KR102298395B1 - System and Method for User Behavior Analysis and Agent for Collection of Event - Google Patents

Abstract

The user behavior analysis system according to an embodiment of the present technology is provided in the user terminal, and as the user terminal is started, the program by the user from the execution time to the end time of the program including the operating system and the application executed in the user terminal An event collection agent that collects all operation information for the user over time, generates event information and transmits it through a communication network, and an event analysis server that is connected to one or more user terminals through a communication network and collects event information from the event collection agent It may be configured to include

Description

System and Method for User Behavior Analysis and Agent for Collection of Events

The present technology relates to a technology for collecting data to configure a database and analyzing data based thereon, and more particularly, to a user behavior analysis system and method, and an event collection agent for the same.

With the spread and spread of personal computing devices, various services are being developed through them, and users use various applications executed on the computing devices for work or social activities.

A technique for deriving characteristics by analyzing various behaviors of a specific user or users included in a specific group may be referred to as a user behavior analysis (UBA) technique.

Recently, a technology for analyzing a log of a computing device has been developed for task automation or detection of anomalies on a security network. However, the current behavior analysis technology only collects the manipulation history of previously generated information in fragments, and there is a limit to providing personalized services or improving the service level through this.

Accordingly, there is a need for a method capable of more accurately analyzing various actions of a user through a computing device.

An embodiment of the present technology collects and analyzes all actions of a user through a computing device, for example, input actions through a keyboard and a mouse, and provides feedback helpful for a user's input actions to the computing device. It is possible to provide an analysis system and method, and an event collection agent for the same.

The user behavior analysis system according to an embodiment of the present technology is provided in the user terminal, and as the user terminal is activated, the user an event collection agent that collects all operation information for the program over time, generates event information, and transmits it through a communication network; and an event analysis server that is connected to one or more user terminals and the communication network and collects the event information from the event collection agent.

A user behavior analysis method according to an embodiment of the present technology includes an event collection agent provided in a user terminal to detect a process state change, a key input event, a mouse input event, and a screen capture event of the user terminal to generate event information; A user behavior analysis method in a user behavior analysis system including an event analysis server for receiving the event information from the event collection agent, the method comprising: executing the event collection agent as the user terminal is activated; generating event information by collecting all operation information for the program by the user over time from the execution time to the end time of the program including the operating system and the application executed in the user terminal; and constructing, by the event analysis server, the event information as a database; and analyzing the event information based on the database.

The event collection agent according to an embodiment of the present technology is provided in the user terminal, and the process state change of the user terminal from the execution time to the end time of the program including the operating system and the application executed in the user terminal, an event detection unit for detecting a key input event, a mouse input event, and a screen capture event for the program; an event processing unit configured to process the event information detected by the event sensing unit into a set format; and an event transmission manager configured to transmit the event information generated by the event processing unit to a destination designated for each type of event.

According to the present technology, it is possible to collect all input actions of a user to a computing device, and to build and analyze a database. Accordingly, it is possible to allow the user to easily operate the computing device by helping the user to a difficult (immature) operation.

Furthermore, since it is possible to analyze which applications are used by several users and how and what difficulties they experience in input, each application creator can improve the application so that the user's input behavior is easy.

1 is a block diagram of a user behavior analysis system according to an embodiment.
2 is a block diagram of a user terminal according to an embodiment.
3 is a block diagram of an event collection agent according to an embodiment.
4 is a configuration diagram of an event analysis server according to an embodiment.
5 is a flowchart illustrating a method for collecting user actions according to an embodiment.
6 is a flowchart illustrating a user behavior analysis method according to an embodiment.

Hereinafter, embodiments of the present technology will be described in more detail with reference to the accompanying drawings.

1 is a block diagram of a user behavior analysis system according to an embodiment.

Referring to FIG. 1 , a user behavior analysis system 10 according to an embodiment may include one or a plurality of user terminals 100 and an event analysis server 300 that are interconnected through a communication network 200 .

The user terminal 100 may include a terminal 101 such as a desktop computer, or various computing devices capable of wired or wireless communication such as a laptop computer 103 .

The user terminal 100 may include an event collection agent 160 .

The event collection agent 160 may be automatically or manually executed after booting of the user terminal 100 to collect all actions of the user with respect to the user terminal 100 and transmit it to the event analysis server 300 .

The event analysis server 300 may be configured to receive, store, and analyze event collection information received from the user terminal 100 .

