TW202139665A - Distributed customer behavior clustering system - Google Patents

Abstract

The disclosure provides a distributed customer behavior clustering system. This system includes an analysis application and a cloud server. The analysis application may be installed on a plurality of users’ mobile devices for operating, and the analysis application may calculate each of the plurality of users’ behavior clustering result based on their browsing behaviors and operation manners correspondingly. The cloud server may receive the users’ behavior clustering results from the plurality of mobile devices, so as to accumulate a distributed clustering result. It’s worth mentioning that the browsing behavior may include a detention period of browsing, a monitor swiping of browsing, a tag click on a webpage or an article semantic analysis. The operation manner may include a zoom out/in operation, time of operation, and number of operations.

Description

Distributed user behavior grouping system

The following description is about a grouping system, especially a distributed grouping system that can collect the behaviors of different users on mobile devices and then group them.

Due to the rise of mobile devices, many humans’ work and rest are inseparable from mobile devices and the Internet, such as using Google maps for navigation, online shopping malls, watching YouTube videos on mobile devices, or using mobile devices to watch tribes. Articles, etc., all show how important mobile devices are to humans.

Furthermore, because users have become accustomed to using a large number of mobile devices, there are currently relevant documents discussing how to explore user behaviors or hobbies through mobile devices. , The manufacturer can display appropriate product content or articles on the mobile device according to its preference, which can attract the attention of users and increase the click-through rate, and secondly, it can increase the chance of the product being purchased.

Generally speaking, most of the current methods of investigating user behavior can be carried out by using a server. The method is that when the user uses a mobile device to browse the website set up by this server, the server will record the user's click to Page, and further analyze the content text of this page, and then extract relevant or frequently occurring keyword terms, or directly analyze the label selected by the user to infer what the user is interested in What is the content.

However, the results of this exploration method are often unsatisfactory due to some factors. For example, users sometimes mistakenly select pages or advertisement windows in the website, which may result in incorrect analysis of the content or use it when used. When the user clicks on two different pages, and only one of them is the content that the user is interested in, traditional exploration methods cannot interpret the differences between the two pages. Furthermore, the records of all users using mobile devices are related to the content of this server's website, so the results obtained will also be limited to the content of the website, and it is impossible to further explore the user's true hobbies or interests. .

In summary, the inventor of the present invention thought about and designed a decentralized user behavior grouping system, in order to improve the lack of conventional technology, and further enhance the implementation and utilization of the industry.

Based on the above objective, the present invention provides a distributed user behavior grouping system, which includes an analysis application and a cloud server. The analysis application program is installed on a plurality of mobile devices for operation by a plurality of users, and the analysis application program is calculated to generate a user behavior grouping result based on a browsing behavior and an operation method of each plurality of users . The cloud server receives the user behavior grouping results generated by the plurality of mobile devices to accumulatively generate a distributed grouping result.

Preferably, the browsing behavior includes a browsing page stay time, a browsing page sliding ratio, a page tag click item, or a page article topic analysis.

Preferably, the analysis application includes a pre-stored analysis algorithm, and the pre-stored analysis algorithm is updated by a cloud server.

Preferably, the cloud server updates the pre-stored analysis algorithm according to the distributed grouping result.

Preferably, the analysis application program assigns a weight to the grouping result of the user behavior according to the operation mode.

Preferably, the operation mode includes an enlargement/reduction operation, an operation time, and an operation number.

Preferably, the mobile phone application program includes an e-commerce platform software, a tool application software, or a browser software.

It can be known from the above that the distributed user behavior grouping system of the present invention can achieve the following technical effects.

(1) The analysis application can directly analyze the user and upload the user behavior grouping results generated by the analysis to the cloud server. In this way, the load of the cloud server will be greatly reduced.

(2) The analysis application can further analyze the user's user behavior based on the operation mode and browsing behavior, which can effectively filter out information that may cause misjudgment, thereby improving the accuracy of grouping.

(3) The pre-stored analysis algorithm of the analysis application can be continuously updated by the cloud server. It is worth mentioning that the cloud server is updated based on the content of the distributed grouping result, that is, all users are used Compared with the conventional technology, the data analyzed by the distributed user behavior grouping system of the present invention is also more objective.

In order to help your examiners understand the features, content and advantages of the present invention and the effects that can be achieved, the present invention is described in detail with the accompanying drawings and in the form of embodiment expressions. The figures used therein are as follows: The subject matter is only for the purpose of illustration and auxiliary description, and may not be the true proportions and precise configuration after the implementation of the present invention. Therefore, the proportion and configuration relationship of the attached drawings should not be interpreted as to limit the scope of rights of the present invention in actual implementation.

