KR20210085770A - Apparatus for recommending product based log - Google Patents

Abstract

According to the present invention, a log-based product recommendation device includes: a user terminal for transmitting a log including event information of product inquiry and product purchase; and a recommendation server configured to: collect a log for a preset sampling period; refine the log based on preset cleaning conditions; generate a product log for product purchase and product inquiry from the refined log; extract second product search information from the log received along with a product recommendation request signal, when the log is received along with the recommendation request signal; calculate a similarity between the second product search information and the product log; and generate recommended product information based on the similarity. Accordingly, by reflecting a product purchase propensity according to the product search pattern to recommend the product, it is possible to promote user's purchase of the product.

Description

Log-based product recommendation device {Apparatus for recommending product based log}

The present invention relates to a log-based product recommendation apparatus, and more particularly, to a log-based product recommendation apparatus capable of recommending a product based on a user's log.

With the development of mobile communication networks and the development of terminal specifications, mobile communication terminals have become essential belongings of modern people, out of the scope of conventional simple communication devices or information providing devices, and are evolving into total entertainment devices.

In general, an e-commerce service using a mobile communication terminal or a PC is performed by providing product information sold on an online shopping mall site to customers, and purchasing a desired product after the customer confirms the product information. However, customers have difficulty in searching for their preferred products on large shopping mall sites where there is a vast amount of product information.

Difficulty in product search is a factor that lowers customer satisfaction and willingness to revisit. Accordingly, the e-commerce service provides a service for recommending products that they prefer to customers.

However, the conventional product recommendation method uses collaborative filtering to recommend a product preferred by the customer based on information of customers who have a similar preference to the customer’s preference. Calculation complexity increases, and recommendations for new users are impossible.

In addition, it classifies customers using e-commerce services into groups according to age and gender, measures the product similarity of customers with similar preferences within the classified group, and recommends products to target customers according to the measured product similarity. Although the method is provided, this is only done based on the customer's preference, and there is a problem in that the accuracy and reliability of the recommended product are lowered.

Therefore, there is a need for a method for providing users with higher quality recommended products.

Korean Patent Registration No. 10-1567217

An object of the present invention is to provide a log-based product recommendation apparatus capable of recommending a product by calculating a similarity between product search information in an existing log and product search information in a log received along with a product recommendation signal.

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention not mentioned may be understood by the following description, and will be more clearly understood by the embodiments of the present invention. Moreover, it will be readily apparent that the objects and advantages of the present invention may be realized by the means and combinations thereof indicated in the claims.

A log-based product recommendation apparatus according to the present invention for solving the above technical problem includes: a user terminal for transmitting a log including event information of product inquiry and product purchase; The log is collected for a preset sampling period, the log is refined based on a preset refining condition, a product log for product purchase and product inquiry is generated from the refined log, and a product recommendation request signal from the user terminal When a log is received with the product recommendation request signal, second product search information is extracted from the log received along with the product recommendation request signal, a degree of similarity between the second product search information and the product log is calculated, and recommended product information is obtained based on the similarity degree. It may include a recommendation server to generate.

Preferably, the recommendation server sets reference number information by using the product inquiry number information included in each log, and compares the set number of times information with the product inquiry number information included in any one of the logs. It may be determined whether any one log satisfies a preset refining condition.

Preferably, the recommendation server calculates the average number of inquiry information among the product inquiry number information included in each of the logs, amplifies the average inquiry number information and sets it as the reference number information, and stores the information in any one of the logs. When the included product inquiry count information exceeds the reference count information, it is determined that any one of the logs satisfies a preset refining condition, and the one of the logs may be deleted.

Preferably, the recommendation server extracts product purchase information indicating a purchased product and first product search information indicating a product searched until purchasing the purchased product from the refined log, and extracts the extracted product purchase information and the The first product search information may be mapped to generate the product log.

Preferably, the recommendation server calculates a similarity between the first product search information and the second product search information based on products indicated by each of the first product search information and the second product search information included in the product log. can do.

Preferably, the recommendation server calculates the number of duplicates of the product purchase information mapped with the first product search information including all products indicated by the second product search information for each product indicated by the product purchase information, and the calculated duplicates for each product The number can be calculated as similarity.

