Cognitive presentation advisor

Patent metadata

Field	Value
Publication number	US-9633008-B1
Application number	US-201615279659-A
Country	US
Kind code	B1
Filing date	Sep 29, 2016
Priority date	Sep 29, 2016
Publication date	Apr 25, 2017
Grant date	Apr 25, 2017

Abstract

A computer-implemented method for reviewing content of a presentation is provided. The method includes obtaining, using a processor system of a device, content of an electronic presentation, wherein the content is associated with a first slide of the presentation. The format of the content is determined wherein the content includes natural language content and non-natural language content and the non-natural language content is converted into natural language content using the processor system. The sentiment and tone of the natural language content is analyzed using the processor system. A textual summary is created. The sentiment, tone and textual summary of the natural language content of the slide of the presentation is evaluated and a summary of the natural language content of the slide is presented using a display based upon the evaluation of the sentiment, tone and textual summary.

First claim

What is claimed is:

1. A computer-implemented method for reviewing content of a presentation, the method comprising:
obtaining, using a processor system of a device, content of an electronic presentation, wherein the content is associated with a first slide and a second slide of the presentation;
determining, using the processor system, a format of the content contained on the first slide and the second slide of the presentation wherein the content includes natural language content and non-natural language content;
converting non-natural language content into natural language content using the processor system;
analyzing a sentiment of the natural language content using the processor system;
analyzing a tone of the natural language content using the processor system;
creating a textual summary of the natural language content using the processor system;
evaluating the sentiment, tone and textual summary of the natural language content of the first slide and the second slide of the presentation;
presenting a summary of the natural language content of the first slide and the second slide using a display based upon the evaluation of the sentiment, tone and textual summary.

2. The computer-implemented method of claim 1 , wherein the presentation consists of a plurality of slides and the summary of the natural language content of each slide based upon the evaluation of the sentiment, tone and textual summary of each of the respective slides is compared to the remaining slides to determine an overall summary of the sentiment, tone and textual summary of the entire presentation.

3. The computer-implemented method of claim 2 , wherein at least one of the sentiment, tone or textual summary of one of the slides of the presentation is compared to the sentiment of the one of the slides of the presentation, tone and textual summary of the entire presentation to determine if the sentiment, tone and textual summary of the respective slide is consistent with the sentiment, tone, and textual summary of the entire presentation.

4. The computer-implemented method of claim 3 , wherein the sentiment of a respective slide is scored creating a slide sentiment score, wherein the sentiment of the entire presentation is scored creating a presentation sentiment score, and wherein the method includes recommending a change to the natural language content of the respective slide if the slide sentiment score is of a marginal difference greater than a predetermined value when compared to the presentation sentiment score.

5. The computer-implemented method of claim 3 , wherein the tone of a respective slide is scored creating a slide tone score, wherein the tone of the entire presentation is scored creating a presentation tone score, and wherein the method includes recommending a change to the natural language content of the respective slide if the slide tone score is of a marginal difference greater than a predetermined value when compared to the presentation tone score.

6. The computer-implemented method of claim 1 , wherein the sentiment of the first slide is scored creating a first slide sentiment score, and wherein the sentiment of the second slide is scored creating a second slide sentiment score, and wherein the method includes recommending a change to the natural language content of either the first or second slide if the first slide sentiment score is of a marginal difference greater than a predetermined value when compared to the second slide sentiment score.

7. The computer-implemented method of claim 1 , wherein the tone of the first slide is scored creating a first slide tone score, and wherein the tone of the second slide is scored creating a second tone sentiment score, and wherein the method includes recommending a change to the natural language content of either the first or second slide if the first slide tone score is of a marginal difference greater than a predetermined value when compared to the second slide tone score.

8. The computer-implemented method of claim 1 , wherein the first slide contains visual content and audio content, wherein the visual and audio content is translated to natural language.

9. A system for reviewing content of an electronic presentation having multiple slides, the system comprising:
a processor system;
a cognitive presentation advisor including:
an input module, using the processor system, accessing content of an electronic presentation, wherein the content is associated with a first slide and a second slide of the presentation;
a natural language module, using the processor system, determining a format of the content contained on a first slide and the second slide of the presentation wherein the content includes natural language content and non-natural language content, the natural language module converting non-natural language content into natural language content using the processor system;
a sentiment analysis module, using the processor system, analyzing the sentiment of the natural language content;
a tonal analysis module, using the processor system, analyzing the tone of the natural language content;
a word stemming and summation module, using the processor system, creating a textual summary of the natural language content; and
an evaluation module, using the processor system, evaluating the sentiment, tone and textual summary of the natural language content of the first and second slides of the presentation; and
a display displaying a summary of the natural language content of the first slide and the second slide based upon the evaluation of the sentiment, tone and textual summary.

10. The system of claim 9 wherein the sentiment analysis module analyzes the sentiment of a slide by evaluating the natural language content as being positive or negative.

11. The system of claim 9 wherein the evaluation module evaluates the sentiment, tone and textual summary of the natural language of each slide individually and the slides aggregately.

12. The system of claim 9 including a recommendation module, using the processor system, recommending modifications to the natural language content of a respective slide if the slide has a sentiment score the slides.

13. The system of claim 12 wherein the tone of a respective slide is scored creating a slide tone score, wherein the tone of the entire presentation is scored creating a presentation tone score, and wherein the recommendation module recommends a change to the natural language content of the respective slide if the slide tone score is of a marginal difference greater than a predetermined value when compared to the presentation tone score.

14. A computer program product for reviewing content of a presentation, the computer program product comprising a computer readable storage medium having program instructions embodied therewith the program instructions readable by a processing circuit to cause the processing circuit to perform a method comprising:
obtaining, using a processor system of a device, content of an electronic presentation, wherein the content is associated with a first slide and a second slide of the presentation;
determining, using the processor system, a format of the content contained on the first slide and the second slide of the presentation wherein the content includes natural language content and non-natural language content;
converting non-natural language content into natural language content using the processor system;
analyzing a sentiment of the natural language content using the processor system;
analyzing a tone of the natural language content using the processor system;
creating a textual summary of the natural language conte

Assignees

• IBM

Inventors

• Brand Samantha M
• Dow Eli M
• Fitzsimmons Thomas D

Classifications

• G06F16/345 Primary

  Summarisation for human users · CPC title

• G06F40/30 Primary

  Semantic analysis · CPC title

• G06F17/2785 Primary

  Physics · mapped topic

• G06F17/2264

  Physics · mapped topic

• G06F17/2705

  Physics · mapped topic