TWM627312U - System to wake up non-shopping consumer - Google Patents

Abstract

This creation discloses a system for awakening non-shopping consumers, which uses artificial intelligence to group consumers into groups, and then determines consumers who have not shopped for a long time, and provides products that meet consumers' needs by analyzing the results of the grouping, as consumers' shopping choices Reference, and then arouse consumers' desire to buy goods, so as to awaken consumers who have not shopped for a long time.

Description

A system to wake up non-shopping consumers

This creation involves a system for awakening non-shopping consumers, especially a system that uses artificial intelligence to divide consumers into groups, and then uses the grouping results to awaken long-term non-shopping consumers.

The shopping journey of consumers can be of great help to merchants in marketing. From picking commodities to purchasing commodities, every step contains potential business opportunities. For example, CN106485536A discloses a method and system for determining the time interval for the next purchase. By collecting customer consumption records, analyzing individual purchasing behaviors and crowd purchasing behaviors, the aforementioned two behaviors are used as variables to determine the time interval of individual purchasing behaviors, and when the purchasing activity is in a sleeper state, the aforementioned customer consumption records are used. Analyze and recommend the most appropriate product and recommend it to customers.

However, CN106485536A only calculates the variables of individuals and groups in a single dimension based on the purchase behavior of consumers in a single location, such as time, location, and commodities, infers the cycle of consumers purchasing commodities at each location, and then recommends appropriate commodities. For consumers, therefore, it lacks a multi-dimensional consideration, to study the behavior of consumers, whether it is an individual or a group, in order to recommend the products that consumers need.

In addition, there are other prior art available for reference as follows: (1) CN107767217A "Shopping recommendation method, mobile terminal and storage medium"; (2) CN110751515A "Decision-making method and device based on user consumption behavior"; (3) JPA2019046189 "Extraction device, extraction method and extraction process"; (4) JPA2020047157 "Commodity Recommendation Device, Commodity Recommendation Systema, and びプログラム".

Based on this, how to provide products that meet the needs of consumers based on multi-dimensional considerations and the attributes of products, or increase the accuracy of placing and promoting products, and then arouse consumers' desire to buy products. It's a problem to be solved.

In view of the above problems, the creator is based on years of experience in related industries to improve the system for awakening non-shopping consumers; therefore, the main purpose of this creation is to provide a system for awakening non-shopping consumers. It is mainly based on the behavior of consumers and the attributes of the products, providing products that meet the needs of consumers, and increasing the accuracy of placing and promoting products, thereby arousing consumers' desire to buy products, so as to wake up long-term non-shopping consumption. By.

In order to achieve the above-mentioned purpose, the present invention mainly uses a data processing unit to classify a route data label generated by a user operating an information device based on a plurality of classification labels of a label database, and store it in a route database, After training and learning, an artificial intelligence module converts the path data into a vectorized data, and then classifies the multiple vectorized data into a group data, wherein the path data can be a website trigger event, a website click event, a website Operation behavior, time spent on a website, or a derivative data under the aforesaid website operation behavior or any combination of data.

Secondly, the data processing unit determines whether the user corresponding to the path data is the wake-up target according to the path data containing multiple classification labels in the path database; furthermore, the artificial intelligence module compares the group data with a product database based on the wake-up target At least one product data is matched to generate a matching data; finally, based on the matching data, the data processing unit extracts a relevant product data related to the matching data in the product database, and transmits it to the information device to provide more users Products that may be purchased can be used as a reference for users' shopping choices, or provide users with reference matching data to promote more products, and then accurately place products that consumers are interested in, so as to wake up consumers who have not shopped for a long time.

In order for your reviewers to have a clear understanding of the purpose, technical features and effects of this creation, the following descriptions and illustrations are used for illustration, please refer to.

Please refer to "Fig. 1". Fig. 1 is a system architecture diagram of the creation. As shown in the figure, the waking up non-shopping consumer system 1 of the present creation is connected with an information device 2, which mainly includes a data processing unit 10. It is respectively connected with a tag database 21, a route database 22, a product database 23, and an artificial intelligence module 30; and the information device 2 can be a mobile phone, a tablet computer, a personal computer, etc. One of them, but not limited to this.

