KR20210032759A - Food recommendation system and method using individual taste subjectivity analysis - Google Patents

Abstract

The present invention relates to a food recommendation system and method using subjectivity analysis of personal taste, which can recommend appropriate food by analyzing subjectivity to personal taste. The system may comprise: a communication unit acquiring an evaluation value for personal taste associated with a plurality of foods, or receiving one or more food menus from an individual; and a processor extracting subjective variables for personal taste based on the evaluation value of the personal taste and performing a process of recommending at least one food menu suitable for personal taste based on the subjective variables for personal taste with respect to a food menu inputted by the individual.

Description

Food recommendation system and method using subjectivity analysis of personal taste {FOOD RECOMMENDATION SYSTEM AND METHOD USING INDIVIDUAL TASTE SUBJECTIVITY ANALYSIS}

The present invention relates to a food recommendation system and method, and more particularly, to a food recommendation system and method using subjectivity analysis of personal taste that can recommend appropriate food by analyzing subjectivity with respect to personal taste.

Due to the rapid development of information and communication technologies including the Internet and smartphones, a number of various services have been created, including web services as well as smartphone application services.By building a database of places that individuals want to visit, such as restaurants, travel destinations, and accommodations, Services have also emerged that allow them to be searchable.

Recently, as the interest in gourmet has increased, not only cooking rooms (cooking broadcasts) and eating rooms (eating broadcasts) have become popular on TV, as well as services that provide a variety of services and contents such as ordering delivery food, delivering food ingredients, sharing recipes, and recommending restaurants. It is attracting attention.

However, most of the services provide a list of places or restaurants to individuals as search results that match contextual information such as a keyword or location entered by an individual, so food that suits their taste is not recommended. .

Accordingly, there is a need for a new service capable of recommending foods optimized for individual tastes by analyzing individual tastes.

KR10-2014-0065384A1

Accordingly, the present invention has been conceived to solve the above problem, and an object of the present invention is to provide a food recommendation system and method using subjectivity analysis of personal taste that can recommend appropriate food by analyzing subjectivity with respect to personal taste. .

Other objects and advantages of the present invention will be described below, and will be understood by examples of the present invention. Further, the objects and advantages of the present invention can be realized by means and combinations indicated in the claims.

In order to achieve the above object, a food recommendation method using subjectivity analysis of personal tastes according to the present invention includes: supporting, by a communication unit of a computing device, to acquire evaluation values for personal tastes associated with a plurality of foods; Extracting, by a processor of the computing device, subjectivity variables for personal taste based on an evaluation value for the personal taste; Supporting the communication unit to receive one or more food menus from an individual; And recommending, by the processor, at least one food menu suitable for personal taste based on subjectivity variables for personal taste with respect to the food menu input by the individual.

In addition, subjectivity variables for personal taste are extracted using a method of selecting the similarity of the Euclidean distance between foods.

In addition, the processor extracting, by the processor, p hidden variables representing a degree of correlation between the subjectivity variables by using a pre-learned estimation model, wherein p is 1 or a natural number greater than that; It is characterized by that.

In addition, the processor, by comparing the Euclidean distances at the centers of k food menu groups grouped in advance with the hidden variables, to recommend food suitable for individual taste, wherein k is 1 or a natural number greater than that. It characterized in that it further includes;

In addition, when an individual directly inputs a taste expression associated with a plurality of foods, the method further comprises pre-processing the taste expression using a TF-IDF method.

In addition, the food menu group is characterized by consisting of special meals, stew, rice bowl & fried rice, noodles, soup/tang, convenience meals, and others.

In addition, the special meal includes bulgogi, steamed chicken, chicken stir-fried soup, steak, spring roll, suyuk, agu jjim, ssambap, and stir-fried octopus, and the jjigae includes doenjang stew, kimchi stew, bag stew, sundubu stew, cheonggukjang stew, Dongtae stew , Biji stew, red pepper paste stew, and squid stew, and the rice bowl & fried rice includes curry, bibimbap, omelet rice, kimchi fried rice, jeyuk rice bowl, salmon rice bowl, chicken mayo rice bowl, donburi, and squid rice bowl, and the noodles Includes ramen, tomato spaghetti, cream spaghetti, naengmyeon, feast noodles, bibim noodles, kalguksu, udon, and bean noodles, and the soup/tang includes yukgaejang, chicken dogjang, rice cake soup, beef musk soup, siraegit soup, galbitang, chueotang, samgyetang, and Daegutang Including, the convenience meal includes a sandwich, French toast, tteokbokki, cereal, salad, rice burger, hot dog, convenience store lunch box, kimbap, and tofu sushi, and the others include sushi, jajangmyeon, champon, kanpoonggi, mapo tofu, japchae , Pad Thai, and rice noodles, characterized in that it comprises a hamburger.

