Methods and systems for dynamically generating real-time recommendations
AssigneeWisdo Ltd.
InventorsBoaz Gaon, Ido Jonathan Engel, Arik Gilon, Arie Gofer
Life of this patent
Estimating the term…
Abstract
Systems and methods are provided herein for generating personalized timeline-based feeds to a user. A computer-implemented method for generating feeds to a user may be provided. The method may include generating a timeline comprising a plurality of milestones and needs associated with an event, and providing the feeds based on community wisdom. The feeds may be provided for each milestone on the time-line specific to the user, and may be configured to address the user's needs at each milestone.
Claims30
A computer-implemented system for assisting a plurality of users in navigating one or more life events, the system comprising:
- (1) a matching engine configured to, with aid of one or more processors:
- receive input data from a plurality of computing devices associated with a plurality of users, wherein the input data comprises queries and insights from different users relating to the one or more life events;
- identify the type of life event(s) and a plurality of milestones that one or more users are currently experiencing, have experienced, or are likely to experience, by using a natural language processing (NLP) algorithm for multiclass and multilabel categorization to extract features and derive meaning from the users' queries and insights associated with the life event(s);
- and
- match the insights with the queries to the plurality of milestones located along a plurality of timelines for the one or more users;
- and
- (2) a recommendation engine configured to, with aid of one or more processors:
- predict needs of the one or more users at the plurality of milestones along the plurality of timelines, by analyzing the matched insights and queries to predict the users' needs;
- and
- generate personalized recommendations to the one or more users based on the users' predicted needs at the plurality of milestones as the users undergo the life event(s),
- wherein the timelines, milestones, queries, insights, and/or personalized recommendations are configured to be displayed as a set of visual objects on graphical displays of the computing devices associated with the one or more users.
A computer-implemented method for assisting a plurality of users in navigating one or more life events, the method comprising:
- receiving input data from a plurality of computing devices associated with a plurality of users, wherein the input data comprises queries and insights from different users relating to the one or more life events;
- identifying the type of life event(s) and a plurality of milestones that one or more users are currently experiencing, have experienced, or are likely to experience, by using a natural language processing (NLP) algorithm for multiclass and multilabel categorization to extract features and derive meaning from the users' queries and insights associated with the life event(s);
- matching the insights with the queries to the plurality of milestones located along a plurality of timelines for the one or more users;
- predicting needs of the one or more users at the plurality of milestones along the plurality of timelines, by analyzing the matched insights and queries to predict the users' needs;
- and
- generating personalized recommendations to the one or more users based on the users' predicted needs at the plurality of milestones as the users undergo the life event(s),
- wherein the timelines, milestones, queries, insights, and/or personalized recommendations are configured to be displayed as a set of visual objects on graphical displays of the computing devices associated with the one or more users.
A tangible computer readable medium storing instructions that, when executed by one or more servers, causes the one or more servers to perform a computer-implemented method for assisting a plurality of users in navigating one or more life events selected from a group consisting of (i) diagnosis with a terminal illness, (ii) death, (iii) marriage, (iv) divorce and (v) retirement, the method comprising:
- with aid of a matching engine:
- receiving input data from a plurality of computing devices associated with a plurality of users, wherein the input data comprises queries and insights from different users relating to the one or more life events;
- identifying the type of life event(s) and a plurality of milestones that one or more users are currently experiencing, have experienced, or are likely to experience, by using a natural language processing (NLP) algorithm for multiclass and multilabel categorization to extract features and derive meaning from the users' queries and insights associated with the life event(s);
- and
- matching the insights with the queries to a plurality of milestones located along a plurality of timelines for one or more users;
- and
- with aid of a recommendation engine:
- predicting needs of the one or more users at the plurality of milestones along the plurality of timelines, by analyzing the matched insights and queries to predict the users' needs;
- and
- generating personalized recommendations to the one or more users based on the users' predicted needs at the plurality of milestones as the users undergo the life event(s).
Dependencies are inferred from the claim text.