Abstract: Various embodiments of this disclosure relate generally to generating a customized output based on account data of a user. The method comprises receiving, by one or more processors, user data from one or more databases, wherein the user data includes user browser history and user account data, creating a user record based on the user account data, wherein the user record includes one or more stocks, determining a stock subset of the one or more stocks, retrieving one or more relevant segments from a vector database, wherein the one or more relevant segments correspond to one or more relevant events that are relevant to the stock subset, generating a personalized output script based on the one or more relevant segments, creating an output based on the personalized output script, and displaying the output on one or more user interfaces of a user device.

Abstract: The present teaching relates to content categorization. Supervised training data and unlabeled data clusters are used to generate augmented training data. Each unlabeled data cluster includes data samples with varying features. Weakly labeled training data is created based on supervised training data and the unlabeled data clusters with data samples therein with cluster labels via consistent self-training so that a labeled data sample in the supervised training data and a data sample in the weakly labeled training data with the same label have varying characteristics. Augmented training data is created from the supervised and the weakly labeled training data and is used to train a robust content categorization model via machine learning.

Abstract: The present teaching relates to content categorization. Supervised training data and unlabeled data clusters are used to generate augmented training data. Each unlabeled data cluster includes data samples with varying features. Weakly labeled training data is created with new data samples generated via generative augmentation based on supervised training data and the unlabeled data clusters. Each new data sample is assigned a label from a corresponding data sample from the supervised training data with generated varying characteristics. Augmented training data is created from the supervised and the weakly labeled training data and is used to train a robust content categorization model via machine learning.

Abstract: The present teaching relates to displaying ads. A generative artificial intelligence (AI) model for creating advertisement assets is obtained, via machine learning, based on training data generated based on online feedback information on previously displayed advertisements. Base advertisement information associated with an advertisement of a product specifying some attributes characterizing the product is received. Using the generative AI model, multiple advertisement assets are created with respect to some attribute of the advertisement. Each advertisement asset is a representation of an attribute. These advertisement assets are used to form different asset combinations, each of which can be used to display the advertisement.

Abstract: The present teaching relates to online advertising. Bids directed to a display ad opportunity are received, where the display ad opportunity involves a user and an associated context and each bid includes a candidate advertisement. Auxiliary features are obtained for each bid based on a code generated by an autoencoder based on the bid and a predicted performance metric is determined for the candidate advertisement associated with the bid based on the auxiliary features associated with the bid. A winning advertisement is selected from candidate advertisements of the bids according to a ranking determined based on the respective predicted performance metrics of the candidate advertisements.

Abstract: Techniques for intelligently managing service requests using a service request outcome prediction and a dynamically determined probability threshold are disclosed.

Abstract: Disclosed are systems and methods for generating recommendations to users based on historical travel information and electronic communication data. The disclosed systems and methods provide a novel framework for automating the transmission of electronic travel-related recommendations to users by consistently monitoring electronic messages received at an electronic communication mailbox corresponding to a user. The disclosed framework operates by leveraging historical user data, data parsed from electronic communication mailbox corresponding to a user, or various vendor information, and using the aforementioned data as inputs for travel-related recommendation models, in order to generate and transmit the optimal travel-related recommendations to a user.

Abstract: Disclosed are systems and methods for improving interactions with and between computers in electronic messaging and/or providing systems supported by or configured with personal computing devices, servers and/or platforms. The disclosed systems and methods provide systems and methods for generating electronic message filters and for using electronic message filters comprising item category filtering criteria and having an automatically-determined expiration. The discloses systems and methods filter electronic messages using the item category filtering criteria while an electronic message filter remains active as determined using the automatically-determined expiration information.

Abstract: One or more computing devices, systems, and/or methods for generating dynamic content item recommendations are provided. Content item information, extracted from message data, is aggregated to calculate popularity and attributes of content items. The content items are ranked based upon the popularity and attributes to generate a ranked list of content items. Exploration traffic is served utilizing a set of eligible content items selected from the ranked list of content items. An eligible content item is promoted for participation in auctions for serving non-exploration traffic based upon the eligible content item being served a threshold number of times.

Abstract: One or more computing devices, systems, and/or methods for generating and providing recommendations of products are provided. For example, content is extracted from a message sent to a user. The content is evaluated to identify a product identifier corresponding to a product title of a product. If the product identifier is a truncated version of the product title, then a database of product titles and frequencies of occurrence of the product titles is used to complete the product title. A model is used to infer a product category for the product title. Matching scores are assigned to products within a product category based upon weighted attributes. A recommendation is provided to the user for a product having a matching score greater than a matching threshold.