According to another aspect of the present invention, there is provided a food recommendation system using subjective analysis of individual tastes, comprising: a communication unit that obtains evaluation values for personal tastes associated with a plurality of foods or receives one or more food menus from an individual; And a process of extracting subjectivity variables for personal taste based on the evaluation value for personal taste, and at least one food menu suitable for personal taste based on subjectivity variables for personal taste for the food menu input by the individual. And a processor that performs the recommended process.

As described above, according to the present invention, it is possible to recommend food optimized for personal taste by analyzing subjectivity for personal taste.

1 is a block diagram showing an exemplary configuration of a computing device that performs a food recommendation method using subjectivity analysis of personal taste according to the present invention.
2 is a flowchart illustrating a food recommendation method using subjectivity analysis of personal taste according to the present invention.
3 is a view showing an example of recommending food suitable for personal taste according to an embodiment of the present invention.
4 to 6 are diagrams showing examples of food menu groups according to an embodiment of the present invention.

Details of other embodiments are included in the detailed description and drawings.

Advantages and features of the present invention, and a method of achieving them will become apparent with reference to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms. In the following description, when a certain part is connected to another part, this is not only a case in which it is directly connected. It also includes the case where it is connected through another medium in the middle. In addition, parts not related to the present invention in the drawings are omitted to clarify the description of the present invention, and like reference numerals are attached to similar parts throughout the specification.

Hereinafter, the present invention will be described with reference to the accompanying drawings.

1 is a block diagram showing an exemplary configuration of a computing device that performs a food recommendation method using subjectivity analysis of personal taste according to the present invention.

Referring to FIG. 1, the computing device 100 includes a communication unit 110 and a processor 120, and may directly or indirectly communicate with another computing device (not shown) through the communication unit 110. The communication unit 110 may include a communication interface. For example, communication interfaces include WLAN (Wireless LAN), WiFi (Wireless Fidelity) Direct, DLNA (Digital Living Network Alliance), Wibro (Wireless broadband), Wimax (World Interoperability for Microwave Access), HSDPA (High Speed Downlink Packet Access), etc. Wireless Internet interface and short-range communication interfaces such as Bluetooth™, Radio Frequency Identification (RFID), Infrared Data Association (IrDA), Ultra Wideband (UWB), ZigBee, and Near Field Communication (NFC). I can. In addition, the communication interface may represent all interfaces (eg, wired interfaces) capable of performing communication with the outside.

Specifically, the computing device 100 is a device that may include typical computer hardware (eg, computer processor, memory, storage, input devices and output devices, and other components of existing computing devices; routers, switches, etc.). Electronic communication devices; electronic information storage systems such as network-attached storage (NAS) and storage area networks (SANs) and computer software (i.e., allowing the computing device to function in a specific way). Instructions) may be used to achieve the desired system performance.

The communication unit 110 of such a computing device can transmit and receive requests and responses to and from other computing devices to which it is linked. As an example, such requests and responses may be made by the same transmission control protocol (TCP) session. However, the present invention is not limited thereto, and may be transmitted/received as, for example, a user datagram protocol (UDP) datagram. In addition, in a broad sense, the communication unit 110 may include a keyboard, a mouse, other external input devices, printers, displays, and other external output devices for receiving commands or instructions.

In addition, the processor 120 of the computing device includes a micro processing unit (MPU), a central processing unit (CPU), a graphics processing unit (GPU), a neural processing unit (NPU), or a tensile processing unit (TPU), and a cache memory. ), data bus, and the like. In addition, it may further include an operating system and a software configuration of an application that performs a specific purpose.

The communication unit 110 is configured to obtain evaluation values for personal tastes associated with a plurality of foods, or to receive one or more food menus from an individual.

