WO2022120190A1

WO2022120190A1 - Systems and methods determined by person and product attributes to provide expert informed cosmetic product recommendations

Info

Publication number: WO2022120190A1
Application number: PCT/US2021/061838
Authority: WO
Inventors: Taylor BABAIAN; Narendra PINNAMANEMI
Original assignee: Skin Posi Inc.
Priority date: 2020-12-04
Filing date: 2021-12-03
Publication date: 2022-06-09

Abstract

This paper describes the systems and methods for narrowing cosmetic goods in the marketplace to create product recommendations specific to unique personal attributes and preferences. Patterns of dependency are determined between multiple personal attributes and preferences as well as various product value propositions and properties. Consumer profiles are created by collecting personal user information, concerns, and preferences and weighted by importance. A product recommendation is created through a recommendation system that uses the consumer data and filters through a database of systemized cosmetic products collected from the marketplace and narrows them to suggest items that match consumer profiles.

Description

SYSTEMS AND METHODS DETERMINED BY PERSON AND PRODUCT ATTRIBUTES TO PROVIDE EXPERT INFORMED COSMETIC PRODUCT RECOMMENDATIONS

CROSS-REFERENCES TO RELATED APPLICATIONS

[0001] This application claims benefit of U.S. Provisional Patent Application No. 63/121 ,588 filed December 4, 2020, the specification of which is incorporated herein in its entirety by reference.

FIELD OF THE INVENTION

[0002] The present invention is directed to machine learning systems that utilize a comprehensive set of parameters received from a consumer in order to provide accurate recommendations for cosmetic products.

BACKGROUND OF THE INVENTION

[0003] Recommendation systems in e-commerce like Amazon, provide product recommendations by employing machine learning to narrow products in the marketplace and suggesting products that will greatly increase the likelihood of the consumer purchasing the product. The algorithms used to create predictions are based on the user's previous ratings of past purchases or behaviors. However, the user’s past behaviors and purchases alone cannot determine if a cosmetic product is actually suited for the user as recommendations are not based on the user’s unique skin attributes. Further, repeat purchases of suggested products only validate algorithms minimizing the opportunity for the user to find a better suited product. Thus, there exists a present need for a system capable of taking into account a comprehensive set of parameters specific to a customer in order to provide accurate recommendations for cosmetic products.

BRIEF SUMMARY OF THE INVENTION

[0004] It is an objective of the present invention to provide systems and methods that allow for providing accurate recommendations for cosmetic products, as specified in the independent claims. Embodiments of the invention are given in the dependent claims. Embodiments of the present invention can be freely combined with each other if they are not mutually exclusive.

[0005] The present invention features a system for accepting consumer preferences and attributes in order to provide recommendations for cosmetic products. The system may comprise a device in communication with a user, a product database, and a recommendation engine. The recommendation engine may utilize machine learning based on a training set of potential characteristics of a user matched to ingredients that interact well with said user. The user may provide a plurality of consumer attributes to the device, which may be fed into the recommendation engine, which will utilize machine learning to select a plurality of products from the product database based on a plurality of product attributes that correspond to the plurality of consumer attributes. The plurality of products selected by the recommendation engine may be displayed on the device.

[0006] The present invention features a method for accepting consumer preferences and attributes in order to provide recommendations for cosmetic products. In some embodiments, the method may comprise maintaining a systematic database of cosmetic products able to be queried by a recommendation engine. The method may further comprise receiving a plurality of consumer information and a product recommendation request from the user. The method may further comprise retrieving by a recommendation engine, a plurality of products in a category requested by the user. The method may further comprise narrowing the plurality of products by user concerns, user preferences, and user price preference. The method may further comprise creating a hierarchy of products based on a plurality of weighted variables applied to the narrowed plurality of products. The method may further comprise displaying product recommendations based on the hierarchy of products on the device.

[0007] One of the unique and inventive technical features of the present invention is the use of a machine learning algorithm based on expert informed data to create a narrowed hierarchy of cosmetic products based on a plurality of consumer preferences and attributes (e.g. past product ratings). Without wishing to limit the invention to any theory or mechanism, it is believed that the technical feature of the present invention advantageously provides for accurate cosmetic product recommendations to the consumer that are capable of being updated over use of the present invention. None of the presently known prior references or work has the unique inventive technical feature of the present invention.

[0008] Any feature or combination of features described herein are included within the scope of the present invention provided that the features included in any such combination are not mutually inconsistent as will be apparent from the context, this specification, and the knowledge of one of ordinary skill in the art. Additional advantages and aspects of the present invention are apparent in the following detailed description and claims.

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING(S)

[0009] The features and advantages of the present invention will become apparent from a consideration of the following detailed description presented in connection with the accompanying drawings in which:

[0010] FIG. 1 shows a system for accepting consumer preferences and attributes in order to provide recommendations for cosmetic products.

