| CPC H04L 51/42 (2022.05) [G06F 40/284 (2020.01); G06F 3/04886 (2013.01)] | 18 Claims |
|
1. A system for generating and rendering intent-based actionable content, the system comprising:
an electronic device comprising an input device, at least one hardware processor, a memory unit operably and communicatively coupled to the at least one hardware processor, and a display unit: and
an actionable content generation unit operably coupled to the at least one hardware processor and an input interface on the electronic device, wherein the actionable content generation unit is run and executed on the at least one hardware processor and configured to:
analyze an input message received from the input interface into plurality of words;
generate a plurality of tokens of the input message based on the plurality of words:
generate a static model comprising a criteria for mapping tokens, intent definitions, and a list of campaigns for one or more data access objects:
map the generated plurality of tokens of the input message with a plurality of intent definitions of a static model to identify an intent of the input message:
calculate a confidence level for the input message based on the mapped intent definitions of the plurality of tokens, and wherein the confidence level is any one of a no confidence level, a low confidence level, a medium confidence level, and a high confidence level; and
identity and assign at least one campaign to the input message based on the identified intent and the calculated confidence level;
wherein the at least one campaign is retrieved from a campaign data, and wherein the campaign data comprises definitions of intent and ongoing campaigns:
wherein the actionable content generation unit is run on the at least one hardware processor to generate, render, and invoke an intent-based actionable content on a user device in real time through an input interface, and wherein the actionable content generation is invoked within the input interface on the user device, independent of a user application, wherein the user application is a messaging application or chat application or e-mail application, and wherein the intent of a message is determined on real time using an artificial intelligence, and wherein each of the plurality of tokens is further assigned with an intent contributor and a confidence contributor, and wherein the actionable content generation unit creates a static model class for retrieval of the campaign data and wherein the actionable content generation unit creates the static model class for storing token mappings, intent definitions, and a list of campaigns from one or more data access objects, and wherein the actionable content generation unit executes the static model class on a separate thread for capturing an input operation of the keyboard, comparison, and determination of intent of each sentence of a message entered into the input field by a user, and wherein the actionable content generation unit employs a rule-based system for determining the intent of a message in real time while the user enters the message in an input field of the user application using the input interface invoked on the user device, and wherein the rule-based system includes a system of regular expressions (regex) used for comparing and determining an intent type and an intent name, and wherein the set of regular expressions is provided for intent detection and intent attribute extraction, and wherein mappings of tokens with intent definitions are stored in a definition file, and wherein the definition file contains a set of regular expressions, and a list of operating system application package names, and wherein the list of operating system application package names increases or boosts the confidence level by one level of the determined intent, and wherein the actionable content generation unit generates actionable content based on the determined intent using the campaign data, and wherein the campaign data comprises definitions of intent and ongoing campaigns, and wherein each campaign contains attributes to control how the campaign is rendered or displayed to the user, an action to be taken in response the user's click action, the set of applications in which the campaign is displayed, the set of intent type and intent attributes for which the campaign is displayed, a demographic target comprising age, gender, location of the campaign, prompt, text within prompt, and link to web site, and wherein the campaign further includes a cooling interval that provides a time difference between campaigns that are shown consecutively, the estimated confidence level when intents have confidence level above a threshold level, a display interval that provides a time interval for which the campaign is shown at one point of time, a distribution weight which decides weightage of the campaign, and wherein the intent lists direct application and booster application associated with the intent, and wherein at least a processor is configured to execute a snippet for generating the actionable content based on the determined intent using the campaign data.
|