by other users in the neighborhood (but not the particular user). The collaborative-filtering software rates the item of media, using a weighted vote of the users in the neighborhood. The weighted vote depends at least in part on the mean similarity of the users in the neighborhood who have designated the item of media. Then the collaborative-filtering software records the item of media as a recommendation for subsequent presentation to the other user, if the rating of the item of media is among the highest in comparison to the ratings of other items of media designated by users in the neighborhood.