[0011] FIG. 2 shows an example of a database of consumer preferences and attributes and a database of cosmetic products categorized into product type, target, function, etc.

[0012] FIG. 3 shows a flow chart of a method for providing accurate recommendations for cosmetic products upon receiving a consumer’s preferences and attributes.

[0013] FIGs 4A-4I show a plurality of examples of how a software application utilizing the present invention may operate.

[0014] FIG. 5A shows a flow chart of a method of the present invention for collecting data from a user to fill the user attribute database. FIG. 5B shows a flow chart of a method of the present invention for accepting user feedback for products to improve future recommendations for products to other users.

DETAILED DESCRIPTION OF THE INVENTION

[0015] Following is a list of elements corresponding to a particular element referred to herein: [0016] 100 receiving customer data

[0017] 101 receive product recommendation request

[0018] 102 retrieve products in category

[0019] 103 divide by concerns

[0020] 104 narrow by preferences

[0021] 105 divide by price preferences

[0022] 106 create product hierarchy

[0023] 107 display recommendations

[0024] The present invention features a system for accepting preferences, concerns, and attributes from a user in order to provide recommendations for cosmetic products. In some embodiments, the system may comprise a product database comprising a plurality of cosmetic products. Each cosmetic product may comprise one or more product attributes, a primary function, and one or more ingredients. The system may further comprise a user attribute database comprising one or more user preferences comprising a price preference, one or more user attributes, and one or more user concerns. The one or more user preferences may be weighted and ranked based on importance to the user. The system may comprise a recommendation engine communicatively coupled to the product database and the user attribute database, comprising a processor capable of executing computer-readable instructions and a memory component comprising a plurality of computer-readable instructions. The plurality of computer-readable instructions may comprise weighting, by a machine learning algorithm, for each cosmetic product of the product database, an efficacy based on the primary function. The plurality of computer-readable instructions may further comprise accepting the one or more user preferences and one or more user attributes, accepting a list of products from the product database, filtering the list of products based on the one or more user concerns and the one or more user attributes, filtering the list of products based on the one or more user preferences, filtering the list of products based on the price preference, generating a hierarchy of products from the list of products based on weight, and displaying the hierarchy of products to the user.

[0025] in some embodiments, the machine learning aigorithm may be trained by expert informed data and user reviews. The one or more product attributes may comprise claimed value, category, ingredients, ingredient percentage, price, size, solubility, chemical properties, and brand. The one or more user preferences may comprise if the product is clean, , if a product is organic, if a product is vegan, sustainability, price, and user’s rating of the product. The one or more user attributes may comprise age, sex, skin characteristics, and goals. The one or more user concerns may comprise acne, pores, oily skin, dry skin, fine lines, wrinkles, rosacea, eczema, sag, spots, veins, and dark circles. A cosmetic product may be a toner, moisturizer, exfoliant, cleanser, coiorizer, or active. The user attribute database may be filled by the user through a third party, a computing device, or a survey. The product database comprises of products from at least one brand. The memory component may further comprise computer-readable instructions for allowing the user to order one or more products from the hierarchy of products from one or more locations.

[0026] The present invention features a method for accepting preferences, concerns, and attributes from a user in order to provide recommendations for cosmetic products. The method may comprise accepting, from the user, one or more user preferences, one or more user concerns, and one or more user attributes. The one or more user preferences may be weighted and ranked based on importance to the user. The method may further comprise providing a product database comprising a plurality of cosmetic products. Each cosmetic product may comprise one or more product attributes, a primary function, and one or more ingredients. The method may further comprise weighting, by a machine learning algorithm, for each cosmetic product of the product database, an efficacy score based on the primary function, accepting the one or more user preferences and one or more user attributes, accepting a list of products from the product database, filtering the list of products based on the one or more user concerns and the one or more user attributes, filtering the list of products based on the one or more user preferences, filtering the list of products based on the price preference, generating a hierarchy of products from the list of products based on weight, and displaying the hierarchy of products to the user.

[0027] in some embodiments, the machine learning aigorithm may be trained by expert informed data and user reviews. The one or more product attributes may comprise claimed value, category, ingredients, ingredient percentage, price, size, solubility, chemical properties, and brand. The one or more user preferences may comprise if the product is clean, if a product is organic, if a product is vegan, sustainability, price, and rating. The one or more user attributes may comprise age, sex, skin characteristics, and goals. The one or more user concerns may comprise acne, pores, oily skin, dry skin, fine lines, wrinkles, rosacea, eczema, sag, spots, veins, and dark circles. The primary feature of each cosmetic product may be selected from a group comprising toner, moisturizer, exfoliant, cleanser, colorizer, and active. The user attribute database may be filled by the user through a third party, a computing device, or a survey. The product database may comprise products from only one brand. The method may further comprise allowing the user to order one or more products from the hierarchy of products from one or more locations.