The advantages, features, and technical methods of the present invention will be described in more detail with reference to exemplary embodiments and the accompanying drawings to make it easier to understand, and the present invention can be implemented in different forms, so it should not be understood to be limited to what is here. The stated embodiments, on the contrary, for those with ordinary knowledge in the technical field, the provided embodiments will make this disclosure more thorough, comprehensive and complete to convey the scope of the present invention, and the present invention will only be additional Defined by the scope of the patent application.

Please refer to Figure 1, which is a block diagram of the distributed user behavior grouping system of the present invention. As shown in the figure, the distributed user behavior grouping system 1 of the present invention can include an analysis application 10 and a cloud server 20. The analysis application 10 is installed on a plurality of mobile devices 50 for operation by a plurality of users 40, wherein the plurality of mobile devices 50 can be a smart phone, a tablet computer or a notebook computer with an Internet function, etc. The cloud server 20 can be a workstation, a server, a notebook computer, or a desktop computer.

When the analysis application 10 is executed on the mobile device 50, it can calculate and generate a user behavior grouping result 14 according to a browsing behavior 11 and an operation method 12 of each user 40 on the mobile device 50.

On the other hand, the cloud server 20 can receive the user behavior grouping results 14 generated by the plurality of mobile devices 50 to generate a distributed grouping result 21 cumulatively.

In a preferred embodiment, the distributed user behavior grouping system of the present invention may further include a mobile phone application 30, wherein the mobile phone application 30 may include an e-commerce platform software, a tool application software, or a browser Software, and each user 40 executes the mobile phone application 30 to generate the above-mentioned browsing behavior 11 and operation mode 12, and the analysis application 10 performs monitoring/calculation.

Different from the prior art, the distributed user behavior grouping system 1 of the present invention utilizes the browsing behavior 11 and operation mode 12 of the user 40 to explore the user's preferences, and then further group. In detail, the browsing behavior 11 may include a dwell time of a browsed page, a sliding ratio of a browsed page, a tab click item on a page, or an article topic analysis on a page. The operation mode 12 may include a zoom-in/out operation, an operation time, and a number of operations. The analysis application 10 can assign a weight to the user behavior grouping result 14 according to the operation mode 12, and examples are provided for each. as follows.

When the user 40 browses to a website page of interest, generally speaking, the user 40 will stay on this website page for a longer viewing time, and the sliding method of this page will gradually reach the bottom of the page instead of Going directly to the bottom of the page, instead of just staying at the top of the page, the analysis application 10 can determine whether the website page should be included in the grouping data based on the browsing behavior 11. In addition, if the operation mode 12 of the user 40 has an operation of zooming in/out the screen, or an increase in the number of operations (such as viewing this page repeatedly), then it can be more sure that the current website page is what the user 40 feels. Interested, and the analysis application 10 will weight the grouping results analyzed on this page, so as to increase the weight of the analysis results of this page in the user behavior grouping results 14 of the user 40.

Or, when the user 40 browses a shopping website, since the basic structure of the shopping website is a one-level design, if the user 40 uses a hierarchical way to select page tags, for example, the homepage of the website à home appliances à refrigerator à brand , It can be determined that the user 40 is indeed interested in this home appliance product, but if the user 40 uses a tabbed selection method that escapes this hierarchy, such as the homepage à home appliances à furniture, the user 40 can be determined No interest in home appliances, or the user 40 mistakenly clicked on the home appliance page.

Through the above-mentioned method, the distributed user behavior grouping system of the present invention can accurately determine which browsing pages are of interest to the user, and the browsing information that is not interested or mistakenly clicked will be filtered out. In this way, the analysis application 10 can more accurately calculate the user behavior grouping result 14 of the user 40, where the grouping category can be freely defined, such as general office workers group, petty bourgeoisie group, housewife group Group, middle school student group, high spending power group or e-sports player group, etc. Then, by uploading the user behavior grouping result 14 to the cloud server 20, the grouping results of all users can be cumulatively output, and the decision maker can make the best decision based on the distributed grouping result 21.

Please refer to Figure 2, which is a schematic diagram of the first embodiment of the distributed user behavior grouping system of the present invention. Please also refer to Figure 1. The analysis application 10 in the distributed user behavior grouping system 1 of the present invention may include a pre-stored analysis algorithm 13, and the pre-stored analysis algorithm 13 can be performed by the cloud server 20 through the network Remote software update. Furthermore, this pre-stored analysis algorithm 13 can be mainly used as a basis for clustering in the analysis application 10. It can include the K-means algorithm, the Hierarchical Clustering algorithm, the Density Based Clustering algorithm, etc., but not based on this. limit.

