TWI642024B - Method of providing recommended services and data processing system thereof - Google Patents

Abstract

A data processing system for providing a recommendation service to a user, comprising a label generating unit, configured to use a user data input by the user, a user input log, a user behavior log, and a group to which the user belongs At least one of the group member materials generates at least one tag; and a recommendation service generating unit is configured to recommend at least one of the plurality of services to the user according to the at least one tag.

Description

Recommended service method and related data processing system

The invention relates to a method for providing a recommendation service and a related data processing system, in particular to a method for providing a recommendation service according to user behavior and a related data processing system.

With the development and popularization of electronic devices such as smart phones, tablets, and notebook computers, they are increasingly indispensable in modern life. For example, a user can browse a webpage, play a game, send and receive an email, etc. through a smart phone. In view of the rise in health awareness in recent years, more and more people are paying attention to physical health. For example, the wearer-type device detects and records physiological conditions to assess physical health.

In addition, because physical health is related to living habits and environmental conditions, different types and degrees of diseases may be caused when users are in poor living habits and environmental conditions for a long time. For example, users who lack exercise and stay up all night have a higher risk of developing cardiovascular disease. Users who are in an environment with insufficient illumination and visual targets (such as fonts and images) that are too small or unclear are prone to visual fatigue.

Therefore, how to effectively use the electronic devices around the user to collect user information and related user behaviors to assess the user's living habits, environmental conditions and physical health conditions, and then recommend appropriate services is an emerging issue in the industry.

Accordingly, it is a primary object of the present invention to provide a method of providing a recommendation service and a related data processing system.

The present invention discloses a data processing system for providing a recommendation service to a user, which includes a person database, a log database, a group database, a label generation unit, a label database, a service database, and A recommendation service generation unit. The personal database is used to store a person's data entered by the user. The log database is used to store a user input log and a user behavior log of the user. The group database is used to store a group member data of the group to which the user belongs. The tag generating unit is coupled to the personal data database, the log database, and the group database, and is configured to use the personal data input by the user, the user input log, the user behavior log, and the group member data. At least one of the at least one tag is generated. The tag database is coupled to the tag generating unit for storing the at least one tag. The service database is used to store a plurality of services. The recommendation service generating unit is coupled to the tag database and the service database, and is configured to recommend at least one of the plurality of services to the user according to the at least one tag.

The present invention further discloses a method for providing a recommendation service to a user, the method comprising: according to a user profile of the user, a user input log, a user behavior log, and a group member profile of the group to which the user belongs. At least one generates at least one tag; and generates at least one recommendation service based on the at least one tag.

The data processing system of the present invention can utilize the electronic devices around the user to collect user information and user behavior (ie, personal data, user input logs, user behavior logs, group member data, etc.), thereby generating a description for use. The label of the trait, and then recommend the appropriate service to the user to be close to the user's needs.

Today's smart phones, tablets, notebooks, and wearable devices often have multiple sensors (such as accelerometers, light-sensing components, radios, optical heart rate monitors, body motion recorders, etc.) to generate inductive results. Then, through the processor, data processing is performed on the sensing result to generate a corresponding operation instruction, thereby achieving the purpose of expanding the function of the device. In general, when a user executes an application through an electronic device, the user's operation behavior can be stored as a log file (log file, hereinafter referred to as a log), which is used as a reference for debugging and updating the application in the future. Therefore, the present invention can utilize the electronic device around the user to collect user information and user behavior (ie, the sensing result and the user log), thereby generating a label for describing the user's characteristics, and then recommending an appropriate service to the user. To be close to the needs of users.

In detail, the present invention collects the sensing results generated by the sensors, the user logs generated by the application, and analyzes the user's personal data, analyzes them and creates a relationship to generate tags for describing user characteristics, wherein the user characteristics can be For physiological conditions, living habits, life needs, etc. Further, the present invention recommends an appropriate service to the user in accordance with the label for describing the user's characteristics to be close to the user's needs. For example, when the above information indicates that "the user has more than four days in the week and the last use time is the morning hours", the "Insomnia" label will be generated and the related "Hanging Hypnotic Music" recommendation service will be generated. For example, in an actual application, when a user browses an online video streaming platform, the video streaming platform can preferentially display the "hypnotizing hypnotic music" service to the user.

