1. A UI design method based on big data analysis is characterized in that: collecting driving information data of a user and uploading the driving information data to a cloud background server; the cloud background server performs data processing and analysis on the collected driving information data to form a data report based on the use habits of the user; and designing various vehicle UI interfaces according to the data report of the use habits of the user for the user to select.

2. The method for designing the UI based on the big data analysis as claimed in claim 1, wherein the vehicle and the cloud server need authentication when communicating, and the data is encrypted and transmitted by using an encryption algorithm for ensuring the security of the transmitted data.

3. The big data analysis-based UI design method of claim 1, wherein the cloud backend server performs data processing and analysis on the collected driving information data to form a data report based on the usage habits of the user, comprising the steps of:

step 1, data cleaning, namely, eliminating unusable data through data processing software, identifying effective usable data and obtaining neat data in a specific format;

step 2, data processing, namely extracting and converting the cleaned clean data in the step to obtain a user data set containing frequency and duration data of each function operated by a user;

step 3, data analysis, namely processing the data by using a machine learning mode, and analyzing the individual or group of the user by using a data analysis algorithm to obtain a data statistical report;

and 4, generating a user use habit report, a cluster analysis report and an intelligent recommendation report according to the generated data statistics report.

4. The method for big data analytics based UI design according to claim 3, wherein in step 1, the format includes CSV, JSON, part.

5. The method for designing the UI based on the big data analysis according to claim 1, wherein a plurality of vehicle-mounted machine UI interfaces are designed according to the data report of the user use habit for the user to select, and the method comprises the steps of feeding back the vehicle-mounted machine UI interfaces to a vehicle-mounted machine through a T-BOX after the vehicle-mounted machine UI interfaces are recombined and designed, recommending the UI interfaces which can be selected by the vehicle-mounted machine to the user in a popup window mode, and enabling the user to select the UI interfaces by himself.

6. A system for UI design based on big data analysis is characterized by comprising a data transmission module: the system is used for transmitting data between the vehicle and the cloud background server; a data cleaning module: for identifying valid available data; a data preprocessing module: for selecting clean user information to be analyzed; a data analysis module: the data analysis system is used for analyzing the user data set to the user individuals or groups through a data analysis algorithm to obtain a data statistics report; a data report generation module: and generating a user use habit report, a cluster analysis report and an intelligent recommendation report according to the data statistics report.

7. The big-data-analysis-based UI design system according to claim 6, further comprising a data encryption module for encrypting the transmitted data to ensure the security of data transmission.

8. The system of big-data-analysis-based UI design according to claim 6, wherein the data pre-processing module further comprises a data extraction module and a transformation module: the method is used for selecting neat user information to be analyzed, and facilitates later data analysis.

9. The big-data-analysis-based UI design system according to claim 6, wherein the data analysis module further comprises a data sorting module for sorting data according to set rules; a cluster analysis module: the method is used for continuously improving the portrait precision of the user and laying a foundation for the self-learning of the next step; a self-learning module: and the computer continuously iterates and updates according to the formed data training set, so that the user behavior is predicted in advance, and the intelligence and the accuracy of recommendation are improved.

10. The system of big-data-analysis-based UI design according to claim 6, further comprising an intelligent recommendation module: and recommending a UI design interface based on big data analysis to a user.

Background

As automobiles are increasingly integrated with more functions by adopting a large screen of an MP5 vehicle, including much information such as weather, navigation, multimedia control, bluetooth phone, air conditioning control, vehicle conditions, vehicle control, brand zone, etc., it is important to design a vehicle interface that meets user habits. At present, the car machine interface design mostly depends on the experience of designers or the experience of evaluation group personnel, or the traditional questionnaire survey mode obtains the use frequency of each function. However, the method relying on the experience of the designer cannot represent the actual use habit of the user, and the adoption of the questionnaire mode may cause data deviation due to different subjective understandings of the questionnaire by the person to be investigated, so that the method is not suitable for the situation that a large amount of data needs to be processed, the survey data of a large number of samples cannot be formed, and the final result may be different from the user habit due to the insufficient sample amount.

According to the method, cloud background data used by the vehicle machine is cleaned and mined by utilizing a big data technology to form a user use habit data report, and interface priority level and hierarchy definition design is guided during UI design according to actual use frequency data of different functions. Meanwhile, the big data technology can be combined with the big data of the user to perform cluster analysis, and meanwhile, the automatic learning technology is used to perform intelligent recommendation, so that the system and the interface thereof are more in line with the actual use habits of the user.

Disclosure of Invention

The invention aims to provide a method for designing a vehicle-mounted machine interface which best meets the use habit of a user by utilizing cloud background big data and combining a big data processing technology.

