KR102842457B1 - Customized snack recommendation and provision system - Google Patents

Abstract

The present invention relates to a customized snack recommendation and provision system, and a customized snack recommendation and provision system that recommends snacks to an orderer and packages and provides a plurality of snacks selected by the orderer, comprising: an orderer information collection unit that collects at least one orderer information among the orderer's gender, age, order history, allergies, preferred taste, preferred texture, diet, preferred nutrients, preferred food temperature, preferred packaging material, preferred brand, previously selected snacks, and snack consumption completion time; a snack information collection unit that collects at least one snack information among the category, nutritional content, food type, brand, raw material, and calorie information of the snack; a reward provision unit that provides a reward to the orderer when the orderer performs a preset input operation using a user terminal wirelessly connected to the system; a snack recommendation unit that recommends customized snacks and snack combinations using the orderer information and the snack information; an intake information calculation unit that divides a snack box delivered to the orderer into preset sizes and assigns numbers, and calculates the total calories consumed when the orderer consumes the snack located at the number and displays the calculated total calories consumed on the user terminal; and an intake information calculation unit that calculates the total calories consumed in the snack box collected from the orderer. It is characterized by including a feedback section that assigns a score to the snack in response to feedback information including an evaluation, score, and repurchase intention corresponding to the included snack.

Description

CUSTOMIZED SNACK RECOMMENDATION AND PROVISION SYSTEM

The present invention relates to a customized snack recommendation and provision system, and more particularly, to a customized snack recommendation and provision system for a consumer's order that recommends snacks that are purchased sequentially or simultaneously with the entire ordered product or that the consumer prefers.

Snacks refer to food intake through snacks, drinks, and light meals other than the three regular meals (breakfast, lunch, and dinner) during the day. The purpose of snacks was to supplement the nutrition that was lacking in people with much higher activity levels or children in the growth period compared to the three meals, but in modern times, it has developed into a culture that recognizes snacks as a time to take a break and enjoy simple foods such as snacks, drinks, and tea. Accordingly, snacks that capture the taste buds of consumers with various forms of taste, nutrition, and flavor are still being tried and developed to this day.

Additionally, as times progress, the number of snack production plants that produce large quantities and extend shelf life increases, and as a result, consumers' tastes and preferences for snacks are diversifying as a result of the production of various types of snacks.

Meanwhile, as snack culture develops, the number of companies purchasing various types of snacks and providing them to employees as a form of welfare is increasing, not only in ordinary households and on holidays, but also in offices where work is conducted.

However, when a specific individual in the office purchases a large quantity of snacks, it is difficult to know the tastes of many employees, and it is impossible to satisfy all employees' tastes. Therefore, even if the same amount of various snacks is purchased, some snacks are quickly consumed, while others remain unclaimed for a long time, causing a problem.

Therefore, research is required on a snack recommendation system that recommends snacks according to individual tastes and provides optimal snack recommendations by providing feedback on remaining snack data until the next cycle of snacks provided after recommendation.

Korean Patent No. 10-2299358

The purpose of the present invention is to provide recommended snacks based on objective data by scoring snack recommendation scores corresponding to information of an orderer and displaying the scores on an orderer's terminal.

In addition, the purpose is to collect information on specific ingredient allergies and vegetarianism from the customer's information and match it to the nutritional content of the snack, thereby preventing the customer from ordering snacks containing allergens or animal ingredients, thereby reducing human and economic costs due to returns and complaints.

Additionally, by collecting feedback information from customers after receiving snacks, the purpose is to provide more sophisticated snack recommendations and to optimize the snack items supplied by identifying snacks that are not publicly evaluated, thereby preventing bad inventory.

A customized snack recommendation and provision system according to an embodiment of the present invention is a customized snack recommendation and provision system that recommends snacks to an orderer and packages and provides a plurality of snacks selected by the orderer, comprising: an orderer information collection unit that collects at least one orderer information among the orderer's gender, age, order history, allergies, preferred taste, preferred texture, diet, preferred nutrients, preferred food temperature, preferred packaging material, preferred brand, previously selected snacks, and snack consumption completion time; a snack information collection unit that collects at least one snack information among the category, nutritional content, food type, brand, raw material, and calorie information of the snack; a reward provision unit that provides a reward to the orderer when the orderer performs a preset input operation using a user terminal wirelessly connected to the system; a snack recommendation unit that recommends customized snacks and snack combinations using the orderer information and the snack information; an intake information calculation unit that divides a snack box delivered to the orderer into preset sizes and assigns numbers, and calculates the total calories consumed when the orderer consumes the snack located at the number and displays the calculated total calories consumed on the user terminal; and an intake information calculation unit that collects the information collected from the orderer. The snack box may include a feedback section that assigns a score to the snack in response to feedback information including evaluations, scores, and repurchase intentions corresponding to the snacks included in the snack box.

