KR102534837B1 - Methods and systems for marketing online goods and services - Google Patents

Abstract

According to various embodiments of the present disclosure, an online product marketing server that markets online products registered from a seller terminal to buyers includes a product information collection unit that obtains product information of online products from the seller terminal; an interest data generator configured to obtain behavioral data of the purchaser with respect to the online product and generate interest data of the purchaser with respect to the online product based on the acquired behavioral data; A marketing management unit providing a customized marketing service for the purchaser based on the interest data may be included.

Description

Marketing methods and systems for online products {METHODS AND SYSTEMS FOR MARKETING ONLINE GOODS AND SERVICES}

The present invention relates to a marketing method and system for an online product, and more particularly, to a technique for analyzing data on a buyer who purchases an online product and recommending or marketing an online product based thereon.

Unless otherwise indicated herein, material described in this section is not prior art to the claims in this application, and inclusion in this section is not an admission that it is prior art.

As the digital industry develops and the Internet infrastructure expands, various studies have been conducted on electronic commerce for selling and purchasing products using the Internet or PC communication. For example, the market size of e-commerce in 2015 is 1,548 billion worldwide, showing an annual growth rate of about 15% until 2018, of which product sales platforms account for a significant portion. In particular, the open market, a type of product sales platform, is not only highly accessible to buyers and sellers, but also offers products with high cost-effectiveness by omitting intermediate distribution margins, resulting in exponential demand and application areas. is increasing with

Consumers can easily purchase products of various items due to the recent vitalization of online shopping malls. Sellers can also easily promote their products online, realizing profits in various forms. In order to purchase a desired product, a consumer checks product information in an online shopping mall or uses product purchase evaluation information of other consumers. Consumers can make various choices in purchasing products through a wide range of information, and sometimes select products according to the services provided by online shopping malls.

A seller who sells a product through an online shopping mall may analyze a product having a large sales volume for each product, and may perform marketing for each product based on the analyzed sales volume. However, if a marketing strategy is established based on analysis of simple elements such as sales volume, a wide range of marketing can be performed, and targeting marketing for a specific online product to a specific buyer may be insufficient. Therefore, there is a need for a method to perform marketing by analyzing information about the product itself and behavioral characteristics of the buyer.

An object of the present invention to solve the above problems is to provide a customized online product marketing method by generating interest data of a buyer for an online product registered in a server.

Another object of the present invention for solving the above problems may further provide objects directly or indirectly identified through this document.

According to various embodiments of the present disclosure, an online product marketing server that markets online products registered from a seller terminal to buyers includes a product information collection unit that obtains product information of online products from the seller terminal; an interest data generator configured to obtain behavioral data of the purchaser with respect to the online product and generate interest data of the purchaser with respect to the online product based on the acquired behavioral data; A marketing management unit providing a customized marketing service for the purchaser based on the interest data may be included.

According to various embodiments, the number of times the buyer purchases the first online product registered in the server, the number of times the buyer views the image of the first online product, and the page where the first online product is registered by the buyer The degree of product interest in the first online product of the seller is calculated based on the duration of the stay, the number of times the buyer searches for the first online product name, and the number of times the buyer views reviews of the first online product. can

According to various embodiments of the present disclosure, the marketing management unit determines the first online product of the seller for which the buyer's interest in the product is greater than or equal to a preset first threshold interest level, and determines another online product of the seller that is similar to the determined first online product. Among them, second online products similar to the determined first online product may be recommended to the purchaser.

According to various embodiments of the present disclosure, the server further includes a similar product determining unit configured to determine the first online product and another online product of the seller similar to the first online product, wherein the similar product determining unit determines the seller's other online product. 1 determining similarities between product keywords of online products and product keywords of other second online products of the seller to determine proximate products, and product images of other second online products determined to be proximate products among the other online products; and determining the adjacent product as a similar product based on the image similarity of the product image of the first online product.

According to various embodiments of the present disclosure, the marketing management unit generates sales data for another second online product of the seller determined to be the similar product corresponding to the first online product of the seller, and the first online product of the buyer. An advertising efficiency index for an advertisement marketing the second online product to the buyer may be calculated based on the degree of interest in the product and the sales volume of the product for the second online product.