In one embodiment, the event collection agent 160 collects and stores the status events of all processes executed in the user terminal 100 and all manipulation action events of the user for the running process, and a preset cycle or a preset condition. When this is satisfied, the collected event may be transmitted to the event analysis server 300 .

For example, the event collection agent 160 includes all the operation information for the program by the user from the execution time to the end time of the program including the operating system (OS) and the application executed in the user terminal 100, that is, the keyboard history. , all input information such as mouse input and display configuration change history can be collected over time.

For example, all input information related to OS operation, keyboard/mouse input information in all processes of executing “application A” and then executing “application B”, and display configuration change history, etc. may all be collected.

The event analysis server 300 may database the event information transmitted from the user terminal 100 according to a set index and analyze the event information for each index.

In addition to collecting only the results after a specific application program is executed and manipulated as event data, it is possible to accurately analyze user behavior as all manipulation events for the entire process from execution to termination can be collected.

2 is a block diagram of a user terminal according to an embodiment.

Referring to FIG. 2 , the user terminal 100 according to an embodiment includes a controller 110 , a main memory device 120 , a secondary memory device 130 , an IO device 140 , a communication module 150 , and an event collection agent. (160).

The controller 110 controls the overall operation of the user terminal 100 , and may be configured in a form in which software and hardware for executing the same are combined. In an embodiment, the controller 100 may be configured to run an operating system (OS) to manage each hardware component.

The main memory device 120 may temporarily store a program and data to be executed by the controller 110 , or a result processed by the controller 110 .

The auxiliary storage device 130 may be a disk for storing programs and data to be executed by the controller 110 to be maintained. Depending on implementation, the user terminal 100 may be implemented by omitting the auxiliary storage device 130 .

The IO device 130 may include an input device and an output device. The input device may include at least one of a keyboard, a mouse, a joystick, a touchpad, an electronic pen, and a smart touchpad having a display unit. The output device may include at least one of a monitor, a speaker, a printer, and an image projector.

The communication module 150 may provide an environment in which the user terminal 100 can communicate with other computing devices through the communication network 200 .

The event collection agent 160 monitors which process is executed in the user terminal 100, and from the start of the user terminal 100 until the end, and/or from the execution of a specific application to the end time of the user for the corresponding application. It may be configured to collect all the information of the captured screen according to the operation history and / or set conditions through the IO device 140 .

3 is a block diagram of an event collection agent according to an embodiment.

Referring to FIG. 3 , the event collection agent 160 may include an event detection unit 161 , an event processing unit 163 , and an event transmission manager 165 .

The event detection unit 161 may be configured to detect a change state of the user terminal 100 .

In an embodiment, the event detector 161 may include a process event detector 1611 , a key event detector 1613 , a mouse event detector 1615 , and a screen capture event detector 1617 .

The process event detection unit 1611 may be configured to detect a change in a process executed in the user terminal 100 under the control of the controller 110 .

In an embodiment, the process event detection unit 1611 may generate process event information by collecting status information of processes that are executed, changed, and terminated over time. In an embodiment, the process event detecting unit 1611 may additionally detect the type, version, and user interface configuration information of the executed application in response to a process change as the specific application is executed.

The key event detection unit 1613 may be configured to detect an event that occurs when a key is input through the controller 110 .

In an embodiment, the key event detecting unit 1613 may collect individual key input information input through the keyboard according to time sequence. Here, the key input information to be collected is a character key, a number key, a space key, an enter key, a function key, a shift key, a language conversion key, a backspace key, a delete key, an arrow key ) may be input information for all keys provided on the keyboard. The key event detecting unit 1613 may generate key event information by adding an input time to each key input information. The input time for each key input information may be stored in units of tenths of a second, preferably up to a time of less than or equal to tenths of a second.

The mouse event detection unit 1615 may be configured to detect an event that occurs when a mouse is input through the controller 110 .

In an embodiment, the mouse event detecting unit 1615 may collect input information through a mouse, for example, a right click, a left click, and a click position. In one embodiment, the click position may be an absolute position on the display device or a position in a running application, and for each input information through the mouse, in tenths of a second, preferably in hundredths. Mouse event information can be generated by adding an input time up to a time unit of seconds or less.

The screen capture event detection unit 1617 may be configured to capture the display screen of the user terminal 100 when a defined key or mouse event occurs. In an embodiment, when a predefined capture event occurs through a keyboard or mouse, the screen capture event detection unit 1617 may capture the display screen at the time the capture event occurs.