In addition, the snack information collection unit recognizes the text in the nutritional information table indicated on the snack and calculates a health score according to [Mathematical Formula 1].

[Mathematical Formula 1]

(Here, H _p is the health score, P _r is the protein content (g), V _i is the vitamin content weight, D _f is the dietary fiber content weight, K _c is the calories per 1g (kcal), _Na is the sodium per 1g (mg), S _g is the sugar per 1g (g), F _t is the fat per 1g (g), and C _a is the carbohydrate per 1g (g).)

The health score corresponding to the snack selected by the customer may be displayed on the user terminal as at least one of bad, average, and good based on preset criteria.

In addition, the snack recommendation unit can calculate a snack recommendation level using the orderer information and the snack information according to the following [Mathematical Formula 2].

[Equation 2]

(Here, S _r represents the snack recommendation, H represents the weight of whether it is healthy, A represents the weight of whether it contains allergens, V represents the weight of whether it contains animal ingredients, m _a represents the weight of preferred packaging, CL represents the weight of calorie information, _ET represents the weight of consumption temperature, C represents the weight of whether it is cooked, T represents the weight of taste classification, B represents the weight of the manufacturer's brand, and t represents the weight of the manufacturing type.)

In addition, the snack recommendation unit individually recommends at least one snack among a challenge snack, a new snack, a tasting snack, and a compatible snack to the orderer, wherein the challenge snack is a snack whose snack recommendation is calculated with the lowest value, the new snack is a snack whose first production date is within 2 months and does not contain an allergen corresponding to the collected orderer information, and if the orderer is a vegetarian, does not contain animal ingredients, the tasting snack is a snack individually packaged in a large capacity of 100 g to 150 g, whose minimum packaging weight exceeds a preset weight among the snacks whose snack recommendation is calculated with the highest value, and the compatible snack can provide the orderer with a preset compatible snack corresponding to the customized recommended snack whose snack recommendation is calculated with the highest rate.

In addition, the feedback section calculates the intake score corresponding to each snack in the snack box according to [Mathematical Formula 3] below,

[Equation 3]

(Here, I _s is the intake score, A _s is the evaluation score, R _d is the weight of the intention to reorder, R _s is the rank value, and n is the number of different types of snacks in the snack box.)

The snack recommendation score calculated by the above-mentioned intake score and the above-mentioned snack recommendation degree at a preset ratio is transmitted to the snack recommendation unit, and the snack recommendation unit can perform priority recommendations of the snack to the orderer in order of the highest snack recommendation score.

According to the present invention, a snack recommendation score corresponding to the orderer's information is scored and displayed on the orderer's terminal, thereby providing recommended snacks based on objective data.

In addition, by collecting information on specific ingredient allergies and vegetarianism from the customer's information and matching it to the nutritional content of the snack, the customer can identify and prevent orders for snacks containing allergens or animal ingredients, thereby reducing human and economic costs due to returns and complaints.

Additionally, by collecting feedback from customers after receiving their snacks, we can provide more sophisticated snack recommendations and optimize the snack supply by identifying snacks with poor public reviews, thereby preventing wasteful inventory.

FIG. 1 is a block diagram illustrating a customized snack recommendation and provision system according to an embodiment of the present invention.

Specific details, including the problems to be solved, means of solving them, and effects of the invention, are included in the embodiments and drawings described below. The advantages and features of the present invention, and methods for achieving them, will become clearer with reference to the embodiments described below in detail, along with the accompanying drawings.

The scope of the present invention is not limited to the embodiments described below, and various modifications may be made by a person having ordinary knowledge in the relevant technical field without departing from the technical spirit of the present invention.

Hereinafter, the personalized snack recommendation and provision system of the present invention is described in detail with reference to the attached Figure 1.

First, FIG. 1 is a block diagram illustrating a customized snack recommendation and provision system according to an embodiment of the present invention.

Referring to FIG. 1, a customized snack recommendation and provision system (100) according to an embodiment of the present invention may include an orderer information collection unit (110), a snack information collection unit (120), a reward provision unit (130), a snack recommendation unit (140), an intake information calculation unit (150), and a feedback unit (160).

The personalized snack recommendation and provision system (100) of the present invention can, when an orderer orders a snack through a dedicated application provided in a preset user terminal, recommend a personalized snack to the orderer, pack a number of snacks selected by the orderer in a snack box, and deliver the snacks to the orderer.

In addition, if a dedicated application is installed on the user terminal of the orderer, data transmission and reception can be performed with the customized snack recommendation and provision system (100).

The above orderer information collection unit (110) can collect at least one piece of orderer information from among the orderer's gender, age, order history, allergies, preferred taste, preferred texture, diet, preferred nutrients, preferred food temperature, preferred packaging material, preferred brand, previously selected snack, and snack consumption completion time.