The data processing unit 10 can be used to drive the above modules and databases, and to classify an input data label generated by a user operating the information device 2, such as route data, product data, and has the ability to receive and transmit information signals, It has functions such as logic operation, temporary storage of operation results, and storage of execution instruction positions, and it can be a central processing unit (CPU) or a microcontroller (MCU).

The label database 21, the path database 22, and the product database 23 can be used to store electronic data, which can be a solid state disk (Solid State Disk or Solid State Drive, SSD), a hard disk (Hard Disk Drive, HDD), a static memory (Static Random Access Memory, SRAM), a random access memory (Random Access Memory, DRAM), or a cloud drive (Cloud Drive) any one or a combination thereof.

The tag database 21 mainly stores a plurality of classification tags for the data processing unit 10 to classify the input data tags; the path database 22 mainly stores a path vector learning data, a vector grouping learning data, a history data, and a path data, The above data can be pre-input data from an external database; historical data can be the data calculated and processed by the system itself. After the system has processed the data information, it can be classified as path vector learning data and vector clustering learning data; The path data can be input data generated by the user operating the information device 2, which can be a website trigger event (such as a webpage hyperlink), a website click event (such as clicking an advertisement), a website operation behavior (such as purchasing a product, Search for goods), time spent on a website, or a derivative data (such as shopping cart data, or product data included in purchased goods) under the aforementioned website operation behavior, but not limited to this data; , the aforementioned path data can respectively include a plurality of classification labels; the product database 23 mainly stores a product data, and the product data can be any one or a combination of a product type, a product name, a product price, a product function, but not Limited to this, the product data may be input data generated when the user operates the information device 2 , or product data pre-input from an external database, wherein the product data may include a plurality of classification labels respectively.

The artificial intelligence module 30 can be used for training and learning through the path vector learning data and the vector grouping learning data, converting the path data into a vectorized data, and then classifying a plurality of vectorized data into a grouping data, wherein the artificial intelligence Module 30 can use Supervised Learning, Semi-Supervised Learning, Reinforcement Learning, Unsupervised Learning, Self-Supervised Learning ( Self-Supervised Learning), or Heuristic Algorithms and other machine learning methods (Machine Learning) training and learning, but not limited to this.

Please refer to "Fig. 2". Fig. 2 is a flow chart of the implementation method of the creation. As shown in the figure, the implementation method of the system for awakening non-shopping consumers of this creation is as follows:

Receive path data 201 : The system 1 for awakening non-shopping consumers of the present creation receives a path data generated by a user operating an information device 2 , and a data processing unit 10 converts the path data based on a plurality of classification tags of a tag database 21 . The data is classified by tags, and then the path data with multiple classification tags is sent to a path database 22 for storage, wherein the path data can be a website trigger event (such as a webpage hyperlink), a website click event (such as clicking on an advertisement) ), a website operation behavior (such as purchasing goods, searching for goods), a website stay time, or any data or data combination of a derivative data under the aforesaid website operation behavior, but not limited to this, the above path data also It can be data input in advance from an external database; wherein, the derived data can be any data of a shopping cart, or a product data included in the purchased commodity, or a combination of data.

In one embodiment, please match "Fig. 3a" and "Fig. 3b". Fig. 3a and Fig. 3b are the implementation diagram (1) and the implementation diagram (2) of the creation, respectively. 2. Browsing a website page 301, input mountaineering off-road vehicle in a search unit 302 in the website page 301, and choose to browse 2 kinds of commodities, click a purchase unit 303 to buy one of the titanium alloy road bikes, and trigger an advertisement unit 304 3 advertisements, in which the website trigger event, website click event, website operation behavior, website stay time, and derivative data under the aforesaid website operation behavior generated by the user on the website page 301 will be marked by the data processing unit 10 Multiple classification labels; for example, the data processing unit 10 marks the user's search information "mountain bike" (path data) as a mountaineering label, a bicycle label, etc., or labels the purchased "titanium alloy bicycle" (derived from data) to mark a titanium alloy label, an outdoor sports label, etc., and then transmit the route data with a plurality of classification labels to the route database 22 for storage. The above examples are only examples and are not limited thereto.

