Abstract: A method for generating a model for recommendations from an item data set for a target data set includes embedding a set of targets from the target data set in a shared coordinate space using a first embedding function, embedding a first set of items from the item data set in the shared coordinate space using a second embedding function, selecting at least one target from the set of targets, and identifying a second set of items from the first set of items that are proximate to the at least one target as candidates from the recommendations.

Abstract: A set of data for training a machine learning system can be modified to improve its performance. An item of information can be transmitted. A message can be transmitted that includes an explanation of a determination, by the machine learning system, to transmit the item of information from among a plurality of items of information. A first set of data can have been used to train the machine learning system. A signal can be received that includes an indication of a usefulness of the message, to a user of a user device, in making a decision to perform an action based on a knowledge associated with the item of information. The first set of data can be modified, in response to a receipt of the signal, to produce a second set of data. The machine learning system can be caused to be trained using the second set of data.

Abstract: A set of data for training a machine learning system can be modified to improve its performance. An item of information can be transmitted. A message can be transmitted that includes an explanation of a determination, by the machine learning system, to transmit the item of information from among a plurality of items of information. A first set of data can have been used to train the machine learning system. A signal can be received that includes an indication of a usefulness of the message, to a user of a user device, in making a decision to perform an action based on a knowledge associated with the item of information. The first set of data can be modified, in response to a receipt of the signal, to produce a second set of data. The machine learning system can be caused to be trained using the second set of data.

Abstract: A method for generating a model for recommendations from an item data set for a target data set includes embedding a set of targets from the target data set in a shared coordinate space using a first embedding function, embedding a first set of items from the item data set in the shared coordinate space using a second embedding function, selecting at least one target from the set of targets, and identifying a second set of items from the first set of items that are proximate to the at least one target as candidates from the recommendations.

Abstract: A set of data for training a machine learning system can be modified to improve its performance. An item of information can be transmitted. A message can be transmitted that includes an explanation of a determination, by the machine learning system, to transmit the item of information from among a plurality of items of information. A first set of data can have been used to train the machine learning system. A signal can be received that includes an indication of a usefulness of the message, to a user of a user device, in making a decision to perform an action based on a knowledge associated with the item of information. The first set of data can be modified, in response to a receipt of the signal, to produce a second set of data. The machine learning system can be caused to be trained using the second set of data.

Abstract: A system and method of generating a preview of a file attachment includes receiving the file attachment in an email server. The file attachment can be received either during composition of the email or when receiving an inbound email. Determining if the file attachment is a previewable file type or compressed file type or non-previewable file type and automatically generating a preview of the file attachment including identifying the file attachment, parsing the file attachment to identify content of the file attachment, wherein the identified content includes a pagination of the file attachment. A file structure for each page of the file attachment is created and then saved to an attachment store in the email server.

Abstract: Instances of data associated with hindsight bias in a training set of data for a machine learning system can be reduced. A first set of data, having a first set of fields, can be received. Data in a first field can be analyzed with respect to data in a second field corresponding to an event to be predicted. A result can be that the data in the first field is associated with hindsight bias. A second set of data, having a second set of fields, can be produced. The second set of fields can exclude the first field. One or more features associated with the second set of data can be generated. A third set of data, having the second set of fields and fields that correspond to the one or more features, can be produced. The training set of data can be produced using the third set of data.

Abstract: Disclosed herein is an automated personalized message composition system, method and architecture. A composition style of a user is learned for each of a number of categories, such that each category has a corresponding composition style. The user's composition style determined for a given category can be used to personalize a non-personalized message for the user. A personalized message including elements of the user's composition style. The composition style elements from the user's composition style replacing a number of non-personalized elements in the non-personalized message.

Abstract: A system and method of generating a preview of a file attachment includes receiving the file attachment in an email server. The file attachment can be received either during composition of the email or when receiving an inbound email. Determining if the file attachment is a previewable file type or compressed file type or non-previewable file type and automatically generating a preview of the file attachment including identifying the file attachment, parsing the file attachment to identify content of the file attachment, wherein the identified content includes a pagination of the file attachment. A file structure for each page of the file attachment is created and then saved to an attachment store in the email server.

Abstract: Disclosed herein is an automated personalized message composition system, method and architecture. A composition style of a user is learned for each of a number of categories, such that each category has a corresponding composition style. The user's composition style determined for a given category can be used to personalize a non-personalized message for the user. A personalized message including elements of the user's composition style. The composition style elements from the user's composition style replacing a number of non-personalized elements in the non-personalized message.