FIG. 1 is a schematic diagram of a data processing system 1 according to an embodiment of the present invention. The data processing system 1 includes a mobile device 10, a wearable device 12, and a cloud server 14. The mobile device 10 can be a smart phone, a tablet computer, a notebook computer, or the like. The wearable device 12 can be a wearable device in the form of a wristband, collar, ring, clothing, and the like. The mobile device 10 and the wearable device 12 respectively include a processor, a memory, a sensing module, a wireless communication module, and a display (not shown in FIG. 1) to implement related functions. For example, the mobile device 10 and the wearable device 12 can be connected by Bluetooth wireless communication through a wireless communication module. The mobile device 10 can access the Internet through a wireless communication module via a WiFi or mobile communication system to connect to the cloud server 14. The mobile device 10 and the wearable device 12 can perform data processing and storage through the processor and the memory, respectively, and display the data processing result through the display.

In an embodiment, the mobile device 10 can generate a log based on the sensing result output by the sensing module of the sensing module and the wearable device 12, and the log generated when the software program is executed (for example, web browsing history, usage time, etc.) And user profiles (eg, birthday/age, gender, height, weight, etc.), generating user tags and referral services. In one embodiment, the sensing module of the mobile device 10 is configured to generate an environmental sensing result (eg, ambient brightness) of the user, and the sensing module of the wearable device 12 is configured to generate a physiological sensing of the user. Results (eg, heart rate, blood pressure, sleep quality, stress index, etc.). In one embodiment, the mobile device 10 can learn other environmental conditions (eg, daily and predicted temperature/UV index/rain rate/air quality, etc.) through the Internet and the positioning system.

In an embodiment, the cloud server 14 can serve as a cloud database, wherein the user can connect to the cloud server 14 via the Internet through the user interface of the mobile device 10, and establish a person in the cloud server 14. An account to upload at least one of a user profile, a sensory result, a user log, a user tag, and a recommendation service to the cloud server 14.

In an embodiment, the cloud server 14 can serve as a central processing unit, wherein the mobile device 10 can integrate the sensing result and the user log of the sensing module of the wearable device 12 and upload it to the cloud server 14 . The cloud server 14 is caused to generate a user tag and a recommendation service according to the sensing result.

In an embodiment, the cloud server 14 and the mobile device 10 may further store the group member data of the group to which the user belongs, and let the cloud server 14 or the mobile device 10 according to the user profile, the sensing result, the user log, and the group. Member information, generating user tags and referral services.

FIG. 2 is a functional block diagram of a data processing system 2 according to an embodiment of the present invention. The data processing system 2 includes a user database 20, a log database 21, a group database 22, a label generating unit 23, a label database 24, a service database 25, and a recommendation service generating unit 26.

The user database 20 is used to store a person's data (eg, birthday/age, gender, height, weight, etc.) input by the user, wherein the personal data is input by the user through a user interface of the mobile device 10, and thus the user Personal data is considered fixed if no changes are made.

The log database 21 is configured to store a user input log and a user behavior log of the user, wherein the user can input a user input log (for example, daily or hourly water consumption) through a user interface of the mobile device 10. The mobile device 10 can input a user behavior log according to the operation record generated when the user operates the mobile device 10 (for example, web browsing history, mobile device usage time, ambient brightness, daily and predicted temperature/UV index/raining) Probability / air quality, etc.). In an embodiment, when the user wears the wearable device 12, the mobile device 10 can receive the physiological induction result (eg, heart rate, blood pressure, sleep quality, pressure index, etc.) of the user from the wearable device 12 as A part of the user behavior log. In short, the user input log is generated by the user input, and the user behavior log is generated by a device other than the user (for example, a mobile device or a wearable device).

In an embodiment, the user input log and the user behavior log are used to continuously record the user behavior, and by comparing the change value and the change trend of the user behavior, it is beneficial to track the physiological and environmental conditions of the user for a long time as a possible future. Recommended service basis.

The group database 22 is used to store a group member data (such as birthday/age, gender, height, weight, body mass index, etc.) of the group to which the user belongs. In an embodiment, the group database 22 may additionally store user input logs and user behavior logs of group members.