The UI design method based on big data analysis for realizing one purpose of the invention comprises the following steps: collecting driving information data of a user and uploading the driving information data to a cloud background server; the cloud background server performs data processing and analysis on the collected driving information data to form a data report based on the use habits of the user; and designing various vehicle UI interfaces according to the data report of the use habits of the user for the user to select.

The current car machine interface has integrateed more and more functions, including weather, navigation, multimedia control, bluetooth telephone, air conditioner control, vehicle setting, brand special district etc. a lot of information, and when traditional car machine interface design, often according to designer's experience or evaluation team personnel's experience, each function interface of design is in first level or second level etc. possesses certain subjectivity. But the display interface of the car machine is limited, and the display information can be reasonably arranged in a grading way according to the use habit and the use frequency. According to the method, cloud background data used by the vehicle machine is cleaned and mined by utilizing a big data technology to form a user use habit data report, and interface priority level and hierarchy definition design is guided during UI design according to actual use frequency data of different functions. Meanwhile, the big data of the user can be combined, the big data technology is utilized to perform clustering analysis, various users with different habits are formed, the vehicle interface is defined into various interfaces with different styles, and the user can select the interfaces by himself. Background data used by a user is the truest data capable of reflecting the habits of the user, so that the scheme can be guaranteed to be designed to be in accordance with the use habits of the user. Meanwhile, the intelligent recommendation system and the interface thereof can be combined with the big data of the user, the big data technology, the multidimensional association algorithm and the automatic learning technology to carry out intelligent recommendation, so that the system and the interface thereof are more in line with the actual use habits of the user.

The driving information data comprises the times of rapid acceleration, the times of rapid deceleration, the running time of low vehicle speed, the driving time, the air conditioner starting temperature, the air conditioner wind speed, the multimedia function use frequency, the use time, the Bluetooth phone use frequency and the navigation use frequency.

The further technical scheme includes that authentication is needed when communication is carried out between the vehicle and the cloud server, and meanwhile, data are encrypted and transmitted through an encryption algorithm, so that the safety of data transmission is guaranteed.

The technical scheme includes that the cloud background server performs data processing and analysis on the collected driving information data to form a data report based on the use habits of the user, and the method comprises the following steps:

step 1, data cleaning, namely, eliminating unusable data through data processing software, identifying effective usable data and obtaining neat data in a specific format;

step 2, data processing, namely extracting and converting the cleaned clean data in the step to obtain a user data set containing frequency and duration data of each function operated by a user;

step 3, data analysis, namely processing the data by using a machine learning mode, and analyzing the individual or group of the user by using a data analysis algorithm to obtain a data statistical report;

and 4, generating a user use habit report, a cluster analysis report and an intelligent recommendation report according to the generated data statistics report.

The data processing software in the step 1 comprises data processing software based on Python programming language.

The unusable data in step 1 includes lost, missed, and distorted data that cannot be used for further processing.

The step 1 of obtaining clean data in a specific format refers to data that can be directly analyzed on software, and the format of the data includes csv, xls, and xlsx.

The data analysis algorithm in the step 3 comprises data sorting, cluster analysis, automatic learning and multidimensional association rules; the data statistics report comprises a report with statistical data characteristics, including frequency, average duration and maximum duration.

And 4, generating a user use habit report, a cluster analysis report and an intelligent recommendation report according to the generated data statistics report, wherein the rule of generating the report can be customized according to actual requirements, for example, a user who likes music can be considered to listen to songs for more than 2 hours every day, and a user who likes music can be considered to listen to music for more than or equal to 70% of the time/driving time.

A further technical scheme includes that in the step 1, the format includes CSV, JSON, and partial.

The further technical scheme includes that a plurality of vehicle machine UI interfaces are designed according to the data report of the use habits of the user for the user to select, the vehicle machine UI interfaces are subjected to recombination design and then fed back to the vehicle machine through the T-BOX, the vehicle machine can recommend the UI interfaces which can be selected to the user in a popup window mode, and the user can select the UI interfaces.

The car machine recommends the UI interface that can choose to the user through the form of popup window with fixed cycle or according to the setting of user, and the user selects by oneself.

The UI design system based on big data analysis for realizing the second aim of the invention is as follows: the data transmission module is included: the system is used for transmitting data between the vehicle and the cloud background server; a data cleaning module: for identifying valid available data; a data preprocessing module: for selecting clean user information to be analyzed; a data analysis module: the data analysis system is used for analyzing the user data set to the user individuals or groups through a data analysis algorithm to obtain a data statistics report; a data report generation module: and generating a user use habit report, a cluster analysis report and an intelligent recommendation report according to the data statistics report.