In one embodiment, when a specific menu is clicked with a mouse while using a specific application, the display screen at the clicked moment may be captured or screen information within the corresponding application window may be captured, and through this, what the user clicked may be analyzed. have.

The event processing unit 163 may be configured to process the event information sensed as an event is detected by the event detection unit 161 into a set format. In an embodiment, the event information may be processed in a JSON (JavaScript Object Notation) format, but is not limited thereto.

The event processing unit 163 may include a process event processing unit 1631 , a key event processing unit 1633 , a mouse event processing unit 1635 , and a screen capture event processing unit 1637 .

The process event processing unit 1631 may be configured to process the process change event detected by the process event detection unit 1611 .

The key event processing unit 1633 may process the key input signal detected by the key event detection unit 1613 , and the mouse event processing unit 1635 processes the mouse input history detected by the mouse event detection unit 1615 . can do.

The screen capture event processing unit 1637 may process, for example, resize, or convert the screen captured by the screen capture event detection unit 1617 into a specific format.

The event transmission manager 165 may manage transmission destinations for each type of event, and transmit the event information processed by the event processing unit 163 to the transmission destination, for example, the event analysis server 300 through the communication network 200 . can

As all input information through the user terminal 100 is collected, an input device operation pattern for each specific user can be analyzed, and feedback data can be provided so that the user can more easily operate the terminal 100 . have.

In addition, input signals from a plurality of user terminals may be collected and analyzed according to the type of program. because of this

Since it is possible to analyze which applications how and how many users use which applications and what difficulties they experience in input, it is possible to help users to input easily, and to provide data on application development and improvement.

In an embodiment, the event analysis server 300 may analyze a word having difficulty (immaturity) in keyboard input based on event information of a specific user. In addition, when such words are input or mistyped, a notification signal is output, or input training of the corresponding word, automatic correction, auto-complete, expert advice, etc. are provided so that the user can more easily enter the word having difficulty in input. let it be

4 is a configuration diagram of an event analysis server according to an embodiment.

Referring to FIG. 4 , the event analysis server 300 according to an embodiment includes a controller 310 , a memory 320 , a disk 330 , an IO interface 340 , a communication module 350 , and an event collection unit 360 . ) and an event analysis unit 370 .

The controller 310 may be implemented in the form of a combination of a program necessary to operate the event analysis server 300 and hardware capable of executing the same, and may be configured to control the overall operation of the event collection server 300 .

The memory 320 may temporarily store a program and data to be executed by the controller 310 , or a result processed by the controller 110 , or may temporarily store event information for each user terminal 100 transmitted through the communication network 200 . .

The disk 330 may receive and store event information from the memory 320 .

The IO interface 340 may provide an input/output environment in which an administrator can access, manage, and control the event analysis server 300 .

The communication module 350 provides an environment in which the event analysis server 300 can communicate with the user terminals 100 through the communication network 200 .

The event collector 360 may store event information transmitted from the user terminals 100 in the database 3600 based on a set index. In an embodiment, the database 3600 is a type of Non-Relational Operation Database SQL (NoSQL) DB and may be MongoDB using a JSON format data structure, but is not limited thereto.

In an embodiment, the event collection unit 360 may store event information for each user in the database 3600 using the name of the application executed in the user terminal 100 as an index. In another embodiment, the event collection unit 360 may use the user ID as an index to database event information for each application.

The event analyzer 370 may be configured to extract and classify features from databased event information according to an index.

In one embodiment, the event analyzer 370 divides a word unit input string that is difficult for a specific user to use when a specific application uses a specific application or divides a frequently used word unit input string to facilitate input thereof. You can provide feedback information.

In one embodiment, in order to analyze the event information, the event analyzer 370 connects the event sets collected for each unit time or for a time until a preset event occurrence time according to a continuous collection time, and the connected event set A unit event can be detected from In addition, characteristics of each detected unit event may be analyzed. In one embodiment, an event set may be collected and transmitted when a specific event occurs.

Here, the event set for each unit time may mean a set of event information collected and transmitted for a set time. The linking of the event sets for each unit time may be an operation of linking the event sets of which the collection time is continuous. In addition, detecting a unit event may be a task of classifying a linked event set into a set unit based on a set identifier.

Taking the case of analyzing a key event as an example, the event analyzer 370 extracts and connects as an event set with a continuous collection time from the key input set collected for each unit time, and converts the input string in word units from the connected input set as a unit. Events are classified and detected, and characteristics of each input string of the divided word unit can be analyzed.