Here, the orderer information collection unit (110) can collect the orderer information by guiding the orderer to a welcome survey participation page containing the orderer information when the orderer registers as a member by installing a dedicated application using a user terminal.

In addition, a phrase indicating that the reward provision unit (130) provides benefits to the orderer who moves to the welcome survey participation page can be displayed to encourage more active participation.

The above snack information collection unit (120) can collect at least one or more snack information among snack category, nutritional content, food type, brand, raw material, and calorie information.

Here, the categories of snacks can be classified into sweets, biscuits, crackers, cookies, pies, traditional sweets, rice cakes, chocolate, jelly, caramel, candy, gum, cereal, jam, bread (bakery), jerky, raw snacks, nuts, dried fruits, dried fish, dairy products, ice cream, frozen foods, processed foods, fruits, salads, freeze-dried foods, water, tea, coffee, carbonated drinks, fruit juice, milk, yogurt, soy milk, powdered drinks, etc.

In addition, the nutritional information, raw materials, total weight information, and calorie information can be converted into text and data using OCR (Optimal Character Recognition) technology or a text recognition algorithm using deep learning in images, and then stored, based on the information indicated on the packaging of the snack.

Through this, the snack information collection unit (120) can calculate a health score according to the following [Mathematical Formula 1].

[Mathematical Formula 1]

(Here, H _p is the health score, P _r is the protein content (g), V _i is the vitamin content weight, D _f is the dietary fiber content weight, K _c is the calories per 1g (kcal), _Na is the sodium per 1g (mg), S _g is the sugar per 1g (g), F _t is the fat per 1g (g), and C _a is the carbohydrate per 1g (g).)

At this time, the vitamin content weight (V _i ) can be assigned according to [Table 1] below.

[Table 1]

The above vitamins may refer to the sum of all ingredients classified as vitamins, such as vitamin A, vitamin C, and vitamin E.

The above dietary fiber content weight (D _f ) can be assigned according to [Table 2] below.

[Table 2]

Meanwhile, the protein content (P _r ), calories per 1g (K _c, kcal), sodium per 1g (N _a, mg), sugar per 1g (S _g , g), fat per 1g (F _t , g), carbohydrate per 1g (C _a , g) The health score can be calculated by inputting the content per unit from the textual information written on the packaging of the snack.

For example, if the snack is a salad, and the nutritional information, raw materials, total weight information, and calorie information displayed are total weight 144g, sodium 310mg, carbohydrates 7g, sugars 2g, fat 9g, protein 11g, calories 153kcal, vitamins 0mg, and dietary fiber 12g, the health score (H _p ) can be calculated as approximately 4.94 points ({(11×1.5×1)÷((153+310+2+9+7)÷144)}).

Here, the health score (H _p ) may be displayed on the user terminal of the orderer as at least one of bad, average, and good according to the health score (H _p ) criteria classified in [Table 3] below.

[Table 3]

The above reward provision unit (130) can provide a reward to the orderer when the orderer performs a preset input operation using a user terminal connected to the system by wireless communication.

Here, the preset input action may mean an action of transmitting data to the system (100), such as entering participation in a survey within the welcome survey participation page, transmitting the amount of snacks consumed daily within the system, and providing feedback on a snack package.

At this time, if the input operation entered by the orderer is normally received within the system (100), rewards such as snack discount coupons, priority purchase rights for new snack products, priority delivery rights, and fee exemptions are provided to induce active participation of the orderer, thereby enabling optimization and improvements to be quickly performed through continuous feedback and information collection from the orderer.

The above snack recommendation unit (140) can recommend customized snacks and snack combinations to the orderer using the orderer information and the snack information.

More specifically, the snack recommendation unit (140) can calculate a snack recommendation level for each snack by using the orderer information and the snack information according to the following [Mathematical Formula 2].

[Equation 2]

(Here, S _r represents the snack recommendation, H represents the weight of whether it is healthy, A represents the weight of whether it contains allergens, V represents the weight of whether it contains animal ingredients, m _a represents the weight of preferred packaging, CL represents the weight of calorie information, _ET represents the weight of consumption temperature, C represents the weight of whether it is cooked, T represents the weight of taste classification, B represents the weight of the manufacturer's brand, and t represents the weight of the manufacturing type.)

At this time, the weight (H) for whether the food is healthy may be given a weight of 1 if the health score is indicated as good or average, and 0.3 if the health score is indicated as bad, and the weight for whether the food is allergic may be given a weight of 0 if the snack contains the allergic ingredient of the orderer collected from the orderer information, and 1 if the snack does not contain the allergic ingredient.

In addition, the weight (V) for whether or not the animal ingredient is included may be given a weight of 0 if the snack contains animal ingredients, 2 if it does not contain animal ingredients, and 1 if the orderer is not a vegetarian, if the orderer is a vegetarian.

