# US 12,002,081 B2 · System for generating product recommendations using biometric data > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,002,081 B2 - Document ID: US12002081B2 - Status: Granted - Kind: B2 - Application number: 17/364,310 - Filed: 2021-06-30 - Published: 2021-12-30 - Granted: 2024-06-04 - Earliest claimed: 2020-06-30 - Assignees: L'oreal - Applicants: L'oreal - Inventors: Fred Orsita; Jennifer Lee; Shelby Stewart - Web page: https://patentgrep.com/patent/US12002081B2 ## CPC - G06Q 30/0251: Targeted advertisements - G06Q 30/0631: {Recommending goods or services} - A61B 5/381: Olfactory or gustatory stimuli - A61B 5/384: Recording apparatus or displays specially adapted therefor - A61B 5/6898: {Portable consumer electronic devices, e.g. music players, telephones, tablet computers} - A61B 2503/12: Healthy persons not otherwise provided for, e.g. subjects of a marketing survey - G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) ## Abstract Systems and methods for facilitating product preferences and/or product recommendations are disclosed. Biometric data of the subject are taken into account by the systems and methods in order to determine product preferences and/or product recommendations. Other factors of the subject, although optional, may be also taken into account while determining product preferences and/or product recommendations for a subject. These product preferences or recommendations can be presented to the subject, either automatically via a display device or with assistance from a product consultant. To obtain the biometric data, the subject will be exposed to a number of stimuli, such as fragrance/scent stimuli. Biometric data will then be collected from the subject based on her response to this fragrance/scent stimuli. In some examples, the collected biometric data relates to a subject's event-related potential (ERP)—the measured brain response that is the direct result of a specific sensory, cognitive, or motor event. ## Claims ### Claim 1 A system, comprising: a plurality of electroencephalograph (EEG) electrodes configured to sense an event-related potential of a subject based on a response to an olfactory stimulus; and one or more engines configured to: receive the event-related potential of a subject as EEG signals generated by the EEG electrodes; process the EEG signals to generate EEG data; and generate a product recommendation based at least on said EEG data, wherein the product recommendation is generated by a recommendation engine that includes a machine learning model trained on a set of supervised training data comprising images of electroencephalographs or brain activity maps of subjects exposed to known fragrances. ### Claim 2 (depends on 1) The system of claim 1, wherein the one or more engines are housed in a mobile computing device. ### Claim 3 (depends on 1) The system of claim 1, wherein the generated EEG data is represented as an image. ### Claim 4 (depends on 3) The system of claim 3, wherein the image includes an electroencephalograph (EEG) image or a brain activity map. ### Claim 5 (depends on 1) The system of claim 1, wherein the one or more engines are further configured to generate preferred characteristic parameters of the olfactory stimulus based on the EEG data and to generate the product recommendation based on the preferred characteristic parameters. ### Claim 6 (depends on 5) The system of claim 5, wherein the one or more engines are configured to determine a product recommendation by comparing data indictive of the generated preferred characteristic parameters to product data accessible by the one or more engines. ### Claim 7 (depends on 5) The system of claim 5, wherein the olfactory stimulus is a fragrance and wherein the data indictive of the generated preferred characteristic parameters includes a fragrance profile. ### Claim 8 (depends on 7) The system of claim 7, wherein the fragrance profile is presented to the subject as the product recommendation. ### Claim 9 (depends on 7) The system of claim 7, wherein the generated preferred characteristic parameters represent notes of the fragrance, and wherein the product recommendation is generated by comparing the fragrance profile with a set of fragrance profiles representing, respectively, a set of fragrances, the set of fragrance profiles accessible by the one or more engines. ### Claim 10 (depends on 1) The system of claim 1, wherein the one or more engines are configured to detect the event-related potential based on the response to an olfactory stimulus includes processing circuitry configured to: detect a real-time cognitive process associated with an olfactory stimulus; detect a real-time event-related potential associated with a response to one or more fragrance accords; detect voltage fluctuations indicative of a response to an olfactory stimulus; or detect a postsynaptic potential based on a response to an olfactory stimulus. ### Claim 11 (depends on 1) The system of claim 1, further comprising: a perfume selection unit including processing circuitry configured to generate one or more virtual instances of a fragrance subset based on at least one input associated with the event-related potential. ### Claim 12 (depends on 11) The system of claim 11, wherein the perfume selection unit further includes one of: processing circuitry configured to generate one or more virtual instances of a fragrance subset based on at least one input associated with an electroencephalogram; processing circuitry configured to generate one or more instances of a degree of desirability score or a likeability measure; processing circuitry configured to generate one or more instances of a scent strength; processing circuitry configured to generate one or more instances of aromatic compound concertation; or processing circuitry configured to generate one or more instances of base notes, top notes or middle notes in a fragrance composition. ### Claim 13 A method for recommending a product to a subject, comprising: obtaining biometric data in the form of event-related potential of a subject as electroencephalograph (EEG) signals generated by a plurality of EEG electrodes based on exposure to an olfactory stimulus; processing the EEG signals to generate EEG data; and generating a product recommendation based at least on the EEG data, wherein the product recommendation is generated by a recommendation engine that includes a machine learning model trained on a set of supervised training data comprising images of electroencephalographs or brain activity maps of subjects exposed to known fragrances. ### Claim 14 (depends on 13) The method of claim 13, wherein generating the product recommendation includes presenting a fragrance name to the subject; or presenting a fragrance profile to the subject. ### Claim 15 (depends on 13) The method of claim 13, further comprising obtaining questionnaire data of the subject indicative of a preference of a characteristic parameter of a product, wherein the product recommendation is further based on the questionnaire data. ### Claim 16 (depends on 13) The method of claim 13, wherein the product recommendation is a perfume recommendation. ### Claim 17 (depends on 16) The method of claim 16, wherein the product recommendation includes a fragrance profile based on the EEG data. ### Claim 18 (depends on 17) The method of claim 17 further comprising comparing the fragrance profile with a set of fragrance profiles representing, respectively, a set of fragrances to select a fragrance from the set of fragrances having a fragrance profile most similar to the generated fragrance profile, and presenting the selected fragrance to the subject. ### Claim 19 (depends on 14) The method of claim 14, wherein are positioned at a front left (F7) lobe and front right (F8) lobe of the subject and are configured to detect brain activity associated with approach and avoidance responsive to the exposure to the olfactory stimulus. ### Claim 20 A non-transitory computer-readable medium having stored thereon instructions configured to cause a computer system to perform steps comprising: obtaining biometric data in the form of event-related potential of a subject as electroencephalograph (EEG) signals generated by a plurality of EEG electrodes based on exposure to an olfactory stimulus; processing the EEG signals to generate EEG data; and generating a product recommendation based at least on the EEG data, wherein the product recommendation is generated by a recommendation engine that includes a machine learning model trained on a set of supervised training data comprising images of electroencephalographs or brain activity maps of subjects exposed to known fragrances, wherein the EEG electrodes are positioned at a front left (F7) lobe and front right (F8) lobe of the subject and are configured to detect brain activity associated with approach and avoidance responsive to the exposure to the olfactory stimulus.