Distributed electronic document review in a blockchain system and computerized scoring based on textual and visual feedback

• US 9,870,591 B2
• Filed: 12/21/2016
• Issued: 01/16/2018
• Est. Priority Date: 09/12/2013
• Status: Active Grant

• Current Assignee: Intellectual Frontiers LLC
• Original Assignee: Netspective Communications LLC
• Inventors: Shah, Shahid N.
• Primary Examiner(s): Araque, Jr., Gerardo
• Application Number:
  US15/386,846
• Publication Number:
  US 20170103472A1
• Time in Patent Office: 391 Days
• Field of Search: 705317, 705319, 705320, 705347
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06N 3/02 Neural networks
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/043 Distributed expert systems; Blackboards
• CPC Class Codes: G06Q 10/063112 Skill-based matching of a person or a group to a task
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: G06V 40/174 Facial expression recognition
• CPC Class Codes: G06V 40/18 Eye characteristics, e.g. of the iris
• CPC Class Codes: H04L 9/0637 Modes of operation, e.g. cipher block chaining [CBC], electronic codebook [ECB] or Galois/counter mode [GCM]
• CPC Class Codes: H04L 9/32 including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials

Abstract

A blockchain configured system and a method for facilitating an expertise driven review and scoring of electronic documents in a crowdsourced environment. The system includes a server computer, a memory circuit and a processing circuit. The processing circuit is coupled to the memory circuit and includes or is coupled to a credentialing engine. The system further includes an expert scoring module. The system further includes a document reviewing and scoring engine coupled to the processing circuit. The document review and scoring module associates an aggregate score to the electronic document based on aggregation of the review ratings by crowdsourced experts and aggregate scores of each of the crowdsourced experts based on the set of attributes including one or more of the credentialed expertise, reputation of the expert, and the officiality.

Claims

1. A blockchain configured distributed architecture-based system in a communication network, said system comprising:

a memory circuit communicatively connected to said communication network that stores a plurality of digital profiles associated with a plurality of crowdsourced experts, and further stores a plurality of segmented digital profiles associated with each of said digital profiles, wherein said segmented digital profiles and digital profiles are created based on a plurality of sources distributed and electronically linked across said communication network;

a processor coupled with the memory circuit to execute instructions for evaluating an expert, the instructions comprising;

a credentialing engine that allows a plurality of crowdsourced respondents to respond to said segmented digital profiles associated with each of said plurality of experts and credential said plurality of experts and determine crowdsourced credentialed expertise, wherein the credentialing of each of said segmented digital profiles associated with an expert of said plurality of experts contribute to credentialing of a digital profile of said expert upon collation of said credentialed segmented digital profiles, and wherein said segmented digital profiles associated with said experts are credentialed from a plurality of respondents using a computerized crowdsourcing index, wherein said computerized crowdsourcing index is indicative of number of respondents credentialing an expert and dynamically increases with an increase in said number of respondents;

an expert scoring module to;

determine a set of attributes for said experts, said set of attributes including one or more of said crowdsourced credentialed expertise determined based on said credentialing of said segmented digital profiles of said experts by said respondents, reputation of said experts indicative of a trust of a relevant community on said experts, and officiality indicative of a position or a designation of said experts in a relevant job, wherein each of said attributes are assigned varying computer-calculated weights; and

determine an aggregate score of an expert based on said one or more attributes in association with the assigned weights;

an electronic document scoring engine to receive and process comments and document ratings for an electronic document by said crowdsourced experts, wherein said crowdsourced experts have an aggregate score greater than a defined threshold, the document scoring engine comprising;

a natural language processing-based (NLP-based) analysis engine to process textual information-based reviews and comments generated as part of textual review of said electronic document by said crowdsourced experts;

a visual scoring engine for processing visual and non-textual feedback and reviews by the crowdsourced experts, wherein the visual scoring engine comprises;

an eye tracks processor controlled by a special purpose microprocessor to receive eye track inputs from respective eye tracking systems associated with computing devices of said crowdsourced experts and process said eye track inputs to associate a review score based on predefined eye track patterns; and

a micro expressions processor to receive data indicative of micro facial expressions extracted by respective micro expressions sensors associated with said computing devices of said crowdsourced experts, wherein said micro expressions processor comprises an image processing circuitry and an associated memory to interpret said micro facial expressions and compare them with predefined facial patterns to associate a review score based on said extracted micro facial expressions,

wherein, the document scoring engine further configured to;

associate an aggregate score to said electronic document based on aggregation of individual textual and visual review scores obtained by processing of said textual and said visual reviews by said crowdsourced experts who review the document; and

display on a graphical user interface device, an output indicative of an aggregate score of the document reviewed by said crowdsourced experts along with information about who reviewed and how many times reviewed the document;

an expert identity validation device to verify identities of the crowdsourced experts during or prior to review, wherein said expert identity validation device comprises;

a device patterns assessment device to receive and process device information extracted by respective agent devices associated with said computing devices of said crowdsourced experts and verify the extracted device information with predefined device information for the respective crowdsourced experts;

a network patterns assessment device to receive and process network information extracted by said respective agent devices associated with said computing devices of said crowdsourced experts and verify the extracted network information with predefined network information of the respective crowdsourced experts;

a geo-spatial mapping device to perform geo-tagging of the crowdsourced experts and the documents reviewed by said crowdsourced experts and compare the geo-tags with pre-stored geo-spatial information about the experts for processing validation, wherein the geo-tagging is performed based on geo-spatial information received from a global positioning system (GPS)-based device; and

a facial expression validation device to receive and process facial expressions received from respective facial expression sensors associated with said computing devices of said crowdsourced experts and verify identity in accordance with respective predefined facial patterns of said crowdsourced experts, wherein the facial expression validation device comprises a digital acquisition unit and multichannel amplifiers for pre-processing and amplification of signals transmitted by said facial expression sensors.