Extracting analysis data 202: The data processing unit 10 extracts a plurality of path data in the path database 22 and at least one product data in the product database 23 for analysis by an artificial intelligence module 30, wherein the path data and the product The data respectively includes a plurality of classification labels, and the product data can be input data generated when the user operates the information device 2, or product data pre-input from an external database; Label the outdoor sports label and a waterproof material label on the special bag (input data) in advance; or, connect the external database with pre-labeled product classification by the system of this creation.

Vectorized grouping path data 203: The artificial intelligence module 30 performs vectorization analysis on the path data to generate a vectorized data, and then defines a plurality of vectorized data as a grouped data with a plurality of classification labels.

In one embodiment, please refer to "Fig. 4". Fig. 4 is a schematic diagram (3) of the implementation of the creation. As shown in the figure, the artificial intelligence module 30 stacks and converts a plurality of path data into a multi-dimensional vector matrix. User a stays on the website for 3 minutes and 45 seconds, clicks on 3 products on the website, and watches 2 advertisements set on the website for a total of 30 seconds, then the artificial intelligence module 30 converts the path data of user a into a vectorized data A1[0.33 , 2, 0.3] ([total stay time, number of clicked products, time to watch advertisements]), this creation is represented by a three-dimensional vector matrix, but not limited to this; the vectorized data A1~A6 can be vectors for different users For example, the vectorized data A2 can be the vectorized data of a user b, the vectorized data A3 can be the vectorized data of a user c, etc., and the tangent line t can represent the artificial intelligence module 30. Under a clustering training theme, the vectorized data A1~A6 are divided into two parts, wherein, the vectorized data A1~A3 can be divided into a clustering data G1, because the artificial intelligence module 30 is subject to different paths of vector learning data and vector clustering. The training of the learning data causes the tangent t to be different in slope and direction, which makes the grouped data different. The above examples are only examples and are not limited thereto.

Determine the wake-up target 204 : The data processing unit 10 determines a wake-up target based on a plurality of classification labels of the path data, that is, the data processing unit 10 determines the path data according to the path data with a plurality of classification labels in the path database 22 Whether the corresponding user is a wake-up target, wherein the wake-up target has multiple classification labels.

In an embodiment, please refer to "Fig. 5", which is a detailed flowchart of the implementation method of the creation. As shown in the figure, the data processing unit 10 extracts a piece of route data in the route database 22, and determines the corresponding user , whether the purchase time point is greater than the purchase period, if so, the user is listed as the wake-up target; if not, it is judged whether the purchase time point is greater than the product cycle of the previously purchased product, if so, the user is listed as the wake-up target; If not, the data processing unit 10 extracts another piece of path data in the path database 22, wherein the product cycle can be any one of the product life cycle of the product itself, the related product of the product, and the product life cycle of the related product itself or a combination thereof, but not limited to this; for example, if user a buys stationery once a month, but he has not purchased stationery for more than a month, user a is listed as a wake-up target; user b is one year Buy a mobile phone once, and buy another mobile phone when it is less than one year. Since the product life cycle of the mobile phone itself does not exceed the purchase cycle, it is judged based on the relevance of the product that user b may need related products, such as Bluetooth headsets, Or the mobile phone charging cable needs to be replaced, user b is still listed as the wake-up target.

Matching analysis result 205 : the artificial intelligence module 30 matches the grouping data with the product data based on the wake-up target to generate a matching data, that is, the artificial intelligence module 30 implicitly hides the grouping data according to the classification label of the wake-up target. The classification label of the product is matched with the implicit classification label of the product data, and a matching data is generated.