The tag generating unit 23 is coupled to the personal data base 20, the log database 21, and the group database 22 for generating at least one of at least one of personal data, user input logs, user behavior logs, and group member data. A label, where the label is used to describe user traits (eg, physiological condition, physiological needs, life needs, etc.). The tag database 24 is coupled to the tag generating unit 23 and the recommendation service generating unit 26 for storing tags generated by the tag generating unit 23.

The service database 25 is coupled to the recommendation service generation unit 26, which can be regarded as a service store for storing different types of services, which can be products, knowledge and labor related services, to meet physiological needs and living needs. The recommendation service generating unit 26 is coupled to the tag database 24 and the service database 25 for recommending at least one of the plurality of services to the user according to the tag.

In operation, in an embodiment, the label generating unit 23 can obtain the web browsing history of the user from the user behavior log, and perform word segmentation, classification, and calculation of the frequency of the webpage title and the article content to determine the webpage. Keywords in the middle. Next, the tag generating unit 23 performs weighting rules and conversion calculations on the keywords (for example, by establishing a machine learning or neural network model, and performing weighting rules and conversion calculations on the neural nodes) to generate related tags. For example, when the label generating unit 23 determines that "the article with the "knee" keyword is frequently clicked, and the frequency of searching for the "arthritis" keyword is greater than the average value in the rainy day, the "arthritis patient" label may be generated. Therefore, the recommendation service generation unit 26 can generate a related "arthritis consultation" recommendation service. For example, in practical applications, when a user browses a health theme website, the website can preferentially display the "arthritis consultation" service to the user.

In an embodiment, the tag generating unit 23 can obtain the user's personal data and the member data of the member of the group from the personal database 20 and the group database 22, respectively, and determine the relationship between the user and the group member. For example, when the tag generation unit 23 determines that "the user is a female between the ages of 20 and 40, and the member of the group has a member under the age of 5", the "mother" label may be generated, so it is recommended. The service generation unit 26 can generate a recommendation service related to the "mother" tag. For example, in an actual application, when a user purchases a website online, the website can preferentially display a "mother and baby product" service to the user. Therefore, by referring to the member data of the group member, the object of the recommendation service can be extended to the group members in the group to which the user belongs to expand the service object. For example, the recommended services for the "Mother and Baby Products" service can be "Mom" and "Child".

In this way, the data processing system 2 can use the electronic devices around the user to collect user information and user behavior (ie, personal data, user input logs, user behavior logs, group member data, etc.), thereby generating A label describing the characteristics of the user, and then recommending appropriate services to the user to be close to the user's needs.

The operation modes of the data processing systems 1 and 2 described above can be summarized as a process 3. As shown in FIG. 3, the process 3 can be compiled into a code (for example, a software application) and stored in the mobile device 10 or the cloud server. The memory of 14 is used to instruct the processor of the mobile device 10 or the cloud server 14 to perform the following steps.

Step 30: Start.

Step 31: Generate at least one label according to at least one of a personal data input by the user, a user input log, a user behavior log, and a group member data of the group to which the user belongs.

Step 32: Generate at least one recommended service according to at least one tag.

Step 33: End.

For details on the operation mode of the process 3, refer to the figures 1 and 2 and related descriptions, and details are not described herein.

The operation mode of the tag generating unit 23 can be summarized as a process 4. As shown in FIG. 4, the process 4 can be compiled into a code (for example, a software application) and stored in the mobile device 10 or the cloud server 14. The memory is used to instruct the processor of the mobile device 10 or the cloud server 14 to perform the following steps.

Step 40: Start.

Step 41: Obtain a webpage browsing record of the user's user behavior log, and perform word disassembly and classification on the title of the webpage and the content of the webpage.

Step 42: Repeat the frequency according to the words to determine the keywords in the webpage.

Step 43: Perform weighting rules and conversion calculations on the keywords according to at least one of personal data, user input logs, and group member data to generate at least one tag.

Step 44: End.

For details on the operation mode of the process 4, refer to FIG. 2 and related descriptions, and details are not described herein.

In summary, the data processing system of the present invention can utilize the electronic devices around the user to collect user information and user behavior (ie, personal data, user input logs, user behavior logs, group member data, etc.). A label is generated to describe the characteristics of the user, and an appropriate service is recommended to the user to be close to the user's needs. The above are only the preferred embodiments of the present invention, and all changes and modifications made to the scope of the present invention should be within the scope of the present invention.