2. The system of claim 1, further comprising an expert scoring module, wherein said expert scoring module further comprises a computerized weight module to identify a degree of relevance and significance of an expert attribute with said electronic document to be reviewed and accordingly assign a weight to each of said attributes of said experts based on said identified degree of significance.

3. The system of claim 2, wherein said attribute of credentialed expertise is weighed as the highest by said weight module followed by said officiality, and said attribute of reputation is weighed as the lowest by said weight module.

4. The system of claim 2, wherein said weight module is adapted to dynamically change weight assignment based on type of said electronic document.

5. The system of claim 1, further comprises a reputation assessment engine adapted to determine a degree of reputation of an expert.

6. The system of claim 1, further comprises an officiality assessment engine to determine a degree of officiality of an expert.

7. The system of claim 1 further comprising a federation engine to fragment a digital profile of an expert into a plurality of segmented digital profiles based on commonalities in content of said segmented digital profiles, wherein said segmented digital profiles are treated as distinct profiles for said purpose of credentialing separately by said crowdsourced respondents.

8. The system of claim 1, wherein said document reviewing and scoring module further comprising an electronic document aggregate score assessment engine, comment analysis module, comment aggregator, semantics and analytics engine, and document classification/tagging module.

9. The system of claim 1, wherein said aggregate score of an expert (AES) for one or more attributes is determined based on an empirical relation, said empirical relation being:

AES=EW1+RW2+OW3, wherein ‘

E’

represents credentialed expertise, ‘

R’

represents reputation, and ‘

O’

represents officiality, and W1, W2, W3 represent weightage of said credentialed expertise, reputation, and officiality respectively.

10. The system of claim 9, wherein said credentialing engine evaluates said credentialed expertise (E) for said expert based on an empirical relation, said empirical relation being:

E=(P_F11+P_F12+ . . . +P_F1N)×

(P_F21+P_F22+ . . . P_F2N)X . . . X(P_FZ1+P_FZ2+ . . . +P_FZN),

wherein;

P_F11 is nonzero and represents a credentialed segmented profile score for a first segmented profile of a first expert by a first respondent,

P_F12 is nonzero and represents a credentialed segmented profile score for said first segmented profile of said first expert by a second respondent,

P_F1N is nonzero and represents a credentialed segmented profile score for said first segmented profile of said first expert by an Nth respondent,

P_F21 is nonzero and represents a credentialed segmented profile score for a second segmented profile of said first expert by said first respondent,

P_F22 is nonzero and represents a credentialed segmented profile score for said second segmented profile of said first expert by said second respondent,

P_F2N is nonzero and represents a credentialed segmented profile score for said second segmented profile of said first expert by said Nth respondent,

P_FZ1 is nonzero and represents a credentialed segmented profile score for a Zth segmented profile of said first expert by said first respondent,

P_FZ2 is nonzero and represents a credentialed segmented profile score for said Zth segmented profile of said first expert by said second respondent,

P_FZN is nonzero and represents a credentialed segmented profile score for said Zth segmented profile of said first expert by said Nth respondent, and

wherein said empirical relation above considers profiles scores for entire segmented digital profiles from 1 to Z, wherein said empirical relation above considers all respondents from 1 to N.

11. The system of claim 10, wherein said document reviewing and scoring engine evaluates aggregate crowdsourced document score (ACDS) based on credentialed expertise and other attributes of said crowdsourced experts, based on an empirical relation, said empirical relation being:

ACDS={(E₁+E₂+E₃+ . . . +E_X)W₁+(R₁+R₂+R₃+ . . . +R_X)W₂+(O₁+O₂+O₃+ . . . +O_X)W₃}(D₁+D₂+D₃+ . . . +D_X)CI

wherein;

E₁, E₂, E₃, . . . E_X represent respective credentialed expertise of X number of crowdsourced experts,

R₁, R₂, R₃, . . . R_X represent respective reputation of said X number of crowdsourced experts,

O₁, O₂, O₃, . . . O_X represent respective officiality of said X number of crowdsourced experts,

D₁, D₂, D₃ . . . D_X represent respective document scores earned by said X number of crowdsourced experts, and

CI represents a non-linear crowdsourcing index.

12. The system of claim 11, wherein said CI is defined non-linearly with integral ranges (R) of experts who credential said document, first five of said ranges and corresponding CI being:

CI=1, when R=0-2 experts,

CI=1.2, when R=3-4 experts,

CI=1.5, when R=5-6 experts

CI=1.9, when R=7-8 experts, and

CI=2.5, when R=9-10 experts.

13. The system of claim 11, wherein said CI is calculated based on an empirical relationship that dynamically determines value of said CI with every integral change in number of expert credentialing said electronic document.

14. The system of claim 1, wherein said eye tracks processor is configured to process the eye tracks inputs, wherein said eye tracks inputs includes data indicative of time lapse on specific sections of the document under review, repeat reading, coherence between eye focus and scrolling speed, blinks, expert'"'"'s gaze location, eye fixations, saccades, gaze rate, gaze direction, gaze percentage, number of fixations, scan path directionality, instances of backtracking, frequency of long duration dwells over areas of interest, saccade length, fixation/saccade time ratio, and visual profile of review-flow.

15. The system of claim 1, wherein said facial expressions processor is configured to process the facial expressions, and wherein said facial expressions includes data indicative of sensation or passion of an expert, mental views and psychosomatic aspects originating from bones, tissues, skin, muscles, deformities, facial contractions and expansions.

16. The system of claim 1, wherein said visual scoring engine further comprising an image processor configured to perform image processing and pre-processing tasks.