According to various embodiments disclosed in this document, customized marketing may be provided based on buyer behavior data.

Also, according to various embodiments, the marketing model may be re-evaluated by determining the effectiveness of advertisements according to customized marketing.

In addition to this, various effects identified directly or indirectly through this document may be provided.

1 is a diagram illustrating an online product marketing system according to an exemplary embodiment.
Figure 2 is a diagram showing the main components of the online product marketing server.
3 is a view related to marketing of other products of a seller by generating interest data of a buyer.
4 is a process of determining proximate products by comparing product keywords of a first online product with product keywords of other online products of a seller, and determining similar products by comparing product images of proximate products with product images of the first online product. is a drawing showing
FIG. 5 is a diagram showing the hardware configuration of the online product marketing server according to FIG. 1 .

Since the present invention can make various changes and have various embodiments, specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, and should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention. Like reference numerals have been used for like elements throughout the description of each figure.

Terms such as first, second, A, and B may be used to describe various components, but the components should not be limited by the terms. These terms are only used for the purpose of distinguishing one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element, without departing from the scope of the present invention. The terms and/or include any combination of a plurality of related recited items or any of a plurality of related recited items.

It is understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, but other elements may exist in the middle. It should be. On the other hand, when an element is referred to as “directly connected” or “directly connected” to another element, it should be understood that no other element exists in the middle.

Terms used in this application are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, the terms "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the present application, they should not be interpreted in an ideal or excessively formal meaning. don't

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram illustrating an online product marketing system 10 according to an embodiment. Referring to FIG. 1 , the online product marketing system 10 may include an online product marketing server 100, a buyer terminal 200, and the like. Operations described below may be performed or implemented through a platform (eg, a web page and/or application) controlled by the online product marketing server 100 . In other words, the online product marketing server 100 is a website through which a buyer can access the online product marketing server 100 through a network using the purchaser terminal 200 to input, register, and output various information. It is possible to provide, and by being installed and executed in the purchaser terminal 200, it is possible to provide an application capable of inputting, registering, and outputting various information.

The online product marketing server 100 obtains product information of online products from the seller terminal 300, receives a registration request for online products, and registers online products sold by the seller of the seller terminal 300 in response to the registration request. can do. The online product marketing server 100 may be, for example, an online commerce market server that brings together various sellers, or a platform server that registers and sells only online products of a specific seller, and its form is limited. It may not be.

When recommending the online products, the online product marketing server 100 may recommend specific online products to the buyer by arranging the online products according to a specific algorithm on the interface. The online product marketing server 100 obtains real-time product category ranking information within the online product marketing server 100 and recommends products corresponding to a higher rank in the real-time product category ranking to a purchaser by referring to the real-time product category ranking. Similar products similar to products corresponding to the top ranks may be determined and recommended to the purchaser, or a specific online product may be recommended to the purchaser based on the purchaser's behavioral data regardless of the category ranking.

The purchaser terminal 200 is a terminal that purchases and pays for a product, and allows the buyer to connect to the online product marketing server 100, search for a product in real time, pay for and purchase the searched product. The online product marketing server 100 enables the seller's product to be marketed, and through this, the seller's online product can be promoted or recommended to the buyer so that the transaction can be promoted.

The purchaser terminal 200 may obtain an input for purchase processing and output a processing result corresponding to the input for purchase processing through an output device such as a display. For example, the purchaser terminal 200 may display products provided by the online product marketing server 100 through a display or the like, and obtain a touch input of a purchaser purchasing at least one product among the displayed products. The buyer may purchase a product corresponding to the buyer's touch input through the online product marketing server 100 .

The purchaser terminal 200 is a communicable desktop computer, a laptop computer, a notebook, a smart phone, a tablet PC, a mobile phone, and a smart watch. (smart watch), smart glass, e-book reader, PMP (portable multimedia player), portable game console, navigation device, digital camera, DMB (digital multimedia broadcasting) player, digital voice It may be a digital audio recorder, a digital audio player, a digital video recorder, a digital video player, and a personal digital assistant (PDA).