In addition, the preferred packaging weight (m _a ) may be given a weight of 1 if it matches the packaging material of the snack among the preferred packaging materials classified as paper packaging, plastic packaging, individual packaging, etc. in the orderer information, and a weight of 0.6 if it does not match, and the calorie information weight (CL) may be given a weight according to [Table 4] below by comparing the calorie (kcal) ratio per g of the snack based on the calorie (kcal) per g selected by the orderer.

[Table 4]

For example, if the calories per gram (kcal) selected by the customer is 10 kcal and the calories per gram of the snack is 12 kcal, the calories per gram (kcal) of the snack is 20% more than the calories per gram (kcal) selected by the customer, so a weighting of 0.9 can be given.

In addition, the above-mentioned intake temperature weight (E _T ) refers to the temperature at which the orderer consumes the snack, and at least one item among room temperature, low temperature, and high temperature can be selected, and based on the storage method or intake method described in the snack, it can be classified into room temperature, low temperature, and high temperature, and if it matches, a weight of 1 can be given, and if it does not match, a weight of 0.25 can be given. At this time, in the case of a snack such as cup ramen, which has a storage temperature of room temperature or low temperature but a intake temperature of high temperature (a snack consumed at a temperature of 50℃ or higher, such as by pouring boiling water), it can be classified based on the intake temperature.

In addition, the weighting (C) for whether or not to cook can be given a weight of 1 if the orderer selects either a snack that requires cooking or a snack that can be consumed immediately, and whether or not cooking is required when consuming the snack is classified, and a weighting of 0.5 if the selection matches, and if the selection does not match, a weighting of 1 is given. At this time, the standard for whether or not to cook can be determined based on whether or not the snack can be consumed immediately after opening the packaging upon receipt, and snacks that are consumed using boiling water or a microwave, such as ramen or frozen hot dogs, are classified as snacks that require cooking, and snacks that can be consumed with or without cooking according to the orderer's preference, such as triangle kimbap, can be classified as snacks that do not require cooking.

In addition, the classification weight (T) of the taste can be given a weight of 1 if the taste of the snack classified as one of salty, sweet, and savory tastes matches the information of the orderer who selected at least one taste among salty, sweet, and savory tastes, and a weight of 0.6 if the taste of the snack matches the information of the orderer who selected at least one taste among salty, sweet, and savory tastes.

In addition, the brand weight (B) can be assigned a weight of 1 if the brand information of the manufacturer of the snack preferred by the customer is collected from the customer information, and the brand information of the manufacturer producing the snack is compared, and a weight of 0.8 if the brand information of the manufacturer of the snack is not matched.

In addition, the manufacturing type weight (t) may be obtained by collecting at least one manufacturing type among boiling, oven processing, drying, and fermentation, which is the manufacturing type preferred by the orderer, from the orderer information, and classifying the manufacturing type of the snack as at least one among boiling, oven processing, drying, and fermentation, and assigning a weight of 1 if the manufacturing type matches, and 0.7 if the manufacturing type does not match.

As described above, the snack recommendation unit (140) calculates a snack recommendation score for each snack using the orderer information and the snack information, and displays the intake score (I _s ) calculated by the feedback unit (160) and the snack recommendation score calculated at a preset ratio in ascending order on the user terminal of the orderer, and can perform priority recommendations of snacks to the orderer in order of the highest snack recommendation score.

Meanwhile, the snack recommendation unit (140) can individually recommend at least one snack among challenge snacks, new snacks, tasting snacks, and compatible snacks to the customer.

Here, the challenge snack is a snack whose recommendation level is calculated to be the lowest, and the new snack may mean a snack that does not contain an allergen that causes an allergy to the orderer based on the collected orderer information among snacks produced within 2 months of the first date of production, and does not contain animal ingredients if the orderer is a vegetarian.

Here, the first production date may refer to the time when the snack was first produced as an official product (e.g., January 19, 2022), rather than the manufacturing date.

In addition, the above-mentioned tasting snack may mean a large-capacity snack individually packaged in 100g to 150g, with a minimum packaging weight exceeding a preset weight among the snacks calculated with the highest snack recommendation value.

Here, the large-capacity snack may refer to a snack among the same snack product group whose minimum weight exceeds 1 kg, and the minimum weight classified as large-capacity may be set to at least 500 g.

The above compatibility snack may refer to a preset compatibility snack corresponding to the snack calculated as having the highest snack recommendation value to the orderer.

Here, the preset compatible snack is a snack with the highest recommendation among the snacks recommended to the customer, and when another customer orders the same snack, the snacks ordered together are individually counted, and the snack with the highest count value among the snacks ordered together with the snack collected from the other customers can be set as the compatible snack.

That is, the above-described combination snack may refer to the snack with the highest recommendation rating among the snacks recommended to the customer and the snack that is most frequently ordered at the same time.