Here, the input set for each unit time may mean a set of key inputs collected for a set time. The linking of the input sets for each unit time may be an operation of linking the input sets in which the collection time is continuous. Separating the input string in units of words may be an operation of dividing the linked input set into units of a set, for example, a unit of a word or a word, based on a set delimiter. In one embodiment, the delimiter for separating the input string in units of words may include at least one of an enter key input event, a space bar input event, a comma (,) or punctuation (.) input event, and a language conversion key input event. However, the present invention is not limited thereto.

In an embodiment, the event analyzer 370 may analyze an input string in units of words that the user has difficulty in inputting from the key event information.

For example, in a document editing application, the user tried to type the string 'the lowest value of input difficulty' (the space bar is entered after the value), but an error occurred in the input of 'value' during typing, and the user entered the space bar for 'A'. Then, a case in which the space bar and 'a' are deleted with backspace from the back and input as 'value' (the space bar after the value is not input) will be described.

In this case, the event analysis unit 370 receives from the user terminal 100, 'ㅇㅣㅇㅇㅇㅇㅇ ㅇTTㅁㅇㅡㅣ_cㅗㅣjㅓahav_[bs][bs]aavo' (_ is a space). Bar, [bs] can receive the input set (backspace).

According to a time interval at which the user terminal 100 transmits event information to the event analysis server 300 , the user's input set may be divided into two or more and transmitted. For example, the input set includes a first input set for each unit time 'ㅇㅣㅇㅇㅕㄱ_ㅇㅓㄹㅕ' and a second input set for each unit time 'ㅇTTㅁㅇㅡㅣ_chㅗㅣjㅓahav_[bs][bs] It can be divided into 'A'.

The event analyzer 370 checks whether a separator set at the beginning or the end of the input set for each unit time is included in order to connect the input sets for each unit time in which the collection time is continuous, and if the separator is not included, the unit for the collection time is continuous Hourly input sets can be linked.

That is, no space bar, enter key, comma, or punctuation input information was collected at the end of the first input set for each unit time 'ㅇㅣㅇㅇㅇㅇㅓㄹㅕ', and the second input set for each unit time 'ㅇTTㅁㅇㅡㅣ_ch Since the space bar, enter key, comma, or punctuation input information at the beginning of ㅗㅣj ㅓaa _[bs][bs]aa o ' was not collected, 'ㅇㅓㅕ' and 'ㅇTTㅁㅇㅡㅣ' are not connected. It can be judged as what should be the target.

After the input sets are connected, the input string in units of words may be detected based on the delimiter included in the connected input set. In this case, it is also possible to ignore the space bar, the enter key, commas, or punctuation which are determined to be deleted during the editing process.

Therefore, from the result of connecting the first and second input sets for each unit time,

(1)'ㅇㅣㅇㅇㅇㅇㅇㅇ_'

(2)'ㅇㅓㅕㅇTTㅁㅇㅡㅣ_'

(3) A three-word unit input string of 'cㅗㅣjㅓaaf_[bs][bs]aaaw' may be detected.

Among these, the case of the third word unit input string will be described. When key input is performed without error, 9 key input events occur, but 15 key input events occur due to erroneous input. Accordingly, it can be seen that it is highly likely to take a longer time than in the case of normal input.

Here, the input time required for the input string for a specific word unit is, for example, the time (for example, 15:33:01:07) appended to the last key input information to the time (for example, 15:33:01:07) appended to the first key input information of the input string. For example, it can be identified as the time interval (1 second 10) up to 15:33:02:17).

The event analysis unit 370 detects a specific document, an entire document, or a word unit input string included in a database collected for a certain period of time, and includes the frequency of use of each word unit input string, the number of occurrences of key input events, and the input required time. According to the characteristics of the word unit input string can be clustered.

And, based on the clustering result, an input string that the user has difficulty with may be classified.

In an embodiment, the event analyzer 370 may classify an input event based on a method selected from among various clustering methods, for example, K-Means, hierarchical cohesive clustering, and non-hierarchical clustering.

Therefore, by classifying the input string of words that are frequently used but having difficulty in input, a notification signal is output when the user inputs such a word, or input training (practice) of the corresponding word, automatic correction, automatic completion, expert advice etc. may be provided so that the user can more easily input a word having difficulty in input.

In addition, when input events from a plurality of user terminals are collected, it is possible to analyze which applications and how many users use which applications and what difficulties they experience in input, thus helping users to easily input actions, while We can provide data on development and improvement, etc.

