Implementations of the present specification provide a digital avatar recommendation method and recommendation system. The digital avatar recommendation system includes a computer-simulated digital avatar, and the corresponding recommendation method includes: obtaining current state data, where the state data includes user information of a target user, scenario information of a current scenario, and history information of an interaction between the target user and the digital avatar; mapping, by an agent in the digital avatar, the state data to a target action in a candidate action set based on a current policy obtained through reinforcement learning, where a candidate action in the candidate action set corresponds to a to-be-recommended content category, and the target action corresponds to a target content category; and performing, by the digital avatar, target interaction with the target user, where the target interaction is used to recommend the target content category. As such, individualized recommendation is provided for the target user by using the digital avatar.

Examples provide a system for crowdsourced item selection. A request fulfillment component selects a set of one or more items from a plurality of requested items for retrieval by an identified user. A set of items are selected based on an analysis of item data, per-user retrieval score(s), user-provided data, and/or per-user restriction(s). The request fulfillment component analyzes sensor data associated with a set of retrieved items received at an intake area to identify the items retrieved by the identified user and/or a level of compliance with retrieval criteria for the set of items. The request fulfillment component assigns a reward to the identified user, one or more retrieval scores ranking quality of items retrieved by the user, and/or one or more retrieval restrictions are assigned to the user based on the number of retrieved items and the level of compliance with the retrieval criteria.

Some embodiments of the invention provide tools for promoting a user's financial well-being by encouraging positive financial behavior through timely feedback, counseling and instruction. For example, some embodiments of the invention may provide real-time (e.g., in response to a transaction being performed, or in response to the user indicating he/she is contemplating a transaction) useful analysis and advice designed to encourage the type of ongoing financial behavior that leads over time to financial well-being.

A system and method for providing shopping information are provided. The system includes a survey product selector for selecting a product selected by a product registrant as a survey product, a preference score collector for collecting preference scores for the survey product from a plurality of survey participants among users, a winner selector for calculating the average of the preference scores, and selecting at least one survey winner from among the plurality of survey participants based on the average, and a winner reward decider for determining a reward to be offered to the at least one survey winner.

Embodiments of the present disclosure provide mechanisms for selection of a user survey data object from a set of user data objects, and processing of survey engagement data associated with a selected user survey data object. The user survey data object selected is appropriate for providing associated with a particular user data object, and the survey engagement data received associated therewith enables programmatic selection and use of particular ranking model(s) for use in generating and providing an output ranked item data object set. Example embodiments utilize selected ranking model(s) of a set of ranking models to programmatically generate and output an output ranked item data object set for a particular user profile.

Techniques are disclosed for determining application experience of a user. One embodiment presented herein includes a computer-implemented method, which includes receiving, at a computing device, eye tracking data of a user interacting with at least a first page of an application. The computer-implemented method further includes determining, based at least on the eye tracking data, at least a current user experience regarding the first page. The computer-implemented method further includes predicting, based on evaluating the current user experience, that the user is likely to discontinue use of the application. The computer-implemented method further includes determining, based at least on the prediction, an intervention that reduces a likelihood of the user discontinuing use of the application, and interacting with the user according to the intervention.

A method performed by a computing system that has access to a mapping of a plurality of user intentions to a plurality of services. The method includes receiving a communication from a computing device operated by a user and identifying one of the plurality of user intentions based at least in part on the communication. The computing system uses the mapping to identify one of the plurality of services mapped to the identified user intention and provides the identified service.

A method includes receiving a request and request data associated with a user from a web server and analyzing the request data to identify one or more data gaps associated with the request. One or more third-party services are called to fill at least a portion of the one or more data gaps. A question set is prepared based on determining that the one or more data gaps remain at least partially unfilled. The question set is selected by a machine-learning component trained to adapt a sequence and content of the question set over a plurality of interactions with a plurality of users. The question set is transmitted to the web server for presentation to the user. Data exchanges can be authenticated using tokens.

The integrity of data (e.g., review data) provided via a computer network (e.g., the Internet) and relied upon by online consumers when making purchase decisions is assured using a blockchain protocol that conducts a machine learning enabled validation flow on a public blockchain with reward mechanisms designed to weed out fake and incentivized reviews. A machine learning (ML) annotator evaluates a review and provides a score that is passed to human validators along with annotations for validation. The human validators stake tokens against the annotated reviews in a proof-of-stake consensus model that provides financial incentives to the human validators to provide an increased number of trustworthy reviews. The review protocol is implemented via smart contracts that are stored and replicated on a blockchain network and that are used to facilitate, verify, and enforce the negotiation or performance of the terms of an agreement between the participants and the review protocol.

Some embodiments of the invention provide tools for promoting a user's financial well-being by encouraging positive financial behavior through timely feedback, counseling and instruction. For example, some embodiments of the invention may provide real-time (e.g., in response to a transaction being performed, or in response to the user indicating he/she is contemplating a transaction) useful analysis and advice designed to encourage the type of ongoing financial behavior that leads over time to financial well-being.