[0028] Referring now to FIG. 1 , the present invention features a system for accepting consumer preferences and attributes in order to provide recommendations for cosmetic products. The system may comprise a device in communication with a user, a product database, and a recommendation engine. The user may communicate with the device through a software application on a phone, computer, etc. The recommendation engine may utilize machine learning based on a training set of potential characteristics of a user matched to ingredients that interact well with said user. The training set may comprise expert informed recommendations and user reviews of products contained in the product database. The user may provide a plurality of consumer attributes to the device, comprising sex, age, skin characteristics, concerns, goals, preferences, and product ratings. FIG. 2 shows examples included in the user and product databases In some embodiments, the plurality of consumer attributes may be weighted in terms of importance in order to aid in product retrieval. The plurality of consumer attributes may be fed into the recommendation engine, which will utilize machine learning to select a plurality of products from the product database based on a plurality of product attributes that correspond to the plurality of consumer attributes. The plurality of product attributes may comprise value propositions, category, price, ingredients, brand attributes, solubility, and chemical properties. FIG. 2 also shows a table of various options that each product may fall into for product attributes. The plurality of products selected by the recommendation engine may be displayed on the device.

[0029] Referring now to FIG. 3, the present invention features a method for accepting consumer preferences and attributes in order to provide recommendations for cosmetic products. In some embodiments, the method may comprise maintaining a systematic database of cosmetic products able to be queried by a recommendation engine. The method may further comprise receiving (100), by a device, a plurality of consumer profile personal data, attributes, and preferences. The method may further comprise receiving (101), by the device, a product recommendation request from the user. The method may further comprise retrieving (102), by a recommendation engine, a plurality of products in a category requested by the user. The method may further comprise dividing (103), by the recommendation engine, the plurality of products by one or more concerns listed by the user. The method may further comprise narrowing (104), by the recommendation engine, the divided plurality of products by one or more preferences listed by the user. The method may further comprise dividing (105), by the recommendation engine, the narrowed plurality of products by price preference received from the user. The method may further comprise creating (106), by the recommendation engine, a hierarchy of products based on a plurality of weighted variables applied to the divided and narrowed plurality of products. The method may further comprise displaying (107), by the device, product recommendations based on the hierarchy of products.

[0030] FIGs 4A-4I show a plurality of examples of a software application implementing the system of the present invention. FIGs 4A-4B show an opening screen of the software application and user login set up. As seen in FIG. 4C, the software application may request a photograph of a user in order to automatically recognize skin characteristics through the use of a facial recognition algorithm. FIG. 4D shows an example of ranking user preferences in order to aid a recommendation engine in generating a hierarchy of products based on said preferences. FIGs 4E-4F show a request for the user to upload and rate one or more products currently used by the user in order to further aid the recommendation engine. FIGs 4G-4H show a product page, comprising a price, description, ratings, and an option to order the product from one or more locations. FIG. 41 shows an option to re-order previously ordered products and receive samples of certain products. In some embodiments, the present invention may be employed by pre-existing shopping sites, such as Amazon.

[0031] Referring now to FIG. 5A, the present invention features a method for collecting data from a user to fill the user attribute database. The method may comprise collecting data from a user (e.g. age, sex, concerns, preferences, etc.). The method may further comprise creating a unique user identifier and preparing the data from the user as empirical data. This empirical data (e.g. concerns and preferences) may be weighted according to expert-informed attributes. The method may further comprise matching products from the product database to the user based on the empirical data. The method may further comprise displaying the top results from the product database that match the user. If the user profile is missing any data, the method may further comprise requesting additional data from the user to improve future predictions. The method may further comprise accepting user feedback on products recommended to them to update future recommendations.

[0032] Referring now to FIG. 5B, the present invention features a method for accepting user feedback for products to improve future recommendations for products to other users. The method may comprise collecting data (e.g. ingredients, price, brand) from all products contained in the product database. The method may further comprise creating a unique product identifier for each product and preparing the product data as empirical data. This empirical data may be analyzed, categorized into qualitative and quantitative variables, and system variable matches may be identified. Each product may be weighted, based on expert informed parameters, on their efficiency at their primary objective (e.g. moisturizer, cleanser, etc.). The weighted products may be matched to users through a match predictor if the product attributes align with the preferences, concerts, and attributes of the user. These recommendations may be displayed to the user.