In addition, the processor 120 is a process of extracting subjectivity variables for personal taste based on the evaluation value for personal taste, and for food menus input by the individual, suitable for personal taste based on subjectivity variables for personal taste. A process of recommending at least one food menu may be performed.

Here, the subjectivity variables for personal taste may be extracted using a method of selecting the similarity of the Euclidean distance between foods, as described later.

2 is a flowchart illustrating a food recommendation method using subjectivity analysis of personal taste according to the present invention.

Referring to FIG. 2, first, the communication unit 110 of the computing device supports to obtain an evaluation value for personal taste associated with a plurality of foods (S100).

Here, the evaluation value for personal taste associated with a plurality of foods has the same meaning as the subjectivity variables for personal taste related to the food menu to be described later, for example, in addition to sweet, sour, salty, bitter, and spicy, There may be temperature, fragrance, color, price (upper, medium, or lower), but is not limited thereto.

For example, each individual may assign an evaluation value to sweet, sour, salty, bitter, spicy, and temperature with respect to individual tastes associated with a plurality of foods. At this time, the evaluation value may be a value in the range of 1-5.

For example, referring to FIG. 4, for the bulgogi menu, it can be seen that as an evaluation value for individual taste, sweet taste 5, sour taste 1, salty taste 3, bitter taste 1, spicy taste 1, temperature 4 were evaluated, and these values are It may be stored in the database in the form of a food menu group (eg, special meals, stew, rice bowl & fried rice, etc.) (see FIGS. 4 to 6)

Next, the processor 120 of the computing device extracts subjectivity variables for personal taste based on the evaluation value for the personal taste (S110).

Next, the communication unit 110 receives one or more food menus from an individual (S120).

Next, the processor 120 recommends at least one food menu suitable for personal taste based on subjectivity variables for personal taste with respect to the food menu input by the individual (S130).

Here, the subjectivity variables for personal taste may be extracted using a method of selecting the similarity of the Euclidean distance between foods.

To this end, the processor 120 extracts p hidden variables representing the degree of correlation between the subjectivity variables by using a pre-learned estimation model. In this case, p is 1 or a natural number greater than that.

Specifically, the computing device may extract the hidden variable by using the analysis model as shown in Equation 1 below.

... (1)

... (One)

Here, xil is the l-th subjectivity variable of the i-th individual, λlp is the load on the p-th hidden variable of the l-th subjectivity variable, ylp is the p-th hidden variable of the l-th individual, and uil is explained by other hidden variables. It is a specificity constant that cannot be observed and represents an unobservable number that only affects the i-th individual.

In addition, the processor 120 may recommend food that suits a personal taste by comparing the hidden variables with the Euclidean distance at the center of the k food menu groups grouped in advance. In this case, k is 1 or a natural number greater than that.

3 is a view showing an example of recommending food suitable for personal taste according to an embodiment of the present invention.

Referring to FIG. 3, the communication unit 110 receives one or more food menus from an individual.

Next, the processor 120 recommends at least one food menu suitable for personal taste based on subjectivity variables for personal taste with respect to the food menu input by the individual.

For example, when an individual selects steamed chicken or chicken fried meat, at least one food menu suitable for personal taste based on subjectivity variables for personal taste for steamed chicken and chicken fried meat Dongtae stew, biji stew, kanpoonggi, mapo tofu, tomato spaghetti, curry, bibimbap, pad thai, etc.) can be recommended. At this time, subjectivity variables for individual tastes for steamed chicken and chicken stir-fry may be averaged.

For example, referring to FIG. 3, when the subjectivity variable value of sweetness of steamed chicken is 4 and the subjectivity variable value of sweetness of Dakbokkeumtang is 2, 3 may be output as the average value.

In addition, when the subjective variable value of the sour taste of steamed chicken is 1 and the subjective variable value of the sour taste of Dakbokkeum-tang is 1, 1 may be output as the average value.

In addition, when the subjectivity variable value of the salty taste of steamed chicken is 4 and the subjective variable value of the salty taste of Dakbokkeum-tang is 3, 4 may be output as the average value.

In addition, when the subjectivity variable value of the spicy taste of steamed chicken is 3 and the subjectivity variable value of the spicy taste of Dak-bokkeum-tang is 5, 4 may be output as the average value.