Preferably, the recommendation server may generate, as the recommended product information, the product purchase information corresponding to the preset number of recommended products in the order of increasing the similarity.

According to the present invention, by reflecting the product purchase propensity according to the product search pattern and recommending the product, it is possible to promote the user's product purchase.

1 is a diagram showing the configuration of a log-based product recommendation system according to an embodiment of the present invention.
2 is a block diagram showing the configuration of a user terminal according to the present invention.
3 is a block diagram showing the configuration of a recommendation server according to the present invention.
4 is a diagram for explaining a process in which the recommendation server generates recommended product information according to the present invention.

Hereinafter, various embodiments of the present invention will be described with reference to the accompanying drawings. However, this is not intended to limit the present invention to specific embodiments, and it should be understood that various modifications, equivalents, and/or alternatives of the embodiments of the present invention are included. In connection with the description of the drawings, like reference numerals may be used for like elements.

In this document, expressions such as "have", "may have", "includes", or "may include" refer to the presence of a corresponding characteristic (eg, a numerical value, function, operation, or component such as a part). and does not exclude the presence of additional features.

In this document, expressions such as “A or B”, “at least one of A or/and B”, or “one or more of A or/and B” may include all possible combinations of the items listed together. . For example, "A or B", "at least one of A and B", or "at least one of A or B" means (1) includes at least one A, (2) includes at least one B; Or (3) it may refer to all cases including both at least one A and at least one B.

Expressions such as "first", "second", "first", or "second" used in this document may modify various elements, regardless of order and/or importance, and may modify one element. It is used only to distinguish it from other components, and does not limit the components. For example, the first user equipment and the second user equipment may represent different user equipment regardless of order or importance. For example, without departing from the scope of the rights described in this document, a first component may be referred to as a second component, and similarly, the second component may also be renamed as a first component.

A component (eg, a first component) is "coupled with/to (operatively or communicatively)" to another component (eg, a second component) When referring to "connected to", it will be understood that the certain element may be directly connected to the other element or may be connected through another element (eg, a third element). On the other hand, when it is said that a component (eg, a first component) is "directly connected" or "directly connected" to another component (eg, a second component), the component and the It may be understood that other components (eg, a third component) do not exist between other components.

The expression “configured to (or configured to)” as used in this document, depending on the context, for example, “suitable for”, “having the capacity to )", "designed to", "adapted to", "made to", or "capable of" . The term “configured (or set up to)” may not necessarily mean only “specifically designed to” hardware. Instead, in some contexts, the expression “a device configured to” may mean that the device is “capable of” with other devices or parts. For example, the phrase "controller configured (or configured to perform) A, B, and C" may refer to a dedicated processor (eg, an embedded processor) for performing the corresponding operation, or by executing one or more software programs stored in memory. , may mean a generic-purpose processor (eg, a CPU or an application processor) capable of performing corresponding operations.

As used herein, the term “unit” refers to a hardware component such as software, FPGA, or ASIC, and “unit” performs certain roles. However, “a” is not meant to be limited to software or hardware. "~" may be configured to be in an addressable storage medium. Thus, as an example, “a” refers to components such as software components, object-oriented software components, class components, and task components, and processors, functions, properties, and procedures. , subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables. The functions provided within the components and “units” may be combined into a smaller number of components and “units” or further separated into additional components and “units”.

In particular, in the present specification, the “control unit” may include one or more of a central processing unit (CPU), an application processor (AP), and a communication processor (CP).

In this specification, a computer means all types of hardware devices including at least one processor, and may be understood as encompassing software configurations operating in the corresponding hardware device according to embodiments. For example, a computer may be understood to include, but is not limited to, smart phones, tablet PCs, desktops, notebooks, and user clients and applications running on each device.

Terms used in this document are only used to describe specific embodiments, and may not be intended to limit the scope of other embodiments. The singular expression may include the plural expression unless the context clearly dictates otherwise. Terms used herein, including technical or scientific terms, may have the same meanings as commonly understood by one of ordinary skill in the art described in this document. Among the terms used in this document, terms defined in general dictionary may be interpreted with the same or similar meaning to the meaning in the context of the related art, and unless explicitly defined in this document, have an ideal or excessively formal meaning. not interpreted In some cases, even terms defined in this document cannot be construed to exclude embodiments of this document.