The seller terminal 300 is a terminal used by the seller, and may be a normal terminal having a computing function, such as a PC, laptop, netbook, smart phone, tablet, or tablet PC. The seller terminal 300 may request the online product marketing server 100 to register products to be sold. When the online product marketing server 100 approves the request, the seller terminal 300 may provide product information of the product to the online product marketing server 100 . The product information may include a product name, a product price, a promotional text of the product, a product keyword, and a product image. Hereinafter, a description of a seller performing a certain task may be understood as a task performed through the seller terminal 300 .

The online product marketing server 100, the buyer terminal 200, and the seller terminal 300 are connected to a communication network, respectively, and may transmit/receive data with each other through the communication network. For example, communication networks include Local Area Network (LAN), Metropolitan Area Network (MAN), Global System for Mobile Network (GSM), Enhanced Data GSM Environment (EDGE), High Speed Downlink Packet Access (HSDPA), and W-CDMA. (Wideband Code Division Multiple Access), CDMA (Code Division Multiple Access), TDMA (Time Division Multiple Access), Bluetooth, Zigbee, Wi-Fi, Voice over Internet Protocol (VoIP) , LTE Advanced, IEEE802.16m, WirelessMAN-Advanced, HSPA+, 3GPP Long Term Evolution (LTE), Mobile WiMAX (IEEE 802.16e), UMB (formerly EV-DO Rev. C), Flash-OFDM, iBurst and MBWA (IEEE Various types of wired or wireless networks may be used, such as 802.20) systems, HIPERMAN, Beam-Division Multiple Access (BDMA), World Interoperability for Microwave Access (Wi-MAX), and 5G.

2 is a diagram showing the main components of the online product marketing server 100 . The online product marketing server 100 may include a product information collection unit 101, an interest data generation unit 102, a marketing management unit 103, a similar product determination unit 104, and the like.

The product information collecting unit 101 may obtain product information about online products from the seller terminal 300 . The product information may include a product name, a product price, a promotional text of the product, a product keyword, and a product image. The product information collecting unit 101 may classify categories of online products based on the obtained product information.

The product information collecting unit 101 may obtain search inputs of buyers and analyze the search inputs of buyers. The product information collecting unit 101 may classify search data into a plurality of categories through analysis of search inputs of buyers. For example, when a search input such as cargo pants or slacks is obtained, the product information collecting unit 101 accumulates search data as a bottom category, and when a search input such as a twisted knit or a spring cardigan is obtained, the top category Search data can be accumulated as

The product information collecting unit 101 may generate real-time product category ranking information based on the accumulated search data. The product information collection unit 101 may obtain the generated real-time product category ranking information and product information of a product corresponding to the real-time product category ranking.

The product information collecting unit 101 may obtain real-time product ranking information indicating a product ranking for each online product within the classified category and product information of online products corresponding to the real-time product ranking information. For example, the item category may include tops, bottoms, hats, shoes, underwear, watches, jewelry, and the like.

The product information collection unit 101 can search or crawl products identical to online products registered by the seller in an external server, and consists of product keywords for promoting the same products registered in the external server. Information on promotional phrases may be obtained. The external server may include a plurality of external product distribution servers. For example, a plurality of external product distribution servers may include search engine servers, SNS servers, other platform servers, and the like. For example, search engine servers may be Naver, Daum, Nate, and the like. The SNS server may be Twitter, Instagram, Facebook, Kakao Band, or the like.

The product information collection unit 101 may determine product keywords constituting the promotional text. For example, product keywords may include keywords such as daily wear, daily wear, bending, camo, spandex, cargo, check, jogger, semi, full, and cotton. For example, the product image is an image for visually promoting the product to buyers, and may include a picture of wearing the corresponding online product and/or a front, side, or back picture of the corresponding online product.

3 is a view related to marketing of other products of a seller by generating interest data of a buyer.