For example, for snack A, which received the highest snack recommendation among the snacks recommended to the above-mentioned orderer, when orderer 1, an orderer other than the above-mentioned orderer, orders snacks B, C, D, and E that were ordered together with snack A, they can be counted. At this time, if another orderer, orderer 2, orders snacks C, N, and M that were ordered together with snack A, snack C can be counted the highest (twice) when ordering snack A by ordering orderers 1 and 2, and can be set as a compatible snack.

Through this, the customer can try new types of snacks in small quantities by receiving recommendations for snacks that he or she does not normally prefer, newly released snacks, snacks with a large minimum packaging weight, and snack combinations that other customers consume when ordering the snacks.

The above-mentioned intake information calculation unit (150) divides the snack box delivered to the customer into preset sizes and assigns numbers, and when the customer consumes the snack located at the assigned number, calculates the total calories consumed and displays it on the user terminal.

More specifically, when the system (100) provides snack boxes by packaging snacks of different types according to the order information of the customer, partitions are provided in the snack box according to the quantity of snacks ordered, so that snacks of the same type are gathered and packaged, but the snack box is divided by snack size, and numbers are assigned to each divided section and stored in the system (100), so that the quantity of snacks of each number can be displayed in the customer's dedicated application.

Through this, the orderer receives the snack box, and each time he or she consumes a snack, he or she inputs the amount of reduction (or remaining amount) in the number of snacks displayed in the dedicated application installed on the orderer's user terminal, and the system (100) recognizes that the reduced snack has been consumed by the orderer, and adds up the calories of the snacks recognized as having been consumed and displays them on the orderer's user terminal.

At this time, the total calories can be updated based on 24 hours, and after 24 hours, the previously total calories can be displayed as 0.

For example, if the above orderer receives a snack box consisting of three snacks of A, each containing 70 kcal, in box 1, two snacks of B, each containing 100 kcal, in box 2, and four snacks of C, each containing 150 kcal, in box 3, and consumes two snacks of A and one snack of B on November 10 and enters one snack of A, one snack of B, and four snacks of C in the remaining snack quantity field on the user terminal, the total calories consumed today may be displayed as 240 kcal. Afterwards, if the amount of snack reduction is not entered on October 11, the total calories consumed today may be displayed as 0 on the user terminal, and if only one C snack is consumed on that day and the amount of reduction is entered before October 12, only 150 kcal, which is the calorie content of one C snack, may be displayed as the total calories consumed today.

In addition, the above-mentioned intake information calculation unit (150) can display an alarm such as ‘Caution against excessive snack intake’ when the total calories consumed by the customer exceeds a preset value.

The above feedback unit (160) can assign a score to the individual snack in response to feedback information including an evaluation, score, and repurchase intention corresponding to the individual snack included in the snack box collected from the orderer.

Here, the feedback information may refer to information entered by the orderer through the user terminal.

In addition, the feedback unit (160) can calculate the intake score corresponding to each snack in the snack box received by the customer according to the following [Mathematical Formula 3].

[Equation 3]

(Here, I _s is the intake score, A _s is the evaluation score, R _d is the weight of the intention to reorder, R _s is the rank value, and n is the number of different types of snacks in the snack box.)

At this time, the evaluation score (A _s ) is a score assigned by the orderer using a user terminal between 1 and 10 points, and the reorder intention weight (R _d ) is a weight assigned by the orderer selecting one of 'yes', 'no', and 'I don't know' as to whether he or she intends to reorder the snack through the user terminal, and if the reorder intention is 'yes', a weight of 2 is assigned, if 'no', a weight of 0 is assigned, and if 'I don't know', a weight of 1 is assigned.

In addition, the ranking value (R _s ) can be calculated by having the orderer rank the entire list of snacks in the snack box from 1st to nth using a user terminal, and dividing the ranking by the total number of each _snack .

For example, if there are 10 different snacks in the snack box, the ranking value (R _s ) of the snack ranked 1st by the customer can be calculated as 10 (10/1), and the ranking value of the snack ranked 5th can be calculated as 2 (10/5).

Meanwhile, the feedback unit (160) can calculate a snack recommendation score as a percentage by calculating the calculated intake score (I _s ) and the snack recommendation degree (S _r ) calculated by the snack recommendation unit (140) at a preset rate and transmit the calculated snack recommendation score to the snack recommendation unit (140).

Here, the above preset ratio is the sum ratio of the above intake score (I _s ) and the above snack recommendation (S _r ), and can be basically summed at a ratio of 5:5, but the ratio can be adjusted depending on the work environment of the orderer and the place of order (school, office, kindergarten, etc.).

Through this, the feedback unit (160) can perform priority recommendations to the customer in order of the highest calculated recommendation score.

Meanwhile, the feedback unit (160) can provide a smart bracelet to an orderer who has performed all of the preset input operations requested by the reward provision unit (130) to the orderer's user terminal, and the smart bracelet can be linked to the user terminal to measure the body temperature and heart rate of the orderer, display the health data on the user terminal, and automatically transmit the data to a dedicated application of the system (100) to collect the orderer's health data.

