System and method for transferring data, scheduling appointments, and conducting conferences

Abstract

An efficient and secure process by which users may enter sensitive information into an electronic information system. When information is required from a user, the electronic information system may be configured to generate a unique access link (uniform resource locator, or URL) for that user. The link may be sent to the user via electronic communication, such as a text message or email. When the user follows the link with a web browser, the system prompts the user to enter an additional piece of personal information that is not known to the general public. Once identity is verified, the user may be required to electronically sign agreements. The user is then prompted to enter the required information. This may allow a user to deposit sensitive information into the system without requiring the user to provide full login credentials.

What is claimed is: 
     
         1 . A method for predicting high risk appointments, the method comprising:
 accessing a data store comprising patient data and appointment data;   creating a read replica of the data store;   pre-processing data from the read replica resulting in normalized data;   providing the normalized data to a base classifier group;   predicting risk probabilities by the base classifier group resulting in initial predictions;   analyzing the initial predictions by false negative classifiers;   adjusting the initial predictions for appointments if a false negative is detected, resulting in adjusted predictions;   providing the adjusted predictions to a meta-classifier; and   returning a final prediction from the meta-classifier.   
     
     
         2 . The method of  claim 1 , further comprising assigning a human-readable confidence string based on a magnitude of a risk score. 
     
     
         3 . The method of  claim 1 , further comprising collecting a final prediction for each appointment instance tasked for a prediction in a communicable payload and communicating the payload such that a user interface is updated to reflect the final prediction for each appointment. 
     
     
         4 . The method of  claim 1 , wherein pre-processing comprises stripping superfluous strings, dropping unnecessary columns or features, applying a cyclical encoder to certain data, correcting missing values, and label encoding. 
     
     
         5 . The method of  claim 1 , wherein the base classifier group comprises a random forest ensemble, a gradient boosted machine, and a neural network. 
     
     
         6 . The method of  claim 1 , wherein the meta-classifier is a neural network. 
     
     
         7 . The method of  claim 1 , wherein patient data and appointment data is at least one of demographic data, appointment data, patient scheduling data, and patient form data. 
     
     
         8 . The method of  claim 1 , further comprising automatically performing a prediction for each scheduled appointment at a predetermined period in advance of a scheduled appointment time or date. 
     
     
         9 . The method of  claim 1 , further comprising receiving prediction requests through a user interface. 
     
     
         10 . A method for scheduling an appointment comprising:
 providing a scheduling interface for at least one of a patient and a provider;   collecting appointment information through the scheduling interface;   creating an appointment using the collected information;   storing appointment data and user data in a data store;   running a prediction application to identify appointments at risk for cancelations or no-shows; and   presenting an attendance probability data through a provider interface.   
     
     
         11 . The method of  claim 10 , further comprising creating an audio and/or video conferencing session to be launched for the scheduled appointment; and communicating an access link to the audio and/or video conferencing session to at least one participant. 
     
     
         12 . The method of  claim 10 , further comprising automatically requesting, by a software scheduler, a prediction from the prediction application for upcoming appointments. 
     
     
         13 . The method of  claim 10 , further comprising creating a read replica of the data store for processing by a prediction application. 
     
     
         14 . The method of  claim 10 , further comprising pre-processing data from the read replica resulting in normalized data; providing the normalized data to a base classifier group; predicting risk probabilities by the base classifier group resulting in initial predictions; analyzing the initial predictions by false negative classifiers; adjusting the initial predictions for appointments if a false negative is detected, resulting in adjusted predictions; providing the adjusted predictions to a meta-classifier; and returning a final prediction from the meta-classifier. 
     
     
         15 . The method of  claim 14 , wherein pre-processing comprises stripping superfluous strings, dropping unnecessary columns or features, applying a cyclical encoder to certain data, correcting missing values, and label encoding. 
     
     
         16 . The method of  claim 15 , wherein the base classifier group comprises a random forest ensemble, a gradient boosted machine, and a neural network; and wherein the meta-classifier is a neural network. 
     
     
         17 . A non-transitory computer-readable medium comprising computer program code that, when executed, causes a computer system comprising an input system, an output system, a processor, and a memory to perform the steps of:
 accessing a data store comprising patient data and appointment data;   creating a read replica of the data store;   pre-processing data from the read replica resulting in normalized data;   providing the normalized data to a base classifier group;   predicting risk probabilities by the base classifier group resulting in initial predictions;   analyzing the initial predictions by false negative classifiers;   adjusting the initial predictions for appointments if a false negative is detected, resulting in adjusted predictions;   providing the adjusted predictions to a meta-classifier; and   returning a final prediction from the meta-classifier.   
     
     
         18 . The computer-readable medium of  claim 17 , wherein the steps further comprise collecting a final prediction for each appointment instance tasked for a prediction in a communicable payload and communicating the payload such that a user interface is updated to reflect the final prediction for each appointment. 
     
     
         19 . The computer-readable medium of  claim 17 , wherein pre-processing comprises stripping superfluous strings, dropping unnecessary columns or features, applying a cyclical encoder to certain data, correcting missing values, and label encoding. 
     
     
         20 . The computer-readable medium of  claim 17 , wherein the base classifier group comprises a random forest ensemble, a gradient boosted machine, and a neural network, and wherein the meta-classifier is a neural network.