1 is a diagram showing the configuration of a log-based product recommendation system according to an embodiment of the present invention.

Referring to FIG. 1 , a log-based product recommendation system 10 according to the present invention may include a user terminal 100 and a recommendation server 200 .

The user terminal 100 may be connected to the recommendation server 200 through a communication network to transmit/receive a log including all data for providing a recommended product based on the log. In particular, the user terminal 100 according to the present invention may receive information about a recommended product from the recommendation server 200 . Then, the user terminal 100 detects an input signal for a selection event such as a product inquiry related to a product, adding to a shopping cart, or purchasing a product, and according to the input signal, a log including information on the corresponding product is stored in the recommendation server 200 ) can be transmitted. Meanwhile, the user terminal 100 may be a device used by any one user, but may be a different device used by a plurality of different users. That is, there may be a plurality of user terminals 100 .

The recommendation server 200 may be connected to the user terminal 100 through a communication network to transmit/receive a log including all data for providing a recommended product based on the log. In particular, the recommendation server 200 according to the present invention extracts data related to the occurrence of a selection event for a plurality of products from the user terminal 100 from the log, and extracts the extracted data from the log in order to provide a recommended product related to a specific product. Data can be purified. Here, the data includes information related to selection or purchase of a specific product, and some data may be erased by preset refining conditions. In addition, the event may be a user action such as inquiring a product for a specific product, adding it to a shopping cart, or making a purchase. The recommendation server 200 may collect the log by receiving the log from the user terminal 100 during a preset sampling period. For example, the sampling period may be one year.

The recommendation server 200 analyzes the refined data to generate product recommendation reference data, and generates recommended product information based on the log received along with the product recommendation signal from the user terminal 100 and the generated product recommendation reference data. can

Meanwhile, the recommendation server 200 may calculate an association score between related products for each user by using the sorted data. That is, the recommendation server 200 uses the sorted data for each user, a first score related to previous Uniform Resource Locator (URL) information of a specific product, a second score related to a time difference between occurrence of a selection event between products, A third score determined according to the type of event related to the product may be calculated to calculate a related score. In addition, the recommendation server 200 may collect the related scores by classifying them for the same product. Thereafter, the recommendation server 200 may calculate a related score between the collected products and the number of users using the product to determine a final score, and apply the final score to provide a recommendation function for other products.

Through this, the present invention calculates a correlation score between products related to each user by using data related to product purchase, divides the correlation score according to the same product pair, and collects the correlation score between the collected product pairs and the corresponding product pair. Since the final score determined by calculating the number of users using the product is applied to provide a recommendation function for other products, reliable and accurate product recommendation is possible. In addition, since a product recommendation is provided based on the user's log records, the weight of the meaningless log of users can be lowered and the quality of the recommendation service can be improved.

2 is a block diagram showing the configuration of a user terminal according to the present invention.

Referring to FIG. 2 , the user terminal 100 according to the present invention may include a control unit 110 , an input unit 120 , a display unit 130 , a storage unit 140 , and a communication unit 150 .

The input unit 120 may receive various information such as number and character information, and transmit signals input in connection with setting various functions and controlling functions of the user terminal 100 to the control unit 110 . In addition, the input unit 120 may be configured to include at least one of a keypad and a touchpad that generate an input signal according to a user's touch or manipulation. In this case, the input unit 120 may be configured in the form of a single touch panel (or touch screen) together with the display unit 130 to simultaneously perform input and display functions. In addition, the input unit 120 includes a keyboard, All types of input means that can be developed in the future may be used in addition to input devices such as a keypad, mouse, joystick, etc. In particular, the input unit 120 according to the present invention detects an input signal for inquiring or purchasing a product, and the control unit ( 110) can be forwarded.

The display unit 130 may display information about a series of operation states and operation results that occur while the function of the user terminal 100 is performed. Also, the display unit 130 may display a menu of the user terminal 100 and user data input by the user. In this case, when the display unit 130 is configured in the form of a touch screen, the display unit 130 may perform some or all of the functions of the input unit 120 . In particular, the display unit 130 according to the present invention may output product information related to product inquiry or purchase. Also, the display unit 130 may display information on a recommended product provided from the recommendation server 200 .