The interest data generating unit 102 may obtain behavioral data of the buyer in the online product marketing server 100 . Behavioral data may be data on online work performed by the buyer after accessing the online product marketing server 100 . For example, an online action may be an explicit action such as clicking on a product, looking up a review, adding or deleting a shopping cart, attempting a payment, entering a search term, clicking on an advertisement, or social activity such as liking or sharing. In addition, online behavior can also include any implicit behavior that seems likely to infer an online product of interest, such as adjusting the mouse wheel, staying on a particular page for a period of time, or revisiting another online product belonging to the same or similar category. there is. At this time, the online behavior is not limited to the above examples. The interest data generating unit 102 may separate brand information, price range information, and the like corresponding to a product that is a target of a buyer's online behavior. Behavioral data may be acquired in real time as soon as the buyer accesses the online product marketing server 100 and acts. In addition, the behavioral data may be collected in the form of a stream, and may undergo a pre-processing process to be processed into data in a form necessary for identifying data of interest.

According to an embodiment, the interest data generation unit 102 may click on a product for a specific product, search for a review of the specific product, add the specific product to a shopping cart, input a search word for the specific product, or search for a review of the specific product. An interest matrix may be generated based on elements included in each of the behavioral data, such as clicks on advertisements, likes for the specific product, social activities such as sharing the specific product, and time spent on a specific page on which the specific product appears. . For example, the interest data generating unit 102 designates 1 if the review is searched for a specific product, 0 if not, and designates 1 if the shopping cart has been added to the shopping cart and 0 if not. 1 if the advertisement for the specific product is clicked, 0 if not clicked, 1 if the user likes the specific product, 0 if not, and sharing of the specific product If it is not 1, 0 may be designated, and the time itself may be designated as an interest matrix element according to the time (minutes) staying on the specific page. The interest data generator 102 may generate a purchase probability model according to a buyer's behavior pattern based on the interest matrix and whether or not the specific product is purchased.

According to an embodiment, the interest data generating unit 102 may calculate or determine the buyer's interest level for each online product based on the buyer's interest data. The interest data generating unit 102 determines the number of times the purchaser purchased the online product registered in the online product marketing server 100, the number of times the purchaser viewed an image of the online product, and the purchaser registered the online product. Based on the number of minutes the buyer stays on the page, the number of times the buyer searches for the online product name, and the number of times the buyer views reviews of the online product, the buyer's interest in each online product (sold by the seller) is determined. can be calculated or determined.

The interest data generating unit 102 may determine the buyer's product interest in the online product through Equation 1 below.

In Equation 1, T is the degree of interest in the product, M is the number of times the buyer purchases the online product registered in the server, n is the number of times the buyer views the image of the online product, k _- is the number of times the buyer has purchased the online product registered in the server A is the number of times the purchaser searches for the name of the online product, and b is the number of times the purchaser views reviews of the online product. w ₁ is the first weight for the number of times the image has been viewed, w ₂ is the second weight for the staying time, w ₃ is the third weight for the number of times the product name is searched, and w ₄ is the weight for the number of times the review has been viewed. It may mean a fourth weight. The sum of w ₁ to w ₄ is 1, and may be greater in order of w ₁ , w ₂ , w ₃ , and w ₄ .

The marketing management unit 103 may perform a customized advertisement to the buyer based on the buyer's behavioral data and/or interest data. In an embodiment, the marketing management unit 103 may determine an online product of the seller for which the buyer's interest in the product is equal to or higher than a preset first threshold level of interest, and may advertise or market another online product of the seller similar to the determined online product to the buyer. there is. For example, the marketing management unit 103 may recommend the other online product similar to the seller's online product to the buyer. In another embodiment, the marketing management unit 103 uses the online product marketing server 100, determines other buyers having similar behavioral data and/or similar interest data to the buyer, and the other buyers' interest in the product is set in advance. An online product of the seller having a degree of interest higher than or equal to a second threshold may be determined, and another online product of the seller similar to the determined online product may be advertised or marketed to the buyer. For example, the marketing management unit 103 may determine an online product of the seller for which the interest level of the other buyers is equal to or higher than a preset second threshold level of interest, and recommend another online product of the seller similar to the determined online product to the buyer. can

The marketing management unit 103 may generate sales data for another second online product of the seller determined to be a similar product corresponding to the seller's first online product. The marketing management unit 103 determines the advertising efficiency of an advertisement marketing a second online product based on the first online product, based on the product interest level of the buyer for the first online product and the product sales volume for the second online product. index can be calculated. In other words, the marketing management unit 103 may determine that the current advertisement efficiency is high when the product sales volume for the second online product is high regardless of the buyer's interest in the first online product, and the buyer's first online product If the interest in the product is high but the product sales volume for the second online product is low, it may be determined that the current advertising efficiency is low.