Abstract: Disclosed are systems and methods for improving interactions with and between computers in content providing, searching and/or hosting systems supported by or configured with devices, servers and/or platforms. The disclosed systems and methods provide a novel framework for analyzing messages associated with an inbox of a user, and providing functionality to the inbox for alerting the user to specific forms of content included within each message. Such functionality can include displaying messages and/or the content contained therein in a novel, dedicated and modified and/or embedded portion of a message inbox. The messages can also be modified upon display within in the inbox to indicate that the message includes content of a specific type. Also, messages that include specific content types can also trigger the disclosed framework to generate and provide alerts to the user indicating the inbox has received or includes specific forms of such content.

Abstract: The present teaching relates to method, system, medium, and implementations for product recommendation. When communication content is received from a service provider operating on a platform, information related to a product is identified from a webpage accessed based on a link included in the communication content. Based on the information related to the product, a recommendation of the product is generated and sent to some service providers on different platforms for distribution of the recommendation to intended targets.

Abstract: Disclosed are systems and methods for generating recommendations to users based on historical travel information and electronic communication data. The disclosed systems and methods provide a novel framework for automating the transmission of electronic travel-related recommendations to users by consistently monitoring electronic messages received at an electronic communication mailbox corresponding to a user. The disclosed framework operates by leveraging historical user data, data parsed from electronic communication mailbox corresponding to a user, or various vendor information, and using the aforementioned data as inputs for travel-related recommendation models, in order to generate and transmit the optimal travel-related recommendations to a user.

Abstract: Disclosed are systems and methods for improving interactions with and between computers in electronic messaging and/or providing systems supported by or configured with personal computing devices, servers and/or platforms. The disclosed systems and methods provide systems and methods for generating electronic message filters and for using electronic message filters comprising item category filtering criteria and having an automatically-determined expiration. The discloses systems and methods filter electronic messages using the item category filtering criteria while an electronic message filter remains active as determined using the automatically-determined expiration information.

Abstract: One or more computing devices, systems, and/or methods for generating and providing recommendations of products are provided. For example, content is extracted from a message sent to a user. The content is evaluated to identify a product identifier corresponding to a product title of a product. If the product identifier is a truncated version of the product title, then a database of product titles and frequencies of occurrence of the product titles is used to complete the product title. A model is used to infer a product category for the product title. Matching scores are assigned to products within a product category based upon weighted attributes. A recommendation is provided to the user for a product having a matching score greater than a matching threshold.

Abstract: Disclosed are systems and methods for improving interactions with and between computers in content providing, searching and/or hosting systems supported by or configured with devices, servers and/or platforms. The disclosed systems and methods provide a novel framework for analyzing messages associated with an inbox of a user, and providing functionality to the inbox for alerting the user to specific forms of content included within each message. Such functionality can include displaying messages and/or the content contained therein in a novel, dedicated and modified and/or embedded portion of a message inbox. The messages can also be modified upon display within in the inbox to indicate that the message includes content of a specific type. Also, messages that include specific content types can also trigger the disclosed framework to generate and provide alerts to the user indicating the inbox has received or includes specific forms of such content.

Abstract: Disclosed are systems and methods for generating voice renderings of machine-generated electronic messages. The disclosed systems and methods provide a novel framework for organizing often fragmented machine-generated electronic messages and providing mechanisms for a virtual assistant to produce voice-renderings data extracted from electronic messages. The disclosed system may implement steps for receiving user queries via virtual assistants, extracting data from machine-generated electronic messages, converting the extracted data to purposeful organizational schemas, and generating human perceivable voice renderings based on the user queries and extracted data.

Abstract: One or more computing devices, systems, and/or methods for mailbox-based coupon display are provided. For example, a plurality of email messages stored in a mailbox of a first email account of a first user may be identified, and may be evaluated to identify a plurality of coupons. The plurality of coupons may be analyzed to determine one or more coupon features. User data and/or activity data associated with the first user may be analyzed to determine one or more user features. A predictive model may be tuned based upon the one or more coupon features and/or the one or more user features. A set of scores may be generated for the plurality of coupons based upon the predictive model. A graphical user interface may be controlled based upon the set of scores to display one or more coupons.

Abstract: Disclosed are systems and methods for improving interactions with and between computers in content providing, searching and/or hosting systems supported by or configured with devices, servers and/or platforms. The disclosed systems and methods provide a novel framework that provides an optimized ranking methodology for query suggestion and auto-completion tasks that messaging systems can implement and embed into their functionality. The disclosed framework can leverage the mail query logs on a personal level or on the aggregate (e.g., across all or a set of users on a network), and the mailbox content of each user to produce a comprehensive ranking function that generates ranked search suggestions for a mail search query for both person and company entities.

Abstract: One or more computing devices, systems, and/or methods for generating dynamic content item recommendations are provided. Content item information, extracted from message data, is aggregated to calculate popularity and attributes of content items. The content items are ranked based upon the popularity and attributes to generate a ranked list of content items. Exploration traffic is served utilizing a set of eligible content items selected from the ranked list of content items. An eligible content item is promoted for participation in auctions for serving non-exploration traffic based upon the eligible content item being served a threshold number of times.