[0033] Although there has been shown and described the preferred embodiment of the present invention, it will be readily apparent to those skilled in the art that modifications may be made thereto which do not exceed the scope of the appended claims. Therefore, the scope of the invention is only to be limited by the following claims. In some embodiments, the figures presented in this patent application are drawn to scale, including the angles, ratios of dimensions, etc. In some embodiments, the figures are representative only and the claims are not limited by the dimensions of the figures. In some embodiments, descriptions of the inventions described herein using the phrase “comprising” includes embodiments that could be described as “consisting essentially of” or “consisting of”, and as such the written description requirement for claiming one or more embodiments of the present invention using the phrase “consisting essentially of” or “consisting of” is met.

[0034] The reference numbers recited in the below claims are solely for ease of examination of this patent application, and are exemplary, and are not intended in any way to limit the scope of the claims to the particular features having the corresponding reference numbers in the drawings.

Claims

WHAT IS CLAIMED IS:

1 . A system for accepting preferences, concerns, and attributes from a user in order to provide recommendations for cosmetic products comprising: a. a product database comprising a plurality of cosmetic products; wherein each cosmetic product comprises one or more product attributes, a primary function, and one or more ingredients; b. a user attribute database comprising one or more user preferences comprising a price preference, one or more user attributes, and one or more user concerns; wherein the one or more user preferences are weighted and ranked based on importance to the user; and c. a recommendation engine communicatively coupled to the product database and the user attribute database, comprising a processor capable of executing computer-readable instructions and a memory component comprising a plurality of computer-readable instructions for: i. weighting, by a machine learning algorithm, for each cosmetic product of the product database, an efficacy based on the primary function; ii. accepting the one or more user preferences and one or more user attributes; ill. accepting a list of products from the product database; iv. filtering the list of products based on the one or more user concerns and the one or more user attributes; v. filtering the list of products based on the one or more user preferences; vi. filtering the list of products based on the price preference; vii. generating a hierarchy of products from the list of products based on weight; and viii. displaying the hierarchy of products to the user.

2. The system of claim 1 , wherein the machine learning algorithm is trained by expert informed data and user reviews. The system of claim 1 , wherein the one or more product attributes comprise claimed value, category, ingredients, ingredient percentage, price, size, solubility, chemical properties, and brand. The system of claim 1 , wherein the one or more user preferences comprise cleanliness of a product, if a product is organic, if a product is vegan, sustainability, price, and rating. The system of claim 1 , wherein the one or more user attributes comprise age, sex, skin characteristics, and goals. The system of claim 1 , wherein the one or more user concerns comprise acne, pores, oily skin, dry skin, fine lines, wrinkles, rosacea, eczema, sag, spots, veins, and dark circles. The system of claim 1 , wherein the primary feature of each cosmetic product is selected from a group comprising toner, moisturizer, exfoliant, cleanser, colorizer, and active. The system of claim 1 , wherein the user attribute database is filled by the user through a third party, a computing device, or a survey. The system of claim 1 , wherein the product database comprises products from only one brand. The system of claim 1 , wherein the memory component further comprises computer-readable instructions for allowing the user to order one or more products from the hierarchy of products from one or more locations. A method for accepting preferences, concerns, and attributes from a user in order to provide recommendations for cosmetic products comprising: a. accepting, from the user, one or more user preferences, one or more user concerns, and one or more user attributes; wherein the one or more user preferences are weighted and ranked based on importance to the user; b. providing a product database comprising a plurality of cosmetic products; wherein each cosmetic product comprises one or more product attributes, a primary function, and one or more ingredients; c. weighting, by a machine learning algorithm, for each cosmetic product of the product database, an efficacy based on the primary function; d. accepting the one or more user preferences and one or more user attributes; e. accepting a list of products from the product database; f. filtering the list of products based on the one or more user concerns and the one or more user attributes; g. filtering the list of products based on the one or more user preferences; h. filtering the list of products based on the price preference;

I. generating a hierarchy of products from the list of products based on weight; and j. displaying the hierarchy of products to the user. The method of claim 11 , wherein the machine learning algorithm is trained by expert informed data and user reviews. The method of claim 11 , wherein the one or more product attributes comprise claimed value, category, ingredients, ingredient percentage, price, size, solubility, chemical properties, and brand. The method of claim 11 , wherein the one or more user preferences comprise cleanliness of a product, if a product is organic, if a product is vegan, sustainability, price, and rating. The method of claim 11 , wherein the one or more user attributes comprise age, sex, skin characteristics, and goals. The method of claim 11 , wherein the one or more user concerns comprise acne, pores, oily skin, dry skin, fine lines, wrinkles, rosacea, eczema, sag, spots, veins, and dark circles. The method of claim 11 , wherein the primary feature of each cosmetic product is selected from a group comprising toner, moisturizer, exfoliant, cleanser, colorizer, and active. The method of claim 11 , wherein the user attribute database is filled by the user through a third party, a computing device, or a survey. The method of claim 11 , wherein the product database comprises products from only one brand, The method of claim 11 , wherein the method further comprises allowing the user to order one or more products from the hierarchy of products from one or more locations.