Here, the subjectivity variables for personal taste associated with the food menu may include temperature, aroma, color, price (high, medium, low) in addition to sweet, sour, salty, bitter, and spicy, but is not limited thereto.

4 to 6 are diagrams showing examples of food menu groups according to an embodiment of the present invention.

4 to 6, the food menu group may include special meals, stew, rice bowl & fried rice, noodles, soup/tang, convenience meals, and others.

The special meal includes bulgogi, steamed chicken, chicken stir-fry, steak, spring rolls, suyuk, agu jjim, ssambap, and stir-fried octopus, and the jjigae includes doenjang stew, kimchi stew, bag stew, sundubu stew, cheonggukjang stew, Dongtae stew, biji Including stew, red pepper paste stew, and squid stew, and the rice bowl & fried rice includes curry, bibimbap, omelet rice, kimchi fried rice, jeyuk rice bowl, salmon rice bowl, chicken mayo rice bowl, donburi, and squid rice bowl, and the noodles include ramen, Tomato spaghetti, cream spaghetti, naengmyeon, feast noodles, bibim noodles, kalguksu, udon, and bean noodles, and the soup/tang includes yukgaejang, chicken dogjang, rice cake soup, beef mud soup, siraegit soup, galbitang, chueotang, samgyetang, and Daegutang. And, the convenience meal includes a sandwich, French toast, tteokbokki, cereal, salad, rice burger, hot dog, convenience store lunch box, kimbap, and tofu sushi, and the others include sushi, jajangmyeon, champon, kanpoonggi, mapo tofu, japchae, pot Thai, and rice noodles, may include hamburgers.

On the other hand, instead of assigning evaluation values to values in the range of 1 to 5 for individual tastes associated with a plurality of foods, sweet, sour, salty, bitter, spicy, and temperature, an individual is associated with a plurality of foods. You can directly enter the taste expression.

In this case, the processor 120 may pre-process the taste expression directly input by the individual using a TF-IDF (Term Frequency-Inverse Document Frequency) method.

That is, if the taste expression is different for individual tastes related to food, it is impossible to select and score only specific words. Therefore, it assists the analysis to be performed in the right direction by grouping words with similar meanings.

TF counts how often a specific word appears in the document, and the higher the value is, the more important the word is, and the importance of the word in the document is as shown in Equation 2.

...(2)

...(2)

는 해당 문서(

)에서 특정한 단어(

)가 출현하는 횟수를 의미하며,

는 해당 문서(

)에서 출현하는 모든 단어 수를 말한다.here

Is the corresponding document (

) In a specific word (

Means the number of occurrences of ),

Is the corresponding document (

) Is the number of all words that appear in.

However, just because the TF value is high, the word cannot be judged as a keyword of the document. This is because, when constructing a sentence, the weight of surveys connecting words and words increases, or common words may emerge as keywords in the sentence. In order to prevent this and find key words, IDF (reverse document frequency) is used, which is the reciprocal of the number of counts and the logarithm of how many unique documents a particular word came from.

...(3)

...(3)

는 전체 문서의 개수이며,

는 특정한 단어가 나오는 문서의 개수를 의미한다.here,

Is the total number of documents,

Means the number of documents in which a specific word appears.

Through these two methods, the TF-IDF value is finally obtained. The higher the frequency of specific words in the document, and the smaller the number of documents including specific words among all documents, the higher the TF-IDF value.

In order to give a specific score through TF-IDF, a TermDocument Matrix is needed to create a bunch of words and organize them by the frequency of occurrence of words. To this end, words having similar meanings need to be grouped together, because if words having similar meanings are not grouped together, analysis may be performed in the wrong direction.

In the present invention, words of taste expressions having similar meanings are bundled, and the bundled words are as follows.

-Expression of sweetness

Before treatment: sweet; so sweet; Sweet; Sweet; Sweet; It's bitter and bitter

After treatment: sweet

-Expression of sour taste

Before treatment: sour; Sour; Sour; Sour hairy; Sweet and sour; Sour; It's noisy; Sour

After treatment: Sida

-Expression of salty taste

Before treatment: squeeze; Spicy; Salty; Slightly salty; Grumpy; Salty; Intermittently

After treatment: weave

-Expression of bitter taste

Before treatment: bitter; Bitter; Bitter; Bitter; Bitter

After treatment: write

-Expression of spicy taste

Before processing: map demap; Spicy; Spicy; Spicy; Spicy

After treatment: spicy

As described above, according to the present invention, it is possible to recommend food optimized for personal taste by analyzing subjectivity with respect to personal taste.