1, 2‧‧‧ data processing system

10‧‧‧Mobile devices

12‧‧‧Wearing device

14‧‧‧Cloud Server

20‧‧‧User database

21‧‧‧Log database

22‧‧‧Group Database

23‧‧‧ Label Generation Unit

24‧‧‧Label database

25‧‧‧Service Database

26‧‧‧Recommended service generation unit

3, 4‧‧‧ process

30, 31, 32, 33, 40, 41, 42, 43, 44‧ ‧ steps

FIG. 1 is a schematic diagram of a data processing system according to an embodiment of the present invention. FIG. 2 is a functional block diagram of a data processing system according to an embodiment of the present invention. FIG. 3 is a flowchart of a recommended service flow according to an embodiment of the present invention. FIG. 4 is a flowchart of a process for generating a label according to Embodiment 1 of the present invention.

Claims (10)

A data processing system for recommending a service to a user, comprising: a person database for storing a person data input by the user; a log database for storing a user input log of the user and a a user behavior log, wherein the user behavior log is a mobile device inputting the user behavior log to the log database according to an operation record generated when the user operates the mobile device; a group database for storing the usage a group member data of the group to which the group belongs; a tag generating unit coupled to the personal data database, the log database, and the group database for inputting the user input log according to the personal data input by the user At least one of the user behavior log and the member information of the group generates at least one tag, wherein the at least one tag is used to describe a user trait of the user; a tag database coupled to the tag generating unit For storing the at least one tag; a service database for storing a plurality of services; and a recommendation service generating unit coupled The tag database and the service database are configured to recommend at least one of the plurality of services to the user according to the at least one tag; wherein the user behavior log includes a web browsing record, and the tag generating unit is configured according to the webpage Navigating the record, disassembling, classifying, and calculating the frequency of the title of the webpage and the content of the article to determine at least one keyword of the webpage, and the label generating unit according to the personal data, the user input log, and the group At least one of the member materials performs weighting rules and conversion calculations on the at least one keyword to generate the at least one tag. The data processing system of claim 1, wherein the user inputs the personal data to the personal database through a user interface of the mobile device and inputs the user input log to the log database. The data processing system of claim 2, wherein the user behavior log includes usage time of the mobile device, ambient brightness, daily and predicted air temperature, ultraviolet index, rain rate, air quality, and the user behavior log is based on A sensor module of the mobile device is produced. The data processing system of claim 2, wherein the user behavior log includes at least one of a heart rate, a blood pressure, a sleep quality, and a stress index of the user, and the user behavior log is based on a sensing of a wearable device Generated by the module. The data processing system of claim 1, wherein the personal data and the group member data respectively comprise at least one of a height, a weight, a birthday, and a gender of the user and the member of the group. A method for providing a recommendation service to a user, the method comprising: at least one of a user profile, a user input log, a user behavior log, and a group member profile of the group to which the user belongs Generating at least one tag, wherein the at least one tag is used to describe a user trait of the user, wherein the user behavior log is a mobile device inputting the user behavior according to an operation record generated when the user operates the mobile device Logging to the log database; and generating at least one recommendation service according to the at least one tag, wherein the user behavior log includes a webpage browsing record, including: disassembling the title of the webpage and the content of the article according to the browsing history of the webpage , classifying, and calculating the frequency of occurrence to determine at least one keyword of the web page; and according to at least one of the personal data, the user input log, and the member information of the group, The at least one keyword performs a weighting rule and a conversion algorithm to generate the at least one tag. The method of claim 7, wherein the user inputs the personal data to the personal database through a user interface of the mobile device and inputs the user input log to the log database, and the user behavior log is The mobile device inputs the user behavior log to the log database based on an operation record generated by the user when the mobile device is operated. The method of claim 7, wherein the user behavior log includes usage time of the mobile device, ambient brightness, daily and predicted temperature, ultraviolet index, rain rate, air quality, and the user behavior log is based on the action The sensing result produced by a sensing module of the device is generated. The method of claim 7, wherein the user behavior log includes at least one of a heart rate, a blood pressure, a sleep quality, and a stress index of the user, and the user behavior log is a sensing module through a wearable device. The resulting inductive results are produced. The method of claim 6, wherein the personal data and the group member data respectively comprise at least one of a height, a weight, a birthday, and a gender of the user and the group member.