The similar product determining unit 104 may determine the seller's registered online product and the seller's other online product similar to the online product. The similar product determining unit 104 may determine similar products by determining similarities between product keywords of the seller's first online product and product keywords of other online products of the seller. The similar product determining unit 104 may determine similar products among the adjacent products based on image similarities between the product image of the first online product and the product image of the online product determined to be adjacent products among the other online products.

4 is a process of determining proximate products by comparing product keywords of a first online product with product keywords of other online products of a seller, and determining similar products by comparing product images of proximate products with product images of the first online product. is a drawing showing

The similar product determining unit 104 may compare product information of the first online product with product information of other online products of the seller to determine adjacent products. Specifically, the similar product determining unit 104 may compare a first product keyword of a first online product registered in the server 100 with a second product keyword of another online product of the seller registered in the server 100. .

When comparing the first product keyword of the first online product with the second product keyword of the other online products, the similar product determining unit 104 may exclude a specific keyword according to the purchaser's selection.

The similar product determining unit 104 may determine the keyword similarity through Equation 2 below.

Referring to Equation 2 above, S is the similarity of keywords, d is the number of keywords corresponding to the first online product, d' is the number of keywords corresponding to the second online product, and r is the number of keywords corresponding to the first online product and the number of keywords corresponding to the first online product. The number of identical keywords among keywords corresponding to the 2 online products, r' may be the number of similar keywords among keywords corresponding to the first online product and keywords corresponding to the second online product. h _i may be a correlation coefficient of the i th keyword with a product, and v _i may be a weight applied to the i th keyword. The correlation coefficient of a keyword with a product is set as the standard value if the keyword is related to the name, abbreviation, or designation of the product, and if the keyword is related to the nature or effect of the product, the correlation coefficient is set to a value smaller than the standard value. It can be. For example, if the product is cargo pants and the keyword is pants, the relevance coefficient is high, and if the keyword is a promotion-related keyword (e.g., elastic best, price discount, cost-effectiveness, etc.), the relevance coefficient is not related to pants. may be small.

The similar word may be a predetermined word. For example, words similar to the word pants may be pants, pants, bottom, and the like. Words similar to the word span could be span, spanky, span, and the like.

In Equation 2, the weight (v _i ) for the keyword may be set in advance to a value input from a manager for each product keyword. As another example, the weight (v _i ) for the keyword may be determined based on the purchaser's voice acquired from the purchaser terminal 200 and the obtained purchaser's voice. The buyer's voice may be a voice recorded when the buyer makes a phone call, a voice recorded when the buyer makes a clip image, or a buyer's daily voice.

Specifically, the similar product determining unit 104 selects keywords obtained by converting the buyer's voice into text (STT, Speech To Text) from among the first product keyword of the first online product or the second product keyword of the other online product. A preset reference weight may be applied to first keywords identical to or similar to , and a weight smaller than the reference weight may be applied to the remaining keywords.

Also, the similar product determining unit 104 applies a first weight greater than a reference weight to second keywords corresponding to a positive expression among the first keywords, and applies a third weight corresponding to a negative expression among the first keywords. A second weight smaller than the reference weight may be applied to the keywords.

In addition, the similar product determining unit 104 applies a third weight higher than the first weight to a keyword that is the same as or similar to a keyword that is mentioned in the buyer's voice more frequently than the first frequency among the second keywords, and applies a third weight higher than the first weight to the third keyword. Among them, a fourth weight lower than the second weight may be applied to a keyword that is identical to or similar to a keyword that is mentioned in the buyer's voice more frequently than the first frequency.

The similar product determining unit 104 may determine the other online product having a keyword similarity (s) with the first online product higher than a first threshold value as an adjacent product. The similar product determining unit 104 may perform the operation of performing the keyword comparison as described above with respect to all other online products of the seller registered in the server 100 and determine a plurality of adjacent products.

