Document Analysis Using Machine Learning And Neural Networks

Abstract

A method that tracks a gaze of a specific user as the user reads displayed electronic document to determine pauses in reading greater than a specified period of time. Objects on the displayed document are correlated to the pauses. Features are identified based on the objects and textual analysis of the electronic document. A descriptor is obtained from the user defining each of the identified features, each having a value of a relative importance or applicability. An overall value is obtained from the user indicating an overall applicability of the document with respect to specific requirements of the user, and a pre-trained, baseline machine learning model, the identified features, the descriptor for each of the identified features, the value for each of the identified features, and the overall value for each of the displayed plurality of electronic documents are combined as training data for the machine learning model.

What is claimed is: 
     
         1 . A method comprising:
 receiving, by a processing system including a processor, an eye tracking input from an eye movement tracking device;   tracking, by the processing system according to the eye tracking input, a gaze of a specific user as the specific user reads each of a displayed plurality of electronic documents on a display to determine indications of pauses in reading each of the displayed plurality of electronic documents by the specific user for greater than a specified period of time by a specific user, the pauses indicative of the gaze of the specific user;   correlating, by the processing system, objects on each of the displayed plurality of electronic documents to the pauses in reading each of the displayed plurality of electronic documents by the specific user;   identifying, by the processing system and according to a pre-trained, baseline machine learning model, features to obtain identified features based on the objects and textual analysis of each of the displayed plurality of electronic documents;   obtaining, by the processing system, a descriptor from the specific user defining each of the identified features and a value for each of the identified features indicating a relative importance or applicability of each of the identified features;   obtaining, by the processing system, an overall value from the specific user for each of the displayed plurality of electronic documents indicating an overall applicability of each of the displayed plurality of electronic documents with respect to specific requirements of the specific user; and   combining, by the processing system, the pre-trained, baseline machine learning model, the identified features, the descriptor for each of the identified features, the value for each of the identified features, and the overall value for each of the displayed plurality of electronic documents as training data for the pre-trained, baseline machine learning model.   
     
     
         2 . The method of  claim 1 , wherein a plurality of same features and a plurality of different features are identified among the displayed plurality of electronic documents. 
     
     
         3 . The method of  claim 1 , further comprising:
 determining, by the processing system, that a value associated with an identified feature is an indication of bias; and   identifying, by the processing system, the indication of bias in the training data for mitigation during subsequent electronic document evaluation.   
     
     
         4 . The method of  claim 1 , wherein the displayed plurality of electronic documents comprise job applications or resumes, and
 wherein the features comprise one or more of logical skill development information, career progression information, misleading phrases related to employment experience, and gaps in employment history.   
     
     
         5 . The method of  claim 1 , wherein the objects are one of a word, a phrase, and a picture. 
     
     
         6 . The method of  claim 1 , wherein a determining of the indications of pauses in reading further comprises:
 tracking, by the processing system, from the eye tracking input obtained from the eye movement tracking device, eye positions of eyes of the specific user on each of the displayed plurality of electronic documents as the specific user reads each of the displayed plurality of electronic documents; and   detecting, by the processing system, pauses in movement of the eyes of the specific user for greater than the specified period of time on a displayed electronic document or detecting repeated eye movements between different portions of the displayed electronic document.   
     
     
         7 . The method of  claim 6 , wherein the correlating of the objects on each of the displayed plurality of electronic documents to the pauses in movement of the eyes of the specific user comprises:
 determining, by the processing system, based on data from the eye movement tracking device, the eye positions of the specific user's eyes on the display;   correlating, by the processing system, portions of each of the displayed plurality of electronic documents with the eye positions of the specific user's eyes; and   identifying, by the processing system, the objects corresponding to the portions of each of the displayed plurality of electronic documents.   
     
     
         8 . The method of  claim 1 , wherein the display is a touch-sensitive display,
 wherein the eye movement tracking device comprises the touch-sensitive display,   wherein the specific user follows text of each of the displayed plurality of electronic documents across the touch-sensitive display with a finger or a stylus in contact with the touch-sensitive display as the specific user reads each of the displayed plurality of electronic documents, and   wherein a determining of the indications of pauses in reading comprises determining, based on data from the touch-sensitive display, pauses in movement of the finger or the stylus in contact with the touch-sensitive display for the specified period of time.   
     
     
         9 . The method of  claim 8 , wherein the correlating an object on each of the displayed plurality of electronic documents to the pauses of the specific user in reading each of the displayed plurality of electronic documents further comprises:
 determining, by the processing system, based on the data from the touch-sensitive display, positions on the touch-sensitive display of the finger or the stylus in contact with the touch-sensitive display;   correlating, by the processing system, portions of each of the displayed plurality of electronic documents with the positions of the finger or the stylus; and   identifying, by the processing system, the objects corresponding to the portions of each of the displayed plurality of electronic documents.   
     
     
         10 . The method of  claim 1 , wherein the display is a touch-sensitive display configured to provide haptic feedback in response to a specified amount of pressure exerted on the display by a finger or a stylus,
 wherein the specific user follows text of each of the displayed plurality of electronic documents with the finger or the stylus as the specific user reads each of the displayed plurality of electronic documents, and   wherein the determining indications of pause in reading comprises exerting the specified amount of pressure on the touch-sensitive display with the finger or the stylus to generate the haptic feedback for the specified period of time.   
     