The storage unit 140 is a device for storing data, includes a main storage device and an auxiliary storage device, and may store an application program required for a functional operation of the user terminal 100 . The storage unit 140 may largely include a program area and a data area. Here, when activating each function in response to a user's request, the user terminal 100 may provide each function by executing corresponding application programs under the control of the controller 110 . In particular, the storage unit 140 according to the present invention may store an operating system for booting the user terminal 100 , a program for providing an e-commerce service for product selection or purchase, and the like. Also, the storage unit 140 may store information on recommended products and logs related to product selection or purchase.

The communication unit 150 may perform a function for transmitting and receiving data through a communication network with the recommendation server 200 . The communication unit 150 may include at least one of a wireless communication module (not shown) and a wired communication module (not shown). In addition, the wireless communication module is a configuration for transmitting and receiving data according to a wireless communication method, and when the user terminal 100 uses wireless communication, any one of a wireless network communication module, a wireless LAN communication module, and a wireless fan communication module is used. Thus, data may be transmitted/received to and from the recommendation server 200 . In addition, the wired communication module is for transmitting and receiving data by wire. The wired communication module may connect to a communication network through a wire and transmit/receive data to and from the recommendation server 200 . In particular, the communication unit 150 according to the present invention may communicate with the recommendation server 200 to transmit/receive a log including data for providing a recommended product based on the log.

The control unit 110 may be a process device for driving an operating system (OS) and each component. In particular, the controller 110 according to an embodiment of the present invention may receive information on a recommended product related to a specific product from the recommendation server 200 . Then, the control unit 110 detects an input signal for at least one selection event among product inquiry, shopping cart addition, or purchase related to the product, and according to the input signal, a log including information on the corresponding product is sent to the recommendation server ( 200) can be transmitted.

3 is a block diagram showing the configuration of a recommendation server according to the present invention, and FIG. 4 is a diagram for explaining a process in which the recommendation server according to the present invention generates recommended product information.

3 and 4 , the recommendation server 200 according to an embodiment of the present invention may include a control unit 210 , a storage unit 220 , and a communication unit 230 .

The communication unit 230 may communicate with the user terminal 100 through a communication network to transmit/receive a log including data for providing a recommended product based on the log.

The storage unit 220 may store all programs according to the function execution of the recommendation server 200 . In particular, the storage unit 220 according to the present invention includes a program for collecting a log including data related to product selection or purchase for each user, a program for sorting or refining the collected data, and a correlation score between products based on the sorted data. It is possible to store a program that calculates , a program that collects related scores for the same product for each user, a program that calculates the collected related scores to determine the final score, and a program that provides recommended products by applying the final score. . In addition, the storage 220 may store the related score, the final score, and a log related to product selection or purchase.

The control unit 210 may be a process device that drives an operating system and each component. In particular, the control unit 210 according to an embodiment of the present invention may collect data related to product purchase from the user terminal 100 and sort the collected data in order to provide a recommended product related to a specific product. Here, in sorting the collected data, the controller 210 may select a product or sort the product in the order of time and date at which the product purchase is requested. On the other hand, the controller 210 considers data of a user who has performed a selection event of less than a certain number of products from among the collected data and data of a product that has received a selection event from a user less than a certain number of products on a product basis as meaningless data. can be deleted. In this case, the selection event may be a user action such as inquiring a product for a specific product, adding it to a shopping cart, or making a purchase.

The controller 210 may delete a log that satisfies a preset refining condition from among the plurality of logs. Specifically, the control unit 210 checks the product inquiry number information included in all stored logs, calculates the average inquiry number information of the checked product inquiry number information, and the product inquiry number information included in any log among a plurality of logs is When the reference number information set based on the average number of inquiry information is exceeded, it is determined that the log satisfies the preset refining condition, and the log may be deleted.

To this end, the controller 210 may set the reference number information by amplifying the average number of inquiry information by 150%.

Through this, the controller 210 may delete an abnormal log in generating product recommendation information.

The controller 210 may extract first product search information and product purchase information from the refined log, and may generate a product log by mapping the extracted first product search information and product purchase information.