The data analysis algorithm comprises data sorting, cluster analysis, automatic learning and multidimensional association rule analysis; and obtaining a data statistics report, wherein the data report comprises user behaviors and user preferences.

The technical scheme further comprises that the system further comprises a data encryption module used for encrypting the transmitted data and ensuring the security of data transmission.

The further technical scheme includes that the data preprocessing module further comprises a data extraction module and a conversion module: the method is used for selecting neat user information to be analyzed, and facilitates later data analysis.

The further technical scheme comprises that the data analysis module also comprises a data sorting module which is used for sorting the data according to a set rule; a cluster analysis module: the method is used for continuously improving the portrait precision of the user and laying a foundation for the self-learning of the next step; a self-learning module: and the computer continuously iterates and updates according to the formed data training set, so that the user behavior is predicted in advance, and the intelligence and the accuracy of recommendation are improved.

The setting rule comprises the use frequency or the use time length.

The further technical scheme includes that the system further comprises an intelligent recommendation module: and recommending a UI design interface based on big data analysis to a user.

By utilizing the method, the data is more authentic, the potential characteristic habits of different users are mined through the clustering analysis of the big data technology, the use of each user with different control habits can be better met, and the design method is also suitable for designing other terminal interfaces with data storage of the background server.

Drawings

FIG. 1 is a block diagram of a system according to the present invention;

FIG. 2 is a flow chart of data processing according to the present invention.

Detailed Description

The following detailed description is provided for the purpose of explaining the claimed embodiments of the present invention so that those skilled in the art can understand the claims. The scope of the invention is not limited to the following specific implementation configurations. It is intended that the scope of the invention be determined by those skilled in the art from the following detailed description, which includes claims that are directed to this invention.

As shown in fig. 1, in the car-in-car system of the car network, the car display realizes communication interaction with the cloud of the system through a vehicle-mounted wireless communication BOX (T-BOX), when a client operates and uses a function of the car, a system background server can take the operation of the user as a buried point, upload the buried point to the system background server for recording and storage, and analyze the frequency of the user in use for the function operation. For example, the frequency of application use of navigation, music, radio stations, Bluetooth telephones and the like by a user is counted, the preference of the user to theme styles is counted, and the data of the vehicle used by the user is recorded and stored, meanwhile, the system uploads single-time driving data of the vehicle, including the rapid acceleration times, the rapid deceleration times, the air conditioner starting time, the low vehicle speed driving time, the air conditioner starting temperature, the air conditioner wind speed, the multimedia function use frequency, the use time, the Bluetooth telephone use frequency and the navigation use frequency, to a background server through a T-BOX, and the data are obtained and the driving habits of the user are analyzed. These background data form the truest raw big data used by the client. And forming a data report of the use habit of the user through the data transmission module and the data processing system. In the driving process, the driving safety is vital, the information displayed by the instrument panel is limited, and the most common function setting is selected and integrated on the interface which is most convenient to operate of the vehicle machine, so that the driving convenience and safety are greatly improved. According to the data report, design definition of UI interface level and page layout can be guided, and intelligent recommendation schemes for different users can be given.

For example, according to the frequency statistics of the application operation usage, the application with high usage frequency is placed under the first-level interface during interface design, and the user can reach the application at one touch; the statistics of the functional operation frequency of the buttons is combined, the operation buttons which are frequently used are considered to be close to the main driving position during the layout design of the internal interface, and the main driving operation is convenient to use.

The specific implementation manner of the scheme is as follows:

the vehicle-mounted T-BOX and the vehicle-mounted T-BOX are communicated through a CAN line to realize the transmission of instructions and information, the transmission comprises key state information used by the vehicle-mounted function and user control instructions, single driving data including the times of rapid acceleration, the times of rapid deceleration, the time of air conditioner starting, the time of low-speed driving, the time of air conditioner starting, the temperature of air conditioner starting, the air conditioner wind speed, the multimedia function use frequency, the use time, the Bluetooth phone use frequency and the navigation use frequency are simultaneously acquired, and the vehicle-mounted T-BOX is in network communication with the cloud background server through LTE or 5G to realize the information interaction and control of the vehicle-mounted function and the background server.

In order to ensure safe communication, communication authentication is carried out between the vehicle-mounted T-BOX and the cloud server, and meanwhile, communication information is encrypted by adopting an encryption algorithm, so that the safety of data transmission is ensured.

The data transmission module is used for data transmission between the cloud background server and the data processing system. In the data processing system, the data recorded and stored in the background server comprises the most original data (including the frequency of application use and function button operation use by the user) used by each function, and after data preprocessing and data mining are carried out, data extraction and conversion are carried out by combining a clustering algorithm, use frequency sequencing and the like, so that a user data analysis report containing the characteristics of behavior, preference and the like of the user can be formed.