Here, the feedback unit (160) can provide a 'snack intake' item in a dedicated application installed on the user terminal when the orderer consumes the snack, and when the orderer activates the 'snack intake' item when consuming the snack, the body temperature change and heart rate change can be measured when consuming the snack.

At this time, if the heart rate and body temperature changes of the customer exceed a preset rate, the phrase 'suspected allergic reaction' can be displayed on the user terminal, and if the snack consumed by the customer is entered into the dedicated application, the list of ingredients included in the snack can be displayed.

Meanwhile, the smart bracelet may include a fault diagnosis unit (not shown) that determines that an error has occurred in the sensing values collected from the smart bracelet when the average value (A _err ) calculated by the following [Mathematical Formula 4] is greater than a preset limit value (S _err ) by analyzing the sensing values measured through the heart rate measurement sensor and temperature sensor provided in the smart bracelet to determine reliability.

[Equation 4]

Here, A _err represents the average error of the sensing values, T _aver represents the overall average of the sensing values, P _aver represents the partial average of the n-th sensing values of the orderer, and T _σ represents the overall standard deviation of the sensing values.

More specifically, T _aver is the overall average of the sensing values, and means a value calculated by calculating the overall average of the sensing values generated for body temperature or heart rate data measured over a preset period (e.g., one month), and T _σ means a value calculated by calculating the overall standard deviation of the sensing values over the preset period (e.g., one month).

In addition, P _aver is a partial average of sensing value data of n times among the sensing values generated during a preset period (e.g., one month), and in the process of collecting the body temperature sensing value data or the heart rate sensing value data of the customer through the smart bracelet, a preset number (n) of body temperature sensing values or heart rate sensing value data is generated, and the average of the body temperature sensing values or heart rate sensing values of the preset number (n) is calculated, and since it corresponds to an average of some sensing value data, it can be referred to as a partial average.

At this time, if the estimated mean is calculated with 95% confidence using some averages, the estimated mean (μ) is It has a range.

Therefore, the average error (A _err ), which is the difference between the upper or lower limit of the estimated average (μ) and the overall average (T _aver ), can be calculated as in [Mathematical Formula 4] above.

Therefore, since the average error (A _err ) calculated by the above [Mathematical Formula 4] is greater than the preset limit error (S _err ), it means that there is a very high possibility that the body temperature sensor or heart rate sensor collected through the smart bracelet is being input incorrectly due to a malfunction, and therefore, the malfunction diagnosis unit (not shown) can determine that the body temperature or heart rate sensing value data collected through the smart bracelet is not reliable when the above condition is satisfied.

Meanwhile, a fault diagnosis unit (not shown) may compare the average value (A _err ) calculated by [Mathematical Formula 4] with a preset limit value (S _err ) to determine whether there is an error in the body temperature or heart rate sensing value data, and may simultaneously measure the body temperature or heart rate of the orderer to determine whether the measured body temperature or heart rate is outside a preset range, whether the sensing value data shows a pattern that is periodically repeated within a preset section size, and analyze the ratio of the operating time and non-operating time of the body temperature or heart rate sensor corresponding to the activation of the 'snack intake' item of the dedicated application to determine whether the ratio is outside a preset ratio range, and if at least one of the above conditions is satisfied, it may be determined that an error has occurred in the real-time image data.

Below, the specific details of the above error judgment methods are explained in detail.

First, in order to measure the body temperature or heart rate of the user and determine whether the measured body temperature or heart rate is out of a preset range, a temperature sensor capable of measuring the temperature of the smart bracelet is separately provided, and the temperature of the smart bracelet can be monitored in real time using a monitoring unit (not shown) provided in the dedicated application through the temperature sensor. This utilizes the technical principle that the temperature is maintained within an allowable range (preset temperature range) when the smart bracelet is operating normally, and if the temperature of the smart bracelet is excessively high (e.g., 40℃ or higher) or there is no fever at all, it can be assumed that it is overloaded or not operating at all, and in such a case, it can be determined that an error has occurred in the smart bracelet, and this can be assumed to mean that an error has occurred in the body temperature or heart rate sensing value data.

Next, in order to determine whether there is an error in the body temperature or heart rate sensing value data, it is possible to monitor whether the real-time image data shows a pattern that is periodically repeated within a preset interval size. This utilizes the technical principle that if a foreign substance is introduced into the smart bracelet, the body temperature or heart rate sensing value may be repeatedly output in a specific pattern within the preset interval size due to the introduced foreign substance. At this time, the preset interval size may be determined according to [Mathematical Formula 5] below.

[Equation 5]

A _range = {(T _aver + D _max ) - (T _aver - D _max )}*0.3

A _range is a preset interval size, T _aver is the overall average over a preset period, and D _max is the maximum deviation over a preset period.