In an embodiment, please refer to "Fig. 6". Fig. 6 is a schematic diagram (4) of the implementation of the creation. As shown in the figure, the vectorized data B1 is the vectorized data of user d, and the vectorized data B2 is the user's The vectorized data of e, the vectorized data B3 is the vectorized data of user f, and the vectorized data B1 to B3 can be classified into a group of data G2; According to the grouping data G2 where the user e is located, the module 30 corresponds the implicit classification labels of the vectorized data B1 to B3 contained in the grouping data G2 to the classification labels of the product data respectively; After purchasing a tent, it will have a 3C product label, a mobile phone label, a mountaineering label, an outdoor sports label, etc. If the user has watched ski advertisements and purchased carbon fiber trekking poles, he will have a ski label, a carbon fiber label, an outdoor sports label, etc. Sports labels, mountaineering labels, etc., if user f has purchased a diving watch on an outdoor product website, he has a diving label, a 3C product label, an outdoor sports label, etc.; when the data processing unit 10 determines that user e is the wake-up target, according to the user According to the classification label of e, it is inferred that it may purchase a bicycle with an outdoor activity label and a carbon fiber label. At the same time, the artificial intelligence module 30, according to the grouping data G2 where the user e is located, divides the 3C product label, mobile phone and mobile phone implied by the grouping data G2. Labels, mountaineering labels, outdoor sports labels, etc., respectively correspond to the classification labels of product data. For example, the mobile power supply has a 3C product label and a mobile phone label. The artificial intelligence module 30 determines that the user e may need a mobile power supply, and then generates a mobile power supply containing the mobile power supply. matching data.

In one embodiment, please match "Fig. 7a" and "Fig. 7b". Fig. 7a and Fig. 7b are respectively the implementation diagram (5) and the implementation diagram (6) of this creation. As shown in the figure, the awakening of this creation is not The shopping consumer system 1 receives a sales promotion data 700 generated by the user operating the information device 2, wherein the sales promotion data 700 can be product data, and the product data can be a product type, a product name, a product price, and a product function. Any one or a combination thereof, but not limited thereto; the vectorized data C1 is the vectorized data of user g, the vectorized data C2 is the vectorized data of user h, and the vectorized data C1 and C2 may belong to is a grouping data G3; the data processing unit 10 judges that user h is the wake-up target based on the sales promotion data 700, and the artificial intelligence module 30 hides the vectorized data C1 and C2 contained in it according to the grouping data G3 where the user h is located. The category tags included correspond to the category tags of product data; for example, since user g has searched for low-priced Bluetooth earphones and purchased pens, there are 3C product tags, a wireless transmission tag, a price range tag, A stationery label, etc., the user h has clicked on the advertisement hyperlink of a simple home appliance and purchased a notebook computer, and it has a 3C product label, a wireless transmission label, a price range label, a home appliance label, etc.; when the user operates the information device 2. To sell a second-hand mobile phone, the data processing unit 10 determines that the user h is the wake-up target based on the sales data 700, and the artificial intelligence module 30 assigns the implied 3C product label and price range label according to the group data G3 where the user h is located. , wireless transmission labels, stationery labels, etc., corresponding to the classification labels of product data, such as household appliance labels and price range labels corresponding to sweeping robots, wireless transmission labels and stationery labels corresponding to recording pens, etc. The artificial intelligence module 30 determines the user h A sweeping robot and a voice recorder may be required to generate matching data containing the sweeping robot and voice recorder.

In one embodiment, the product price in the product data corresponds to a price range label implicit in the route data, and the price range label is used to define the spending power of the user; for example, if user i buys a high-priced mechanical watch, the With a high price tag and a watch tag, the artificial intelligence module 30 matches the implied high price tag and watch tag with the product data according to the group data of the user i, so the matching data will not include low-priced watches.

Extract product data 206 : The data processing unit 10 extracts a related product data related to the matching data in the product database 23 based on the matching data.

Sending product data 207 : The system 1 for awakening non-shopping consumers of this creation sends relevant product data to the information device 2 operated by the user, so as to provide more products that the user may purchase, as a reference for the user's shopping choice, or to provide Users refer to the matching data to promote more products to consumers.

