Allstate Insurance Company

Ryan Dunn

Michael A LaVigne

Eric D Huls

David MacInnis

Cristina I Gheorghe

Alexander Doyle

Christopher Joseph DeCleene

Soren K Lundsgaard

Anthony Pham

Ben Phillabaum

Aspects of the disclosure relate to computing platforms that utilize improved natural language processing techniques for performance evaluation and training. A computing platform may automatically determine, based on audio transcription files, a model for dynamic performance evaluation and training, which may be dynamically updated as additional audio transcription files are received. The computing platform may receive and analyze an additional audio transcription file using natural language processing and the model, which may result in proficiency scores. Based on the proficiency scores, the computing platform may calculate an overall proficiency score and identify areas for improvement associated with the first individual. Based on the areas for improvement, the computing platform may determine performance feedback tailored to the individual and may send, to a user device associated with the individual, the performance feedback.

Natural language processing platform for automated training and performance evaluation

A computer-assisted method for providing re-quotations for insurance coverage may include receiving a list of insurance leads corresponding to individuals who received a previous quotation for insurance coverage but did not purchase the insurance coverage and identifying a difference between the previous quotation and a new quotation. This difference may include an increase in offered insurance coverage and/or a reduction in cost. A computing device may calculate a probability for each of the individuals on the list using a regression model based, at least in part, on the identified difference. In some cases, the regression model may be associated with individual states. In other cases, the regression model may correspond to a plurality of states. The regression model may output a probability that a resident of a particular state will purchase insurance in response to a re-quotation for insurance coverage, where individuals may then be ranked based on the probability.

Prioritization of insurance requotations

Aspects of the disclosure relate to using machine learning for optimized call routing. A computing platform may receive requests to establish a voice call session. Based on corresponding phone numbers, the computing platform may identify demographic information for corresponding clients. Using a machine learning model and based on the demographic information and representative performance data, the computing platform may score potential client-representative combinations to indicate likelihoods of a successful outcome resulting from establishing a voice call session between the respective client and representative. Scoring the potential client-representative combinations may be based on fall off rates, indicating changes in representative effectiveness as hold time increases. The computing platform may adjust the scores based on a historical frequency of interaction between each representative and clients corresponding to the identified demographic information. Based on the adjusted scores, the computing platform may select at least one of the potential client-representative combinations.

Machine learning system for routing optimization based on historical performance data

Growing community of inventors

Grayslake, IL, United States of America

Ryan Dunn