For example, if the overall average for the heart rate sensing value over a preset period is 70 and the maximum deviation is 20, then A _range is 12, so if values that are repeatedly output periodically within a range where the difference between the maximum and minimum values does not exceed 12 (e.g., similar values such as 52, 62, 52, 62, 53, 62, 52, 63 are repeatedly output), it can be assumed that a foreign substance has entered the smart bracelet, causing an error in the real-time body temperature or heart rate data.

Lastly, the smart bracelet can analyze the ratio of the operating time and non-operating time of the body temperature or heart rate sensor corresponding to the activation of the 'snack intake' item of the dedicated application to determine whether the ratio is outside a preset ratio range, and based on the result, determine that an error has occurred in the real-time image data.

For example, if the heart rate sensor provided in the smart bracelet is set to collect the heart rate sensing value of the customer every 0.1 seconds for every second to save power, the ratio of the operating time to the non-operating time is 10:1, but if the ratio is reversed to 1:10 or changes to a significantly different ratio (e.g., a ratio change of 30% or more) as a result of analysis through real-time monitoring, it can be determined that there is a malfunction of the heart rate sensor in the smart bracelet and an error has occurred in the real-time heart rate sensing value data.

As described above, in one embodiment of the present invention, a fault diagnosis unit (not shown) can determine whether there is an error in the body temperature or heart rate sensor provided in the smart bracelet by considering all of the multiple fault diagnosis methods, and can be finally determined as an S _total value that reflects all of the multiple fault diagnosis methods as in [Mathematical Formula 6] below.

[Equation 6]

S _total = W1*R _aver + W2*R _tem + W3*R _rep + W4*R _hum + W5*R _rat

Here, S _total is the sum of multiple error judgment values, R _aver is the error judgment value derived by comparing the average value (A _err ) calculated by [Mathematical Formula 4] with the preset limit value (S _err ) (one value between 0 and 1 depending on whether there is an error), R _tem is the value that determines whether there is an error depending on whether the temperature is out of the preset range (one value between 0 and 1 depending on whether there is an error), R _rep is a value that determines whether there is an error based on whether a repeating pattern appears (one value between 0 and 1 depending on whether there is an error), R _hum is a value that determines whether there is an error based on whether the humidity is outside the preset range (one value between 0 and 1 depending on whether there is an error), R _rat is a value that determines whether there is an error by analyzing the ratio of the operating time and the non-operating time (one value between 0 and 1 depending on whether there is an error), W1 is the R _aver item weight, W2 is the R _tem item weight, W3 is the R _rep item weight, W4 is the R _hum item weight, and W5 is the R _rat item weight, respectively.

For example, if S _total is 4 or more, it can be preset that there is an error in the body temperature or heart rate sensing value data, and if the fever temperature is a very important parameter, W2 can be preset to 3, and W1, W3, W4, and W5 can all be preset to 1.

Under these conditions, if R _aver is 1, R _tem is 1, R _rep is 0, R _hum is 0, and R _rat is 0 as a result of monitoring whether there is an error according to the above multiple fault diagnosis methods, then S _total becomes 4, and therefore it can be finally determined that there is an error in the body temperature or heart rate sensing value data.

Through the above process, the customized snack recommendation and provision system collects customer information by transmitting and receiving data through a dedicated application provided on the customer's user terminal, and collects snack information for each snack to objectively produce the optimal snack that meets the customer's needs.

In addition, by receiving feedback data from the customer, the more often the customer orders snacks, the more accurate customized snacks can be presented, and based on the customer's information, challenge snacks excluding snacks containing allergens or animal ingredients for vegetarians, new snacks that are new products, tasting snacks that were hesitant to order because they were large in size, and compatible snacks that were ordered at the same time as many other customers ordering snacks can be recommended, thereby encouraging the customer to experience new snacks.

In addition, the same snacks are placed in the snack box delivered to the customer and assigned numbers for each location, so that when the customer inputs the amount of reduction in the snacks consumed on the day through the user terminal, the calorie information consumed as snacks on the day can be calculated and displayed.

In addition, if the customer periodically inputs customer information, evaluates snacks, indicates intent to reorder, and ranks snacks in the snack box through the user terminal, rewards such as discount coupons, priority purchase rights for new products, priority delivery, and exemption from fees may be provided.

In addition, a smart bracelet is provided to an orderer who satisfies certain conditions, and when the orderer activates 'snack intake' through a dedicated application, body temperature and heart rate are measured to check for allergic reactions, and a malfunction diagnosis unit (not shown) is used to diagnose whether the body temperature sensor or heart rate sensor in the smart bracelet is malfunctioning, so that the orderer can prevent confusion about allergic reactions.

According to one embodiment of the present invention, a snack recommendation score corresponding to the orderer's information is scored and displayed on the orderer's terminal, thereby providing recommended snacks based on objective data.