     
         11 . The method of  claim 10 , wherein the correlating of the objects on each of the displayed plurality of electronic documents to the pauses in reading comprises:
 determining, by the processing system, based on data from the touch-sensitive display, positions of the finger or the stylus exerting the specified amount of pressure on the touch-sensitive display to generate the haptic feedback for the specified period of time;   correlating, by the processing system, portions of each of the displayed plurality of electronic documents with the positions of the finger or the stylus; and   identifying, by the processing system, the objects corresponding to the portions of each of the displayed plurality of electronic documents.   
     
     
         12 . The method of  claim 1 , further comprising identifying, by the processing system, training data for a plurality of different models, each of the plurality of different models trained based on training data identified for a different user. 
     
     
         13 . A system, comprising:
 a processing system including a processor; and   a memory that stores executable instructions that, when executed by the processing system, facilitate performance of operations, comprising:
 tracking a gaze of a specific user according to input obtained from an eye movement tracking device as the specific user reads each of a displayed plurality of electronic documents; 
 determining pauses in reading each of the displayed plurality of electronic documents for greater than a specified period of time, the pauses indicative of the gaze of the specific user; 
 correlating objects on each of the displayed plurality of electronic documents to the pauses in reading each of the displayed plurality of electronic documents by the specific user; 
 identifying features for a pre-trained, baseline machine learning model to obtain identified features based on the objects and textual analysis of each of the displayed plurality of electronic documents; 
 obtaining a descriptor from the specific user defining each of the identified features and a value for each of the identified features indicating a relative importance or applicability of each of the identified features; 
 obtaining an overall value from the specific user for each of the displayed plurality of electronic documents indicating an overall applicability of each of the displayed plurality of electronic documents with respect to specific requirements of the specific user; and 
 combining the pre-trained, baseline machine learning model, the identified features, the descriptor for each of the identified features, the value for each of the identified features, and the overall value for each of the displayed plurality of electronic documents as training data for the pre-trained, baseline machine learning model. 
   
     
     
         14 . The system of  claim 13 , wherein the displayed plurality of electronic documents comprises a job application or a resume, and wherein the identified features comprise one or more of logical skill development information, career progression information, misleading phrases related to employment experience, and gaps in employment history. 
     
     
         15 . The system of  claim 13 , wherein a determining of the pauses in reading further comprises:
 tracking, by the processing system, from the eye tracking input obtained from the eye movement tracking device, eye positions of eyes of the specific user on each of the displayed plurality of electronic documents as the specific user reads each of the displayed plurality of electronic documents; and   detecting, by the processing system, pauses in movement of the eyes of the specific user for greater than the specified period of time on a displayed electronic document or detecting repeated eye movements between different portions of the displayed plurality of electronic documents.   
     
     
         16 . The system of  claim 15 , wherein the correlating objects on each of the displayed plurality of electronic documents further comprises:
 determining based on data from the eye movement tracking device, the eye positions of the eyes of the specific user on the display;   correlating portions of each of the displayed plurality of electronic documents with the eye positions of the eyes of the specific user; and   identifying the objects corresponding to the portions of each of the displayed plurality of electronic documents.   
     
     
         17 . The system of  claim 13 , wherein the display is a touch-sensitive display,
 wherein the eye movement tracking device comprises the touch-sensitive display,   wherein the specific user follows text of each of the displayed plurality of electronic documents across the touch-sensitive display with a finger or a stylus in contact with the touch-sensitive display as the specific user reads each of the displayed plurality of electronic documents, and   wherein a determining of the pauses in reading comprises determining, based on data from the touch-sensitive display, pauses in movement of the finger or the stylus in contact with the touch-sensitive display for the specified period of time.   
     
     
         18 . The system of  claim 17 , wherein the correlating objects on each of the displayed plurality of electronic documents further comprises:
 determining based on the data from the touch-sensitive display, positions on the touch-sensitive display of the finger or the stylus in contact with the touch-sensitive display;   correlating portions of each of the displayed plurality of electronic documents with the positions of the finger or the stylus; and   identifying the objects corresponding to the portions of each of the displayed plurality of electronic documents.   
     
     
         19 . The system of  claim 13 , wherein the operations further comprise:
 further comprising identifying, by the processing system, training data for a plurality of different models, each of the plurality of different models trained based on training data identified for a different user.   
     
     
         20 . A non-transitory, machine-readable medium comprising executable instructions that, when executed by a processing system including a processor operating from a device, facilitate performance of operations, the operations comprising:
 tracking a gaze of a specific user according to input obtained from an eye movement tracking device as the specific user reads each of a displayed plurality of electronic documents to determine pauses in reading each of the displayed plurality of electronic documents for greater than a specified period of time by a specific user, the pauses indicative of the gaze of the specific user;   correlating objects on each of the displayed plurality of electronic documents to the pauses in reading by the specific user;   identifying features to obtain identified features based on the objects and textual analysis of each of the plurality of electronic documents;   obtaining a descriptor from the specific user defining each of the identified features and a value for each of the identified features indicating a relative importance or applicability of each of the identified features;   obtaining an overall value from the specific user for each of the displayed plurality of electronic documents indicating an overall applicability of each of the plurality of displayed electronic documents with respect to specific requirements of the specific user; and   combining a pre-trained, baseline machine learning model, the identified features, the descriptor for each of the identified features, the value for each of the identified features, and the overall value for each of the displayed plurality of electronic documents as training data for the pre-trained, baseline machine learning model.