Systems and methods of natural language processing to rank users of real time communications connections
Inventors
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-9509846-B1
Publication Date
2016-11-29
Expiration Date
Abstract
A computing apparatus configured to perform natural language processing, e.g., by comparing the words in a communication transcript of a user to a list of keywords, to generate an input vector representing a pattern of text in the communication transcript. A predictive model is generated from correlating input vectors to user ranking scores, e.g., for retention. The input vector determined from a communication transcript is applied to the computation model to compute a predicted retention score of the user. The retention score can be used, for example, to select a personalized recommendation for a communication connection to an adviser and/or a targeted offer.
Core Innovation
The invention relates to a method and a computing apparatus for real time communications in which a connection server establishes communication connections between telephonic devices and a web server presents user information. First users are associated with a first set of telephonic devices registered in a data store, and second users are associated with a second set of telephonic devices. The web server presents information of the first users to the second users, receives requests identifying selected ones of the first users, and the connection server establishes communications based on communication references stored in the data store.
After connections are established, the computing apparatus generates transcripts of communications over the communication connections and stores communication records in the data store. The computing apparatus performs natural language processing on the transcripts and on a transcript of a communication between a user and a selected one of the first users to generate input vectors. Ranking scores of the second users are generated from the communication records, and the input vectors are correlated with the ranking scores to establish a predictive model of ranking score.
The predictive model of ranking score is then applied by the computing apparatus using first input vectors derived from the transcript of the new communication between a user and a selected one of the first users, thereby generating a ranking score of the user. In this way, user ranking is derived from natural language processing on communication transcripts and from the correlation between input vectors and ranking scores.
Claims Coverage
The document contains three independent claims. Across these claims, the inventive features center on server-mediated real time communication setup, transcript generation, natural language processing to produce input vectors, and correlating input vectors with ranking scores to establish and apply a predictive model of ranking score.
Server-mediated real time communication between telephonic devices based on communication references
A connection server establishes real time communication connections between telephonic devices. A web server presents information of first users to second users, receives requests identifying selected ones of the first users, and the connection server establishes communication connections between respective telephonic devices based on communication references in the data store.
Transcript generation and communication record storage for established connections
Transcripts are generated of communications over the communication connections, and communication records of the communication connections are stored in the data store.
Natural language processing on transcripts to generate input vectors and building a predictive model of ranking score
Natural language processing is performed on the transcripts to generate input vectors, ranking scores of the second users are generated from the communication records, and the input vectors are correlated with the ranking scores to establish a predictive model of ranking score.
Applying predictive model to generate a ranking score for a user using NLP-derived input vectors from a new transcript
A communication connection is provided between a telephonic device of a user and a telephonic device of a selected one of the first users, a transcript of the communication is generated, natural language processing is performed on the transcript to generate first input vectors, and the first input vectors are applied in the predictive model of ranking score to generate a ranking score of the user.
Instruction storage on a non-transitory computer storage medium implementing the same ranking workflow
Instructions are stored to perform server-mediated real time communication based on communication references, generate transcripts and store communication records, perform natural language processing to generate input vectors, generate ranking scores, correlate input vectors with ranking scores to establish a predictive model, and apply input vectors from a new transcript to generate a ranking score of the user.
Computing apparatus configured with a data store, web server, and connection server for ranking via NLP and predictive modeling
A computing apparatus includes a data store storing communication references and information about first users, a web server configured to present first-user information to second users and receive requests identifying selected ones of the first users, and a connection server configured to establish communication connections based on the communication references, generate transcripts, store communication records, perform natural language processing to generate input vectors, generate ranking scores, correlate input vectors with ranking scores to establish a predictive model of ranking score, and apply input vectors from a new transcript to generate a ranking score of the user.
Across the independent claims, the core claimed coverage is a server-mediated real time communication workflow that uses communication references to connect telephonic devices, generates and stores transcripts and communication records, applies natural language processing to produce input vectors, and correlates those input vectors with ranking scores to establish and apply a predictive model of ranking score to generate ranking scores for users.
Stated Advantages
Documented Applications
Ranking users based on communications over server-established real time communication connections by generating ranking scores from natural language processing of communication transcripts and applying a predictive model of ranking score.
Interested in licensing this patent?