In addition, by collecting information on specific ingredient allergies and vegetarianism from the customer's information and matching it to the nutritional content of the snack, the customer can identify and prevent orders for snacks containing allergens or animal ingredients, thereby reducing human and economic costs due to returns and complaints.

Additionally, by collecting feedback from customers after receiving their snacks, we can provide more sophisticated snack recommendations and optimize the snack supply by identifying snacks with poor public reviews, thereby preventing wasteful inventory.

Although the embodiments of the present invention have been described with limited examples and drawings, the embodiments of the present invention are not limited to the embodiments described above, and various modifications and variations are possible based on this description by those skilled in the art to which the present invention pertains. Therefore, the embodiments of the present invention should be understood solely by the scope of the claims set forth below, and all equivalent or equivalent modifications thereof are deemed to fall within the scope of the present invention.

100: Customized Snack Recommendation and Delivery System
110: Customer Information Collection Department
120: Snack Information Collection Department
130: Reward Provider
140: Snack Recommendation Section
150: Intake information production section
160: Feedback Department

Claims (5)

In a customized snack recommendation and provision system that recommends snacks to a customer and packages and provides multiple snacks selected by the customer,
An orderer information collection unit that collects at least one of the orderer's gender, age, order history, allergies, preferred taste, preferred texture, diet, preferred nutrients, preferred food temperature, preferred packaging material, preferred brand, previously selected snack, and snack consumption completion time;
A snack information collection unit that collects at least one snack information among the category, nutritional content, food type, brand, raw material, and calorie information of the above snack;
A reward provision unit that provides a reward to the orderer when the orderer performs a preset input operation using a user terminal connected to the system by wireless communication;
A snack recommendation unit that recommends customized snacks and snack combinations using the above customer information and the above snack information;
A consumption information calculation unit that divides the snack box delivered to the customer into preset sizes and assigns numbers, and calculates the total calories consumed when the customer consumes the snack located at the number and displays the calculated total calories consumed on the user terminal; and
A feedback unit that assigns a score to the snack in response to feedback information including evaluations, scores, and repurchase intentions corresponding to the snacks included in the snack box collected from the customer;
Including,
The above snack information collection department is,
The text in the nutritional information label of the above snack is converted into text and stored as data using OCR technology or a deep learning-based text recognition algorithm in the image.
Using the saved nutritional information table, calculate the health score according to [Mathematical Formula 1].
[Mathematical Formula 1]

(Here, H _p is the health score, P _r is the protein content (g), V _i is the vitamin content weight, D _f is the dietary fiber content weight, K _c is the calories per 1g (kcal), _Na is the sodium per 1g (mg), S _g is the sugar per 1g (g), F _t is the fat per 1g (g), and C _a is the carbohydrate per 1g (g).)
The health score corresponding to the snack selected by the customer is displayed on the user terminal as at least one of bad, average, and good according to preset criteria.
The above snack recommendation section is,
Using the above customer information and the above snack information, the snack recommendation is calculated according to [Mathematical Formula 2].
[Equation 2]

(Here, S _r represents the snack recommendation, H represents the weight of whether it is healthy, A represents the weight of whether it contains allergens, V represents the weight of whether it contains animal ingredients, m _a represents the weight of preferred packaging, CL represents the weight of calorie information, _ET represents the weight of consumption temperature, C represents the weight of whether it is cooked, T represents the weight of taste classification, B represents the weight of the manufacturer's brand, and t represents the weight of the manufacturing type.)
The above snack recommendation section is,
At least one snack among challenge snacks, new snacks, tasting snacks, and compatibility snacks is individually recommended to the above orderer.
The above challenge snack is the snack with the lowest recommended value.
The above new snack is a snack that does not contain any allergens corresponding to the collected customer information among snacks produced within 2 months of the first date of manufacture, and if the customer is a vegetarian, does not contain any animal ingredients.
The above tasting snack is a large-capacity snack individually packaged in 100g to 150g, with a minimum packaging weight exceeding a preset weight among the snacks calculated with the highest recommendation value.
The above compatibility snack provides the preset compatibility snack corresponding to the customized recommended snack calculated with the highest recommendation rate for the above snack to the above orderer.
The above feedback section,
The intake score corresponding to each snack in the above snack box is calculated according to [Mathematical Formula 3] below,
[Equation 3]

(Here, I _s is the intake score, A _s is the evaluation score, R _d is the weight of the intention to reorder, R _s is the rank value, and n is the number of different types of snacks in the snack box.)
The snack recommendation score calculated by the above intake score and the above snack recommendation score at a preset ratio is transmitted to the snack recommendation section.
The above snack recommendation section is,
A personalized snack recommendation and provision system characterized in that the snacks are recommended to the customer in order of the highest snack recommendation score. delete delete delete delete