For example, as shown in FIG. 4 , the controller 210 controls product purchase information indicating product A purchased from the user from any one of the refined logs and products B, C, and D searched until purchasing product A. The indicated first product search information may be mapped. The controller 210 may extract first product search information and product purchase information from all the refined logs, and map the extracted first product search information and product purchase information.

Thereafter, the controller 210 may extract the second product search information from the log received together with the product recommendation request signal. The controller 210 may calculate a similarity between the product log and the second product information. Specifically, the controller may calculate a similarity indicating a similarity between the first product search information and the second product search information included in all product logs, and generate recommended product information based on the similarity.

Here, the second product search information may be information indicating a product searched by a user who is a target of product recommendation to purchase a product through the user terminal 100 .

In addition, the first product search information may be information indicating a product searched by a user who is a target of product recommendation or a user other than a user who is a target of product recommendation until purchasing a product of product purchase information in the past.

Specifically, the control unit 210 calculates the duplicate number of product purchase information mapped with the first product search information including all products of the second product search information for each product of the product purchase information, and calculates the number of duplicates for each product as a degree of similarity can be calculated as

For example, when the products indicated by the second product search information are C and D, and the products indicated by each of the first product search information recommended from the refined logs and the purchased product information are the same as in Table 1 below, the controller 210 Since the products indicated by the first product search information included in the first log, the third log, and the fourth log include all the products indicated by the second search information, duplicate product A indicated by the purchased product information in the first log. The number may be calculated as “1”, and the duplicate number of product B indicated by the purchased product information of each of the third log and the fourth log may be calculated as “2”.

Log 1st product search information Purchase product information first log B, C, D A 2nd log B, C A 3rd log B, C, D B 4th log C, D, E B

In this case, the controller 210 may calculate the similarity by using the apriori algorithm.

The controller 210 sorts the product purchase information to which the first product search information and the first product search information are mapped in the order of the calculated similarity, and recommends product purchase information corresponding to the preset number of recommended products in the order of the highest similarity. It can be created with product information.

Meanwhile, the control unit 210 according to another embodiment determines whether interest product information is included in the first product search information including all products of the second product search information, and selects all products of the second product search information. When the product of interest information is included in the included first product search information, the number of duplicates may be corrected and calculated based on the number of product information of interest.

Here, the interest product information may be information indicating a product that the user sets a favorite among the products searched for by the user or is separately classified using a shopping cart function.

Referring back to Table 1 above, when product E among the products indicated by the first product search information of the fourth log is the product of interest, the controller 210 includes the product of interest information in the first product search information of the third log. It is determined that there is one, and the number of duplicates may be calculated by performing correction by adding 1 of product information of interest to 1 of duplicates of product B according to the third log.

Meanwhile, the controller 210 according to another embodiment may calculate a correlation score between related products for each user by using the sorted data. At this time, the control unit 210 controls a first score (ref_score) related to previous UR information of a specific product, a second score (time_diff_score) related to a time difference between occurrence of a selection event between products, and a third score determined according to the type of event related to the product (action_score) can be calculated to calculate a log score for the associated score.

Here, in order to calculate the first score, the controller 210 checks previous URL information for a specific product, which includes the main page of the sales site, the sub category page, the search page, the external search engine, It may include a blog site, a news site, various community sites, a social network service site, and the like. However, the previous URL information according to an embodiment of the present invention is not limited to the URL information presented above, and may include various types of URL information for searching or checking a specific product. That is, the control unit 210 identifies the user's intention by using a plurality of previous URL information that can confirm a specific product, and if the user clicks on the product without much meaning, the first score is obtained by applying a low weight. can be calculated For example, the control unit 210 may set a low weight for user actions in this case, since the user may click on a product meaninglessly or without thinking about it or if the previous URL information is the main page of the sales site. have. On the other hand, if the previous UR information is a sub-category page or a search page of the sales site, the user clicks and confirms the product with a specific intention. In this case, the weight for the user action may be set high. .

Also, the controller 210 may calculate a second score by calculating a time difference between the occurrence time of the selection event for the specific product and the occurrence time of the selection event for the next product after the occurrence of the selection event for the specific product. That is, the controller 210 may sequentially check the selection event occurrence time difference for a plurality of products. In addition, the controller 210 determines that the correlation between products is low when the confirmed time difference is large, and can calculate the second score by giving a weight lower than that of the products when the time difference is small.