It can be seen from the above that a system for awakening non-shopping consumers in this creation mainly uses the data processing unit to classify the path data labels generated by the user's operation of the information device based on a plurality of classification labels. The module converts the path data into vectorized data, and then classifies multiple vectorized data into grouped data; secondly, the data processing unit determines whether the user is a wake-up target according to the user's path data, such as purchase cycle and product cycle ;Furthermore, based on multi-dimensional considerations, the artificial intelligence module generates matching data based on the behavior of consumers and the attributes of commodities; finally, the data processing unit provides commodities that meet the needs of consumers based on the matching data, and transmits them. To the information device, it can be used as a reference for the user's shopping choice, or provide the user with reference matching data to increase the accuracy of the promotion of the product, thereby arousing the consumer's desire to buy the product, and awakening the consumer who has not shopped for a long time.

However, the above descriptions are only preferred embodiments of this creation, and are not intended to limit the scope of implementation of this creation; anyone who is familiar with this technique can make equal changes and modifications without departing from the spirit and scope of this creation. , shall be covered by the patent scope of this creation.

To sum up, this creation has the patent requirements of "industrial applicability", "novelty" and "progressiveness"; the applicant shall file an application for a new type patent with the Jun Bureau in accordance with the provisions of the Patent Law.

1: Wake up the non-shopping consumer system 2: Information device 10: Data processing unit 301: Website Page 21: Tag Database 302: Search Unit 22: Path Database 303: Purchase unit 23: Product Library 304: Ad Unit 30: Artificial Intelligence Modules 700: Sales Data 201: Receive path data 202: Extract analytics data 203: Vectorized clustering path data 204: Determine the wake-up target 205: Match Analysis Results 206: Extract product data 207: Send product data A1, A2, A3, A4, A5, A6: vectorized data B1, B2, B3: vectorized data C1, C2: vectorized data G1, G2, G3: grouping data t: Tangent

Figure 1 is the system architecture diagram of this creation. Figure 2 is a flow chart of the implementation method of this creation. Figure 3a is a schematic diagram (1) of the implementation of this creation. Figure 3b is a schematic diagram (2) of the implementation of this creation. Figure 4 is a schematic diagram of the implementation of this creation (3). Figure 5 is a detailed flow chart of the implementation method of this creation. Figure 6 is a schematic diagram of the implementation of this creation (4). Figure 7a is a schematic diagram (5) of the implementation of this creation. Figure 7b is a schematic diagram (6) of the implementation of this creation.

1: Wake up the non-shopping consumer system

10: Data processing unit

21: Tag Database

22: Path Database

23: Product Library

30: Artificial Intelligence Modules

2: Information device

Claims (6)

A system for waking up non-shopping consumers for informational connection with an informational device, comprising: A data processing unit is information-connected with a label database, a route database, and a product database, and is used for determining the corresponding route data based on a plurality of classification labels of a route data in the route database whether a user of the product database is a wake-up target, and extracts a related product data related to a matching data in the product database, and transmits the related product data to the information device; and An artificial intelligence module is connected with the data processing unit in information, and is used for vectorized analysis of the path data into a vectorized data, and then a plurality of the vectorized data are classified into a group data, and based on the wake-up target , matching the grouping data with at least one product data in the product database to generate the matching data. The system for awakening non-shopping consumers according to claim 1, wherein the label database includes the plurality of classification labels, which are used by the data processing unit to classify the labels for an input data transmitted from the information device. The system for awakening non-shopping consumers according to claim 1, wherein the path data is a website trigger event, a website click event, a website operation behavior, a website stay time, or a website operation behavior Derived data of any kind or combination of data. The system for waking up non-shopping consumers according to claim 1, wherein the data processing unit is configured to determine the path data corresponding to a purchase time point, a purchase cycle, and a product cycle corresponding to the category label. Whether the user is the wakeup target. The system for waking up non-shopping consumers according to claim 4, wherein the data processing unit determines that the user is listed as the waking target if the user's purchase time point is greater than the purchase period. The system for waking up non-shopping consumers according to claim 5, wherein the data processing unit determines that the user is listed as the waking target if the user's time to purchase is greater than the product cycle.

Images