Abstract: A non-transitory computer-readable memory may include instructions that, when executed by one or more processors, cause the one or more processors to perform operations including receiving a first requisition by a computer system of a procurement organization. The operations may also include receiving, by the computer system of the procurement organization, first SQL commands that generate a first change to be applied to a budget allocation associated with the first requisition. The operations may additionally include determining, by the computer system of the procurement organization, that the first change to the budget allocation does not violate a threshold. The operations may further include in response to determining that the first change does not violate the threshold, creating, by the computer system of the procurement organization, second SQL commands that generate a purchase order based on the first change.

Abstract: A method of modifying a requisition within approval and budgetary constraints by a procurement organization may include receiving the requisition from a requesting organization, where the requisition may comprise one or more lines, each having been approved by the requesting organization. The method may also include receiving a change to be applied to a first line associated with a first budget allocation, where the change may be received from a buyer in the procurement organization, and the change may be associated with a second budget allocation. The method may additionally include determining that the change violates a threshold, sending the change to the requesting organization for a second approval, receiving, from the requesting organization, an indication as to whether the second approval is granted, and creating a purchase order based on the indication as to whether the second approval is granted.

Abstract: Disclosed is a system and method for integrating search capabilities with message composition functionality. The present disclosure provides an intersection between messaging and search by enabling a user to search for content from differing entities and resources from within the body of a message compose window. The disclosed systems and methods enable a user, from within a messaging application or platform, to perform a search for content, view the results, and select and embed the desired result within the message in a rich format.

Abstract: Techniques are provided that include displaying an offer for an interest-based content subscription on an email application Web site, the offer being selectable and displayed in a native format of the email application, and, upon receiving an indication that the offer has been selected by a user, generating an interest-based content subscription without directing a browser application away from the email application Web site. According to some such arrangements, generating the interest-based content subscription may include contacting a third-party Web site associated with the offer, and communicating at least one of an actual identification and a disposable identification of the user to the third-party Web site. The interest-based content subscription may be limited based on at least one of a time duration and a frequency parameter.

Abstract: Embodiments of the present invention are directed to techniques (e.g., systems and methods) for management and tracking of distribution and schedules of a purchase order associated with a purchase requisition. In certain embodiments, a procurement management system can manage information and relationships between split or new distribution and/or schedule requirements for a purchase order in fulfillment of a purchase requisition. In particular, information about a purchase requisition, such as a requisition identifier (requisition ID), can be stored and used later to manage and track distribution and/or schedules for fulfillment of one or more purchase orders. Using information stored about a relationship between a purchase requisition and the distributions and the schedules for a purchase order, a procurement management system can ascertain details about distribution and schedules of a purchase order for fulfillment of a purchase requisition.

Abstract: A method of reinstating requisition demands using information from a canceled purchase order in an enterprise accounting computer system may include identifying canceled purchase order distributions in the canceled purchase order. The method may also include generating new requisition distributions based on the canceled purchase order distributions, and generating new requisition lines for the new requisition distributions. The method may further include generating one or more new purchase orders based on the new requisition lines.

Abstract: A method of modifying a requisition within approval and budgetary constraints by a procurement organization may include receiving the requisition from a requesting organization, where the requisition may comprise one or more lines, each having been approved by the requesting organization. The method may also include receiving a change to be applied to a first line associated with a first budget allocation, where the change may be received from a buyer in the procurement organization, and the change may be associated with a second budget allocation. The method may additionally include determining that the change violates a threshold, sending the change to the requesting organization for a second approval, receiving, from the requesting organization, an indication as to whether the second approval is granted, and creating a purchase order based on the indication as to whether the second approval is granted.

Abstract: A method of calculating accounting encumbrance adjustments for a purchase order associated with a first allocation of a budget includes determining that the purchase order has been closed, where the closure prevents additional processing associated with the purchase order by an invoice processing system; receiving a request to reopen the purchase order, where the request is associated with an additional cost; causing a determination to be made as to whether a second portion of the budget should be allocated, where the second portion of the budget corresponds to the additional cost; and in response to a determination that the second portion of the budget can be allocated: causing the second portion of the budget to be allocated, reopening the purchase order, and sending an indication to the invoice processing system that the purchase order is reopened.

Abstract: Embodiments of the present invention relate generally to methods and systems for processing purchase requisitions and purchase orders in a procurement system and more particularly systems and methods for returning single or individual requisition lines back to the requester without affecting other associated requisition lines under the same requisition. Generally speaking, embodiments allow users of a procurement system to return individual requisition lines instead of returning all the requisition lines in the purchase requisition. In some cases, policies require all or some subset of all lines of a requisition be processed together when possible. Hence, if there are more than one requisition lines, the buyer can be presented with a choice whether he or she would like to return the individual line or whether he would like to return all or some subset of all of the lines allowing a buyer to make a more informed decision.