The similar product determining unit 104 compares product images of adjacent products with product images of the first online product to determine image similarity, and determines adjacent products having a similarity equal to or greater than a specific threshold value as similar products. For example, the similar product determining unit 104 may compare a first product image of a first online product with a second product image of a nearby product. The similar product determining unit 104 may determine the adjacent product as a similar product when the degree of similarity between the first product image and the second product image is greater than or equal to a second threshold based on the comparison.

The similar product determining unit 104 may detect feature points of the first product image and feature points of the second product image by applying a pretrained deep neural network model. The pre-trained deep neural network model is a learning model trained to determine a feature point using a product image as an input, and may be a convolution neural network (CNN)-based learning model. The feature points may include pixel values constituting the outermost vertices of the product included in the product image, pixel values constituting the seam line of the product, and pixel values constituting the overall color of the product.

The similar product determining unit 104 may determine a feature vector based on the detected feature points. For example, the similar product determining unit 104 may determine a first feature vector of a first product image and a second feature vector of a second product image. The similar product determining unit 104 may calculate a similarity between the first feature vector and the second feature vector by using the cosine similarity. The similar product determining unit 104 may determine whether a similarity between the first product image and the second product image is greater than or equal to a second threshold value based on the similarity between the vectors.

The similar product determining unit 104 may perform the image comparison operation as described above with respect to all nearby products of the seller registered in the server 100 and determine a plurality of similar products.

When comparing the first product image and the second product image, the similar product determining unit 104 compares the product front image included in the first product image with the product front image included in the second product image, and compares the product front image included in the first product image. The product side image included in the product image may be compared with the product side image included in the second product image, and the product rear image included in the first product image may be compared with the product rear image included in the second product image.

The similar product determining unit 104 may generate a 3D image by combining front, back, and side images included in the product image. The similar product determining unit 104 generates a first three-dimensional image based on the front, side, and rear product images included in the first product image, and based on the front, side, and rear product images included in the second product image. A second stereoscopic image may be generated. Here, the similar product determining unit 104 generates the first stereoscopic image and the second stereoscopic image, the buyer's body numerical information obtained from the purchaser terminal 200 (e.g., height, weight, body shape, arm thickness). , thigh thickness, etc.) or a body image including the body of the purchaser, a sense of volume when actually worn by the purchaser is calculated, and a three-dimensional image may be generated based on the sense of volume.

The similar product determining unit 104 may determine whether the similarity between the first product image and the second product image is greater than or equal to a second threshold by comparing not only the flat images but also the first stereoscopic image and the second stereoscopic image. there is.

FIG. 5 is a diagram showing the hardware configuration of the online product marketing server 100 according to FIG. 1 .

Referring to FIG. 5 , the online product marketing server 100 stores at least one processor 110 and instructions instructing the at least one processor 110 to perform at least one operation. It may include a memory (memory) to.

The at least one operation may include at least some of the above-described operations or functions of the online product marketing server 100 and may be implemented in the form of instructions and performed by the processor 110 .

Here, the at least one processor 110 may mean a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor for performing methods according to embodiments of the present invention. can Each of the memory 120 and the storage device 160 may include at least one of a volatile storage medium and a non-volatile storage medium. For example, the memory 120 may be one of a read only memory (ROM) and a random access memory (RAM), and the storage device 160 may be a flash-memory. , hard disk drive (HDD), solid state drive (SSD), or various memory cards (eg,

In addition, the online product marketing server 100 may include a transceiver 130 that performs communication through a wireless network. In addition, the online product marketing server 100 may further include an input interface device 140 , an output interface device 150 , a storage device 160 , and the like. Each component included in the online product marketing server 100 may be connected by a bus 170 to communicate with each other. In FIG. 5, the online product marketing server 100 has been described as an example, but is not limited thereto. For example, a plurality of purchaser terminals may include a component according to FIG. 5 .

The methods according to the present invention may be implemented in the form of program instructions that can be executed by various computer means and recorded on a computer readable medium. Computer readable media may include program instructions, data files, data structures, etc. alone or in combination. Program instructions recorded on a computer readable medium may be specially designed and configured for the present invention or may be known and usable to those skilled in computer software.

