Facilitation of artificial intelligence predictions of telecommunications customers

Abstract

Machine learning (ML) can be used to gather data on prospective customers from publicly available online sources. The ML can provide data on a likelihood of near future, customers and subscribers and a list of people with low interest in the company's products. The ML system can be recursively updated as new public data becomes available online. Thus, the ML system can be used to generate targeted advertising and root cause analysis based on a relationship between people's statuses and their respective interests in certain products and/or services.

What is claimed is: 
     
         1 . A method, comprising:
 labeling, by a server system comprising a processor, social media data, representative of statuses associated with a user identity, with type data representative of respective types of the statuses;   matching, by the server system, first identity data representative of the user identity to second identity data representative of the user identity, wherein the first identity data is obtained from a first source, wherein the second identity data is obtained from a second source that is different from the first source, wherein the first identity data comprises a previous location associated with the user identity, and wherein the second identity data is determined to be representative of the previous location;   generating, by the server system, prediction data, representative of a propensity of utilization of a service based on a credential associated with the user identity, based on the type data, the first identity data, and the second identity data; and   in response to generating the prediction data, facilitating, by the server system, display of information associated with the prediction data at a display screen of a user equipment associated with the user identity.   
     
     
         2 . The method of  claim 1 , further comprising:
 aggregating, by the server system, the statuses associated with the user identity from respective social media account services associated with the user identity, resulting in aggregated status data, wherein the labeling of the social media data comprises labeling the social media data further based on the aggregated status data.   
     
     
         3 . The method of  claim 1 , wherein the social media data is first social media data, wherein the statuses are first statuses, wherein the user identity is a first user identity, and wherein the labeling comprises labeling the first social media data further based on an output of a machine learning model applied to second social media data representative of second statuses, authored prior to the first statuses and associated with second user identities that are distinct from the first user identity. 
     
     
         4 . The method of  claim 3 , wherein the output of the machine learning model is representative of a proportion of the second user identities that have been determined to have subscribed to the service. 
     
     
         5 . The method of  claim 3 , wherein the machine learning model is a supervised machine learning model, wherein the output of the supervised machine learning model is a first output, and wherein the generating of the prediction data comprises generating the prediction data further based on a second output of an unsupervised machine learning model applied to the type data. 
     
     
         6 . The method of  claim 1 , wherein the service is a telecommunications service associated with a telecommunications service provider. 
     
     
         7 . The method of  claim 1 , wherein the first identity data and the second identity data are representative of an email address associated with the user identity. 
     
     
         8 . The method of  claim 1 , wherein the rendering is first rendering, wherein the information is first information, and wherein the method further comprises:
 in response to the propensity of the utilization of the service, as represented by the prediction data, being greater than a threshold, facilitating, by the server system, second rendering of second information relating to the service to the display screen of the user equipment associated with the user identity.   
     
     
         9 . The method of  claim 8 , wherein the facilitating of the second rendering comprises facilitating the second rendering of the information relating to the service via a social media account that is associated with the social media data. 
     
     
         10 . A system, comprising:
 a processor; and   a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations, comprising:
 labeling social media data, representative of statuses associated with a user identity, with type data representative of respective types of the statuses; 
 matching first identity data representative of the user identity to second identity data representative of the user identity, wherein the first identity data is obtained via a first source and comprises a previous location associated with the user identity, and wherein the second identity data is obtained via a second source that is distinct from the first source and is determined to be representative of the previous location; 
 generating prediction data, representative of a likelihood of utilization of a service in connection with the user identity, based on the type data, the first identity data, and the second identity data; and 
 in response to generating the prediction data, facilitating display of information associated with the prediction data at a display screen of a user equipment associated with the user identity. 
   
     
     
         11 . The system of  claim 10 , wherein the operations further comprise:
 receiving the statuses associated with the user identity from a social media account service associated with the user identity; and   generating the social media data by aggregating the statuses as received from the social media account service.   
     
     
         12 . The system of  claim 10 , wherein the social media data is first social media data, wherein the statuses are first statuses, wherein the user identity is a first user identity, and wherein the labeling comprises labeling the first social media data further based on an output of a machine learning model applied to second statuses, the second statuses being authored prior to the first statuses and associated with second user identities that are distinct from the first user identity. 
     
     
         13 . The system of  claim 12 , wherein the output of the machine learning model is representative of a percentage of the second user identities that have been determined to have subscribed to the service. 
     
     
         14 . The system of  claim 12 , wherein the machine learning model is a supervised machine learning model, wherein the output of the supervised machine learning model is a first output, and wherein the generating of the prediction data comprises generating the prediction data further based on a second output of an unsupervised machine learning model applied to the type data. 
     
     
         15 . The system of  claim 10 , wherein the service is a telecommunications service associated with a telecommunications service provider. 
     
     
         16 . A non-transitory machine readable medium, comprising executable instructions that, when executed by a processor, facilitate performance of operations, comprising:
 associating status data, representative of statuses associated with a user identity of a user subscribed to a social media account, with type data representative of respective types of the statuses;   matching first identity data representative of the user identity to second identity data representative of the user identity, wherein the first identity data is obtained via a first source and comprises a previous location associated with the user identity, and wherein the second identity data is obtained via a second source that is distinct from the first source and is determined to be representative of the previous location;   generating prediction data, representative of a probability of utilization of a service by the user identity, based on the type data, the first identity data, and the second identity data; and   in response to generating the prediction data, facilitating rendering of information associated with the prediction data at a display screen of a user equipment associated with the user identity.   
     
     
         17 . The non-transitory machine readable medium of  claim 16 , wherein the user identity is a first user identity, wherein the status data is first status data representative of first statuses associated with the user identity, and wherein the associating of the first status data with the type data comprises associating the first status data with the type data based on an output of a machine learning model applied to second status data representative of second statuses, originating prior to the first statuses and associated with second user identities that are distinct from the first user identity. 
     
     
         18 . The non-transitory machine readable medium of  claim 17 , wherein the output of the machine learning model is representative of a percentage of the second user identities that have been determined to have subscribed to the service. 
     
     
         19 . The non-transitory machine readable medium of  claim 17 , wherein the machine learning model is a supervised machine learning model, wherein the output of the supervised machine learning model is a first output, and wherein the generating of the prediction data comprises generating the prediction data further based on a second output of an unsupervised machine learning model applied to the type data. 
     
     
         20 . The non-transitory machine readable medium of  claim 16 , wherein the service is a telecommunications service associated with a telecommunications service provider.