Furthermore, the cloud server 20 can update the pre-stored analysis algorithm 13 on each mobile device 50 according to the distributed grouping result 21 obtained by calculation. In other words, the distributed user behavior grouping system 1 of the present invention can first collect the grouping results of all users, and then update the analysis application 10 on each mobile device 50 based on the grouping results. In this way, the application can be analyzed 10 can be fine-tuned, and the content of the distributed grouping result 21 on the cloud server 20 can also be gradually corrected.

In addition, the cloud server 20 can also store various product information or product services in advance. After the distributed grouping result 21 is generated, the cloud server 20 can find a corresponding recommended product according to the distributed grouping result 21 Information 22, and then update the recommended product information 22 to the pre-stored analysis algorithm 13. For example, if a consumer belongs to a 3C user group, the cloud server 20 can search for pre-stored 3C product information and update it to the consumer's analysis application 10, and then the consumer will use the analysis application 10 At the time, the 3C product information will be pushed on the mobile device 50 for the consumer to browse, thereby enhancing the consumer's desire to buy.

Based on the above, it can be seen that the distributed user behavior grouping system of the present invention can indeed directly analyze users, and upload the analyzed grouping results to the cloud server for accumulation. In this way, the load of the cloud server will be reduced. In addition, the user’s behavior can be further analyzed according to the operation mode and browsing behavior, which can effectively filter out the information that may cause misjudgment, thereby improving the accuracy of grouping. Moreover, in the present invention The pre-stored analysis algorithm is updated with the results of big data, that is, the results of the behavior analysis of all users are used. Therefore, compared with the conventional technology, the distributed user behavior grouping system of the present invention should indeed be novel Nature, advancement and industrial utilization.

The above-mentioned embodiments are only to illustrate the technical ideas and features of the present invention, and their purpose is to enable those who are familiar with the art to understand the content of the present invention and implement them accordingly. When they cannot be used to limit the patent scope of the present invention, That is, all equal changes or modifications made in accordance with the spirit of the present invention should still be covered by the patent scope of the present invention.

1: Decentralized user behavior grouping system 10: Analyze the application 11: Browsing behavior 12: Operation method 13: Pre-stored analysis algorithm 14: User behavior grouping results 20: Cloud server 21: Distributed clustering results 22: Recommended product information 30: mobile app 40: User 50: mobile device

Figure 1 is a block diagram of the distributed user behavior grouping system of the present invention. Figure 2 is a schematic diagram of the first embodiment of the distributed user behavior grouping system of the present invention.

1: Decentralized user behavior grouping system

10: Analyze the application

11: Browsing behavior

12: Operation method

13: Pre-stored analysis algorithm

14: User behavior grouping results

20: Cloud server

21: Distributed clustering results

30: mobile app

40: User

50: mobile device

Claims (9)

A decentralized user behavior grouping system, which includes: An analysis application is installed on a plurality of mobile devices for operation by a plurality of users. The analysis application is calculated based on a browsing behavior and an operation method of each of the plurality of users to generate a user behavior grouping Result; and A cloud server receives the user behavior grouping results generated by the plurality of mobile devices to accumulatively generate a distributed grouping result. The distributed user behavior grouping system according to claim 1, wherein the browsing behavior includes a browsing page dwell time, a browsing page sliding ratio, a page tag click item, or a page article topic analysis. The distributed user behavior grouping system according to claim 1, wherein the analysis application includes a pre-stored analysis algorithm, and the pre-stored analysis algorithm is updated by the cloud server. The distributed user behavior grouping system according to claim 3, wherein the cloud server updates the pre-stored analysis algorithm according to the distributed grouping result. According to the distributed user behavior grouping system described in claim 3, the cloud server searches for corresponding recommended product information according to the distributed grouping result, and updates the recommended product information to the pre-stored analysis algorithm. The distributed user behavior grouping system according to claim 1, wherein the analysis application program performs a weight assignment on the user behavior grouping result according to the operation mode. The distributed user behavior grouping system according to claim 6, wherein the operation mode includes a zoom in/out operation, an operation time, and a number of operations. The distributed user behavior grouping system according to claim 1, further comprising a mobile phone application program, and each of the plurality of users executes the mobile phone application program to generate the browsing behavior and the operation mode. The distributed user behavior grouping system according to claim 8, wherein the mobile phone application includes an e-commerce platform software, a tool application software, or a browser software.

Images