The preprocessing comprises that the vehicle-mounted computer records data into formats such as CSV, JSON, partial and the like in advance; the data mining comprises the steps that a data engineer processes the data in a machine learning mode, and the obtained data are not original data any more but driving habit reports of users; the clustering algorithm comprises K-Means and DBSCAN algorithms in unsupervised learning, driving habit data of one user, including a data set in a CSV format, is obtained from original data of the user, and if the user is a majority of users, different users can be divided into different user groups; the data extraction and conversion relates to an algorithm principle in machine learning. For example, using the K-means algorithm, a data set for a day is first selected for most users and their respective center points are randomly initialized. The center point is the same length position as each data point vector. This requires that the number of classes (i.e., the number of center points) be known in advance, the distance from each data point to the center point be calculated, and the class into which the data point is closest to which center point. The center point in each class is calculated as the new center point. The above steps are repeated until the center of each class does not change much after each iteration. It is also possible to randomly initialize the center point multiple times and then select the one that has the best run result. After obtaining the processed data, we may see the average depth of braking, the distance of braking, etc. of users of different ages and sexes in one day.

The data processing flow comprises the following steps: background raw data extraction- > data cleaning- > data processing- > data analysis- > data reporting, as shown in fig. 2.

The data recorded and stored in the background server comprises the most original data of each function used by a user, including a plurality of information of weather, navigation, multimedia control, Bluetooth telephone, air conditioner control, vehicle conditions, brand special areas and the like, and the related information of the user driving the vehicle, including the times of emergency acceleration, the times of emergency deceleration, the air conditioner starting time, the low-speed driving time, the air conditioner starting temperature, the air conditioner air speed, the multimedia function use frequency, the use time, the Bluetooth telephone use frequency and the navigation use frequency.

Because data loss, missing recording, distortion and the like may exist in the data recording process, such data needs to be removed and valid data is reserved, and therefore the data needs to be checked and filtered. And then, data extraction and conversion are carried out by combining a clustering algorithm, a use frequency sorting and the like, and data sorting, clustering analysis, unsupervised learning algorithm of machine learning, multidimensional association rule analysis and the like are carried out on the data to be analyzed according to actual needs to form a data report. And discovering the potential habits of the user. The multidimensional association rule analysis refers to the association between the collected data and various types of things that may happen. For example, a person who prefers hard braking may prefer a bluetooth phone.

Because the information displayed on each layer of pages of the car machine is limited, the functions required by common use or regulations need to be selected and placed on the first page or the frequently displayed pages. And the data report is a report of the use frequency of each function of the user, so that the user can identify which functions are frequently used and which functions are not frequently used. Firstly, functions required by regulations and needing to be provided with keys are identified to be placed on a home page or a frequently displayed interface, such as front defrosting and rear defrosting of an air conditioner control function (when no other control panel exists, the frequently displayed interface needs to be defined on a vehicle machine page), and then, the use habits of various types of users can be mined, the frequently used functions of the users can be found, and page definition can be guided. Meanwhile, a plurality of users with different habits can be formed to guide the user to define the vehicle interface into a plurality of interfaces with different styles for the user to select, and on the other hand, a data warehouse can be formed simultaneously for subsequent use.

Automatic learning techniques are utilized simultaneously in the data processing system. The technology integrates the intelligent learning of the user on the click frequency of different functions into a user personalized use information list, and meanwhile, the rule information is generated according to the use condition of the user on the different functions. And a multidimensional association algorithm is adopted to intelligently recommend the user, so that an intelligent recommendation scheme is formed, and the user experience is improved.

When the vehicle UI is designed, a user personalized recommendation system is designed, and the use habit of the current driver is predicted according to the existing user data and the partial habit of the current user. The intelligent recommendation scheme is formed based on background data, after the system conducts recombination design on a vehicle machine interface, the vehicle machine interface is fed back to the vehicle machine through the T-BOX, the vehicle machine can pop up through a popup window to inquire whether a user selects the interface scheme intelligently recommended by the system, and the user selects the interface scheme by himself.

And inquiring whether the user accepts the recommended UI interface or not through a popup window or voice broadcast according to the result of the user data obtained by analysis and the driving habit of the user. The driving habits of each user are different, and the information popped up is also different. For example, the user is reminded of being used to sudden braking, driving is not stable, fatigue driving is easy, and the like, so that the user is helped to drive safely, and man-machine interaction is realized. The query may be made on a periodic basis, such as once a month or once a half month.

The invention is suitable for all-level automatic driving vehicles, and can be power battery vehicles or fuel vehicles.