How Microsoft anticipated wearables, machine learning and cognitive enhancement: Key Neurotech Patent #18

contextual-responses

– Illus­tra­tive image from U.S. Patent No. 6,842,877

Let’s dis­cuss today a fas­ci­nat­ing patent assigned to Microsoft back in 2015. (As men­tioned, we are fea­tur­ing a foun­da­tion­al Per­va­sive Neu­rotech patent a day, from old­er to new­er by issue date)

U.S. Patent No. 6,842,877: Con­tex­tu­al respons­es based on auto­mat­ed learn­ing techniques.

  • Assignee(s): Microsoft Corporation
  • Inventor(s): James O. Robarts Eric L. Matteson
  • Tech­nol­o­gy Cat­e­go­ry: Neu­rocog­ni­tive Training
  • Issue Date: Jan­u­ary 11, 2005

SharpBrains’ Take:

The ‘877 patent dis­clos­es wear­able devices that incor­po­rate tech­niques for user-feed­back loops to auto­mat­i­cal­ly improve a sys­tem’s response, for exam­ple when the com­put­ing sys­tem iden­ti­fies  user’s needs and pref­er­ences through the use of sens­ing com­po­nents, in a vari­ety of envi­ron­men­tal contexts. 

Cog­ni­tive ben­e­fits may occur, when the auto­mat­ed response results in pro­vid­ing a tem­plate for user-improve­ment, such as using a per­son­’s psy­cho­log­i­cal pro­file under the Mey­er-Brig­gs test to train an indi­vid­ual to be more out­spo­ken if they are an intro­vert or be aware that, as an extro­vert, they are spend­ing too much time talk­ing. Although the claims are focused on the adap­tive learn­ing aspect of the wear­able device, this patent is a key non-inva­sive neu­rotech­nol­o­gy patent due to the inter­sec­tion of the fast-grow­ing wear­able tech sec­tor with the cog­ni­tive enhance­ment appli­ca­tions dis­closed there­in, the lengthy and detailed spec­i­fi­ca­tion (with 48 illus­tra­tion sheets and 31 pages of writ­ten mate­r­i­al), and the appli­ca­tion hav­ing received an aver­age of 43 cita­tions per year.

Abstract:

Tech­niques are dis­closed for using a com­bi­na­tion of explic­it and implic­it user con­text mod­el­ing tech­niques to iden­ti­fy and pro­vide appro­pri­ate com­put­er actions based on a cur­rent con­text, and to con­tin­u­ous­ly improve the pro­vid­ing of such com­put­er actions. The appro­pri­ate com­put­er actions include pre­sen­ta­tion of appro­pri­ate con­tent and func­tion­al­i­ty. Feed­back paths can be used to assist auto­mat­ed machine learn­ing in detect­ing pat­terns and gen­er­at­ing inferred rules, and improve­ments from the gen­er­at­ed rules can be imple­ment­ed with or with­out direct user con­trol. The tech­niques can be used to enhance soft­ware and device func­tion­al­i­ty, includ­ing self-cus­tomiz­ing of a mod­el of the user’s cur­rent con­text or sit­u­a­tion, cus­tomiz­ing received themes, pre­dict­ing appro­pri­ate con­tent for pre­sen­ta­tion or retrieval, self-cus­tomiz­ing of soft­ware user inter­faces, sim­pli­fy­ing repet­i­tive tasks or sit­u­a­tions, and men­tor­ing of the user to pro­mote desired change.

Illus­tra­tive Claim 41. A wear­able com­put­ing sys­tem con­fig­ured to improve auto­mat­ed respons­es to a cur­rent con­text for a user, the cur­rent con­text being rep­re­sent­ed by a plu­ral­i­ty of con­text attrib­ut­es that each mod­el an aspect of the con­text, mul­ti­ple defined con­tex­tu­al sit­u­a­tions each spec­i­fy­ing val­ues for at least one of the con­text attrib­ut­es, mul­ti­ple auto­mat­ed respons­es being asso­ci­at­ed with the defined con­tex­tu­al sit­u­a­tions, comprising:

  • a first com­po­nent that is con­fig­ured to repeat­ed­ly, receive an indi­ca­tion of cur­rent con­text infor­ma­tion that includes cur­rent val­ues for each of at least some of the plu­ral­i­ty of con­text attrib­ut­es, deter­mine one of the defined con­tex­tu­al sit­u­a­tions that match­es the indi­cat­ed cur­rent con­text infor­ma­tion, deter­mine one of the auto­mat­ed respons­es that is asso­ci­at­ed with the one defined con­tex­tu­al sit­u­a­tion, receive an indi­ca­tion from the user of an alter­nate auto­mat­ed response, and store an indi­ca­tion of the indi­cat­ed cur­rent con­text infor­ma­tion and the alter­nate auto­mat­ed response; and
  • a sec­ond com­po­nent that is con­fig­ured to auto­mat­i­cal­ly detect a rela­tion­ship between an iden­ti­fied con­tex­tu­al sit­u­a­tion and one of the alter­nate auto­mat­ed respons­es based on that alter­nate auto­mat­ed response being pre­vi­ous­ly indi­cat­ed by the user and to cre­ate an asso­ci­a­tion between the iden­ti­fied con­tex­tu­al sit­u­a­tion and the one alter­nate auto­mat­ed response so that the one alter­nate auto­mat­ed response can in the future be pro­vid­ed to the user for that con­tex­tu­al situation.

To learn more about mar­ket data, trends and lead­ing com­pa­nies in the dig­i­tal brain health space –dig­i­tal plat­forms for brain/ cog­ni­tive assess­ment, mon­i­tor­ing and enhance­ment– check out this mar­ket report. To learn more about our analy­sis of 10,000+ patent fil­ings, check out this IP & inno­va­tion neu­rotech report.

About SharpBrains

SHARPBRAINS is an independent think-tank and consulting firm providing services at the frontier of applied neuroscience, health, leadership and innovation.
SHARPBRAINS es un think-tank y consultoría independiente proporcionando servicios para la neurociencia aplicada, salud, liderazgo e innovación.

Top Articles on Brain Health and Neuroplasticity

Top 10 Brain Teasers and Illusions

Newsletter

Subscribe to our e-newsletter

* indicates required

Got the book?