Those of ordinary skill in the art to which the present invention pertains will appreciate that the present invention can be implemented in other specific forms without changing the technical spirit or essential features thereof. Therefore, it should be understood that the embodiments described above are illustrative and non-limiting in all respects. The scope of the present invention is indicated by the scope of the claims to be described later rather than the detailed description, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts are included in the scope of the present invention. It must be interpreted.

100 computing devices
110 Department of Communications
120 processors

Claims (8)

As a food recommendation method using subjectivity analysis of personal taste,
Supporting, by a communication unit of the computing device, obtaining evaluation values for personal tastes associated with a plurality of foods;
Extracting, by a processor of the computing device, subjectivity variables for personal taste based on an evaluation value for the personal taste;
Supporting the communication unit to receive one or more food menus from an individual; And
The processor recommending, by the processor, at least one food menu suitable for personal taste based on subjectivity variables for personal taste with respect to the food menu input by the individual; food using subjectivity analysis of personal taste, comprising: Recommended way. The method of claim 1,
A food recommendation method using subjectivity analysis of individual tastes, characterized in that the subjectivity variables for individual tastes are extracted using a method of selecting the Euclidean distance similarity between foods. The method of claim 2,
Extracting, by the processor, p hidden variables representing a degree of correlation between the subjectivity variables by using a pre-learned estimation model, wherein p is 1 or a natural number greater than that; Food recommendation method using subjectivity analysis of personal taste. The method of claim 3,
The processor, by comparing the hidden variables and the Euclidean distances of the centers of k food menu groups grouped in advance, to recommend food suitable for personal taste, wherein k is 1 or a natural number greater than that; Food recommendation method using subjectivity analysis of personal taste, characterized in that it further comprises. The method of claim 1,
When an individual directly inputs a taste expression associated with a plurality of foods, the method of recommending food using subjectivity analysis of individual tastes, further comprising pre-processing the taste expression using a TF-IDF method. The method of claim 1,
The food menu group is a food recommendation method using subjective analysis of personal taste, characterized in that consisting of special meals, stew, rice bowl & fried rice, noodles, soup/tang, convenience meals, and others. The method of claim 1,
The special meal includes bulgogi, steamed chicken, chicken stir-fried soup, steak, spring rolls, suyuk, steamed agu, ssambap, and stir-fried octopus,
The stew includes doenjang stew, kimchi stew, bag stew, sundubu stew, cheonggukjang stew, dongtae stew, biji stew, red pepper paste stew, and squid stew,
The rice bowl & fried rice includes curry, bibimbap, omelet rice, kimchi fried rice, jeyuk rice bowl, salmon rice bowl, chicken mayo rice bowl, donburi, and squid rice bowl,
The noodles include ramen, tomato spaghetti, cream spaghetti, naengmyeon, feast noodles, bibim noodles, kalguksu, udon, and bean noodles,
The soup/tang includes yukgaejang, chicken gaejang, rice cake soup, beef musk soup, siraegit soup, galbitang, chueotang, samgyetang, and Daegutang,
The convenience meal includes a sandwich, French toast, rice cake, cereal, salad, rice burger, hot dog, convenience store lunch box, kimbap, and tofu sushi,
The other food recommendation method using subjective analysis of personal taste, characterized in that it includes sushi, jajangmyeon, champon, kanpoonggi, mapo tofu, japchae, pad thai, and rice noodles, hamburgers. As a food recommendation system using subjectivity analysis of individual tastes,
A communication unit that obtains evaluation values for personal tastes associated with a plurality of foods, or receives one or more food menus from an individual; And
The process of extracting subjectivity variables for personal taste based on the evaluation value for personal taste, and recommending at least one food menu suitable for personal taste based on subjectivity variables for personal taste for the food menu entered by the individual A food recommendation system using subjectivity analysis of personal taste, comprising: a processor that performs a process to perform a process.

Images