Also, the controller 210 may calculate a third score related to the event type for the product. That is, when a purchase event occurs among product-related events, the control unit 210 may calculate a weight for the third score to be higher than that of other events (eg, viewing product or adding to shopping cart).

Thereafter, the controller 210 may calculate a pair score for a plurality of products by calculating a log score calculated for each product. In addition, the controller 210 may calculate a normalized score by normalizing the pair score between products, and delete the overlapping normalized score between the products.

The control unit 210 may classify and collect the related scores for the same product. Then, the control unit 210 may determine the final score by calculating the number of users using the collected related scores and the product. In this case, the controller 210 may calculate the final score between the products by dividing the sum of the associated scores for each user by the number of users who have selected or purchased each product. Then, the controller 210 may provide a recommendation function for other products by applying the final score.

In particular, the control unit 210 according to the present invention may function to sort the data related to the purchase of goods collected from the user terminal 100 in the order of time and date. Also, the controller 210 may function to calculate a correlation score between products related to each user by using the sorted data. Also, the control unit 210 may function to classify and collect the calculated related scores for the same product. In addition, the control unit 210 may function to determine the final score by calculating the collected related scores and the number of users using the corresponding product. In addition, the controller 210 may perform a recommendation function for other products by applying the determined final score.

So far, the present invention has been focused on preferred embodiments. Those of ordinary skill in the art to which the present invention pertains will understand that the present invention can be implemented in modified forms without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive point of view. The scope of the present invention is indicated in the claims rather than the foregoing description, and all differences within the scope equivalent thereto should be construed as being included in the present invention.

As described above, although the present invention has been described with reference to limited embodiments and drawings, the present invention is not limited thereto, and the technical idea of the present invention and the following by those of ordinary skill in the art to which the present invention pertains. Of course, various modifications and variations are possible within the scope of equivalents of the claims to be described.

10: Log-based product recommendation device
100: user terminal
200: recommendation server

Claims (7)

a user terminal for transmitting a log including event information of product inquiry and product purchase; and
The log is collected for a preset sampling period, the log is refined based on a preset refining condition, a product log for product purchase and product inquiry is generated from the refined log, and a product recommendation request signal from the user terminal When the log is received with the product recommendation request signal, the second product search information is extracted from the log received along with the product recommendation request signal, the similarity between the second product search information and the product log is calculated, and recommended product information is obtained based on the similarity level. A log-based product recommendation system comprising a; a recommendation server for generating.
According to claim 1,
The recommendation server is
The reference number information is set using the product inquiry number information included in each log, and the preset refining condition is determined by comparing the product inquiry number information included in any one of the logs with the set reference number information. A log-based product recommendation system, characterized in that it is determined whether one log is satisfied.
3. The method of claim 2,
The recommendation server is
Calculate the average number of inquiry information among the product inquiry number information included in each log, amplify the average inquiry number information and set it as the reference number information, and the product inquiry number information included in any one of the logs When the reference number information is exceeded, it is determined that the one log satisfies a preset refining condition, and the one log is deleted.
According to claim 1,
The recommendation server is
From the refined log, product purchase information indicating a purchased product and first product search information indicating a product searched until purchasing the purchased product are extracted, and the extracted product purchase information and the first product search information are mapped A log-based product recommendation system, characterized in that it is generated as the product log.
5. The method of claim 4,
The recommendation server is
Log-based, characterized in that the similarity between the first product search information and the second product search information is calculated based on the products indicated by each of the first product search information and the second product search information included in the product log Product recommendation system.
6. The method of claim 5,
The recommendation server is
Calculating the duplicate number of the product purchase information mapped with the first product search information including all products indicated by the second product search information for each product indicated by the product purchase information, and calculating the calculated number of duplicates for each product as a degree of similarity A log-based product recommendation system that features.
6. The method of claim 5,
The recommendation server is
The log-based product recommendation system, characterized in that the product purchase information corresponding to the preset number of recommended products in the order of the greatest similarity is generated as the recommended product information.

Images