Examples of computer readable media may include hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions may include not only machine language codes generated by a compiler but also high-level language codes that can be executed by a computer using an interpreter and the like. The hardware device described above may be configured to operate with at least one software module to perform the operations of the present invention, and vice versa.

In addition, the above-described method or device may be implemented by combining all or some of its components or functions, or may be implemented separately.

Although the above has been described with reference to preferred embodiments of the present invention, those skilled in the art will variously modify and change the present invention within the scope not departing from the spirit and scope of the present invention described in the claims below. You will understand that it can be done.

100: online product marketing server 200: buyer terminal
300: seller terminal

Claims (5)

An online product marketing server for marketing online products registered from a seller terminal to a buyer,
a product information collection unit acquiring product information of online products from the seller terminal;
an interest data generator configured to obtain behavioral data of the purchaser with respect to the online product and generate interest data of the purchaser with respect to the online product based on the acquired behavioral data;
A marketing management unit providing a customized marketing service for the buyer based on the interest data;
The interest data generating unit,
The number of times the buyer purchased the first online product registered in the server, the number of times the buyer viewed the image of the first online product, the time the buyer stayed on the page where the first online product was registered, the buyer Calculate product interest in the seller's first online product based on at least one of the number of times the name of the first online product is searched and the number of times the buyer has viewed a review of the first online product;
The buyer's product interest in the online product is calculated through the following equation,

In the above equation, T is the degree of interest in the product, M is the number of times the buyer purchases the online product registered in the server, n is the number of times the buyer views the image of the online product, and k _- is the number of times the buyer views the online product. The time (minutes) that the online product stays on the registered page, a is the number of times the buyer searches for the online product name, b is the number of times the buyer views reviews of the online product, and w ₁ is the number of times the image is viewed A first weight, w ₂ is a second weight for the staying time, w ₃ is a third weight for the number of times the product name is searched, w ₄ is a fourth weight for the number of times the review is viewed, and w The sum of ₁ to w ₄ is 1, and is greater in order of w ₁ , w ₂ , w ₃ , and w ₄ ,
The marketing management department,
determining the first online product of the seller for which the buyer's interest in the product is equal to or greater than a preset first threshold interest, and similar to the determined first online product among other online products of the seller similar to the determined first online product; recommend second online products to the purchaser;
The server further includes a similar product determination unit configured to determine a similar product similar to the first online product among the other online products of the seller;
The similar product determining unit,
determining proximity products by determining a keyword similarity between product keywords of the first online product of the seller and product keywords of the second online product of the seller;
The keyword similarity is determined through the following equation,

In the above equation, S is the similarity of keywords, d is the number of keywords corresponding to the first online product, d' is the number of keywords corresponding to the second online product, and r is the number of keywords corresponding to the first online product and the second online product. The number of identical keywords among the corresponding keywords, r' is the number of similar keywords among the keywords corresponding to the first online product and the keywords corresponding to the second online product, h _i is the correlation coefficient of the ith keyword with the product, v _i is A weight applied to the i-th keyword,
The relevance coefficient (h _i ) is set as a reference value in the case of a keyword related to a product name, abbreviation, or title, and is set to a value smaller than the reference value in the case of a keyword related to the nature or effect of a product. is set,
The weight (v _i ) for the keyword is set in advance to a value input from the manager for each product keyword or is determined based on the buyer's voice obtained by acquiring the buyer's voice from the buyer's terminal, and the buyer's voice is Includes the voice recorded when the buyer makes a call, the voice recorded when the buyer makes a clip video, and the buyer's daily voice,
determining the adjacent product as the similar product based on image similarities between a product image of the second online product and a product image of the first online product determined as the adjacent product among the other online products;
The marketing management department,
generating sales data for another second online product of the seller determined to be the similar product corresponding to the first online product of the seller;
calculating an advertising efficiency index for an advertisement marketing the second online product to the buyer based on the purchaser's degree of interest in the first online product and the product sales volume for the second online product; server.

delete delete delete delete

Images