Automatic generation of personalized visually isolated text

• US 10,565,291 B2
• Filed: 10/23/2017
• Issued: 02/18/2020
• Est. Priority Date: 10/23/2017
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Sbodio, Marco Luca, Kotoulas, Spyros, Lopez Garcia, Vanessa, Mulligan, Natalia, Tommasi, Pierpaolo
• Primary Examiner(s): Cortes, Howard
• Application Number:
  US15/790,202
• Publication Number:
  US 20190121841A1
• Time in Patent Office: 848 Days
• Field of Search: None
• CPC Class Codes: G06F 16/335 Filtering based on additional data, e.g. user or group profiles filtering in web context G06F16/9535, G06F16/9536
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 16/358 Browsing; Visualisation therefor
• CPC Class Codes: G06F 40/131 Fragmentation of text files, e.g. creating reusable text-blocks; Linking to fragments, e.g. using XInclude; Namespaces
• CPC Class Codes: G06F 40/169 Annotation, e.g. comment data or footnotes
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G06F 40/279 Recognition of textual entities
• CPC Class Codes: G06F 40/295 Named entity recognition
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

Input text containing a plurality of patient information can be annotated for annotations and extracted. One or more annotations are parsed for relevant contextual information. The one or more annotations are assigned a semantic type. The one or more annotations are visually isolated, personalized to a user profile job, and outputted. The one or more extracted annotations are subjected to natural language processing operations.

Claims

1. A method to automatically generate visually isolated text annotations from a text document, personalized to a user profile, comprising:

receiving an input text;

annotating the input text by;

annotating respective text fragments of the input text for respective concepts of a set of concepts; and

categorizing respective annotated text fragments of the input text for respective categories of a set of categories;

retrieving profile characteristics of the user profile;

visually isolating respective text fragments of the input text based on the characteristics of the user profile;

outputting the visually isolated text to a user device associated with the user profile;

collecting a set of training data; and

generating a model to associate a respective user profile to a respective pattern of visually isolated text fragments, wherein the respective pattern is based on respective concepts and respective categories associated with respective text fragments of the input text;

wherein the set of training data comprises user profile data, visually isolated text fragments from historical data, annotations from historical data, and parse trees from historical data; and

wherein the annotations from historical data further comprise semantic knowledge graphs associated with the annotations from historical data.

2. The method of claim 1, wherein annotating the input text further comprises:

linking respective annotated concepts of respective text fragments of the input text to hierarchical knowledge domains;

filtering the linked text fragments through a natural language processing (NLP) parse dependency tree and assigning a respective semantic identifier to respective linked text fragments; and

linking a respective semantic identifier to respective linked text fragments.

3. The method of claim 2, wherein hierarchical knowledge domains comprise Knowledge Graphs and ontologies.

4. The method of claim 2, wherein the natural language processing (NLP) parse dependency tree identifies respective roles of respective text fragments of the input text.

5. The method of claim 1, wherein the input text comprises patient information, wherein the set of concepts comprises:

activity, disease, labs, medication, person, procedure, and service concepts; and

wherein the set of categories comprises;

clinical, contact, and social categories.

6. A system to automatically generate visually isolated text fragments from a text document, personalized to a user profile, comprising:

a computer readable storage medium storing a corpus of data;

a user interface configured to receive input and present output; and

a processor communicatively coupled to the computer readable storage medium and the user interface and having a memory comprising instructions, which, when executed by the processor, cause the processor to;

receive an input text;

annotate the input text by;

annotating respective text fragments of the input text for respective concepts of a set of concepts; and

categorizing respective annotated text fragments of the input text for respective categories of a set of categories;

retrieve profile characteristics of the user profile;

visually isolate respective text fragments of the input text based on the characteristics of the user profile;

output the visually isolated text associated with the user profile to the user interface;

collect a set of training data;

generate a model to associate a respective user profile to a respective pattern of visually isolated text fragments, wherein the respective pattern is based on respective concepts and respective categories associated with respective text fragments of the input text; and

update the model based on user feedback;

wherein the set of training data comprises user profile data, visually isolated text fragments from historical data, annotations from historical data, and parse trees from historical data; and

wherein the annotations from historical data further comprise semantic knowledge graphs associated with the annotations from historical data.

7. The system of claim 6, wherein the processor is further configured to annotate the input text by:

linking respective annotated concepts of respective text fragments of the input text to hierarchical knowledge domains;

filtering the inked text fragments through a natural language processing (NLP) parse dependency tree and assigning a respective semantic identifier to respective linked text fragments; and

linking a respective semantic identifier to respective linked text fragments.

8. The system of claim 7, wherein hierarchical knowledge domains comprise Knowledge Graphs and ontologies;

and

wherein natural language processing (NLP) parse dependency trees identify respective roles of respective text fragments of the input text.

9. The system of claim 6, wherein the input text comprises patient information, wherein the set of concepts comprises:

activity, disease, labs, medication, person, procedure, and service concepts; and

wherein the set of categories comprises;

clinical, contact, and social categories.

10. A computer program product for automatically generating visually isolated text annotations from a text document, personalized to a user profile, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to:

receive an input text;

annotating the input text by;

annotating respective text fragments of the input text for respective concepts of a set of concepts; and

categorizing respective annotated text fragments of the input text for respective categories of a set of categories;

retrieving profile characteristics of the user profile;

visually isolating respective text fragments of the input text based on the characteristics of the user profile;

outputting the visually isolated text to a user device associated with the user profile;

collect a set of training data; and

generate a model to associate a respective user profile to a respective pattern of visually isolated text fragments, wherein the respective pattern is based on respective concepts and respective categories associated with respective text fragments of the input text;

wherein the set of training data comprises user profile data, visually isolated text fragments from historical data, annotations from historical data, and parse trees from historical data; and

wherein the annotations from historical data further comprise semantic knowledge graphs associated with the annotations from historical data.

11. The computer program product of claim 10, wherein the program instructions configured to annotate the input text are further configured to cause the processor to:

link respective annotated concepts of respective text fragments of the input text to hierarchical knowledge domains;

filter the linked text fragments through natural language processing (NLP) parse dependency tree and assigning a respective semantic identifier to respective linked text fragments; and

link a respective semantic identifier to respective linked text fragments.

12. The computer program product of claim 11, wherein hierarchical knowledge domains comprise Knowledge Graphs and ontologies;

and

wherein natural language processing (NLP) parse dependency trees identify respective roles of respective text fragments of the input text.

13. The computer program product of claim 10, wherein the input text comprises patient information, wherein the set of concepts comprises:

activity, disease, labs, medication, person, procedure, and service concepts; and

wherein the set of categories comprises;

clinical, contact, and social categories.