5 is a flowchart illustrating a method for collecting user actions according to an embodiment.

Referring to FIG. 5 , as the user terminal 100 is booted and started ( S101 ), the event collection agent 160 may be executed ( S103 ).

The event collection agent 160 starts monitoring to detect whether a process change, an input event, or a screen capture event of the user terminal 100 occurs (S105), and when an event occurs, event information can be collected and stored for each event. (S107). For example, the event collection agent 160 may collect and store a process change event, a key event, a mouse event, and a screen capture event, respectively.

In one embodiment, the event collection step ( S107 ) includes all events that occur after the user terminal 100 is booted and started until it is terminated, or all events that occur after a specific process or a specific application is executed until it is terminated. , for example, may be a step of collecting all input signals of the user.

The event collection agent 160 may transmit the collected event to the event analysis server 300 when a set period arrives or a set condition is satisfied (S109).

6 is a flowchart illustrating a user behavior analysis method according to an embodiment.

As event information is transmitted from the event collection agent 160 of the user terminal 100, the event analysis server 300 connects the event sets of consecutive collection times from the event sets collected for each unit time, and from the connected event sets A unit event can be detected. In addition, characteristics of each detected unit event may be analyzed.

An example of a process of analyzing a user behavior based on a key event will be described with reference to FIG. 6 .

Referring to FIG. 6 , the event analysis server 300 may extract (S201) and connect (S203) a set of key inputs collected for each unit time, for example, a set of key inputs collected for a set time. Here, the linking of the input sets for each unit time may be an operation of linking the input sets of which the collection time is continuous.

The event analysis server 300 may extract a delimiter included in the linked input set connected in step S203 (S205). In one embodiment, the delimiter may include at least one of an enter key input event, a space bar input event, a comma or punctuation input event, and a language conversion key input event, but is not limited thereto.

The event analysis server 300 may classify the input string in units of words from the linked input set based on the delimiter extracted in step S205 (S207). In an embodiment, the step S207 of classifying the input string in units of words may be a step of classifying the linked input set in units of words or words.

Thereafter, the event analysis server 300 may cluster the word unit input string according to the set characteristics ( S209 ).

In one embodiment, the step of analyzing the input string in units of words ( S209 ) includes clustering the input strings in units of words according to characteristics including the frequency of use of the input strings in units of words, the number of keystrokes, and the input time required for each word unit, and adding the results of the clustering. The method may include classifying an input string that the user has difficulty in inputting based on.

In an embodiment, the event analyzer 370 may classify an input event based on a method selected from among various clustering methods, for example, K-Means, hierarchical cohesive clustering, and non-hierarchical clustering.

Therefore, by classifying the input string of words that are frequently used but having difficulty in input, a notification signal is output when the user inputs such a word, or input training (practice) of the corresponding word, automatic correction, automatic completion, expert advice By providing feedback information such as S211, it is possible to help the user to easily input a word having difficulty in input.

In addition, when several users analyze how and which applications are used and what difficulties they experience in input, data necessary for input method or user interface (UI) improvement may be provided.

Although the key input event has been described above as an example, it goes without saying that the mouse input event can also be converted into a database and analyzed in a similar manner. In addition, it is self-evident that even when a key input event and a mouse input event are mixed, it is possible to list the input events in chronological order or to list and classify the input events by input interface and in the chronological order.

As such, those skilled in the art to which the present invention pertains will understand that the present invention may be embodied in other specific forms without changing the technical spirit or essential characteristics thereof. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. The scope of the present invention is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

10: User behavior analysis system
100: user terminal
200: communication network
300: event analysis server
160: event collection agent

Claims (24)

It is provided in the user terminal, and as the user terminal is activated, all operation information for the program by the user from the execution time to the end time of the program including the operating system and the application executed in the user terminal is collected over time. an event collection agent that generates event information and transmits it through a communication network; and
an event analysis server connected to one or more user terminals and the communication network to collect the event information from the event collection agent;
including,
The event analysis server receives, from the event collection agent, input signals through all keys included in the keyboard connected to the user terminal as key input event information according to the input time,
At least one key input set having a continuous collection time is connected to a key input set, which is a set of key input signals collected during a preset unit time or a time until a preset event occurrence time, and a word unit from the connected input set A user behavior analysis system configured to classify and classify the input string of The method of claim 1,
The event collection agent is a user configured to collect process change event information executed in the user terminal, key input event information by the keyboard, mouse input event information by a mouse connected to the user terminal, and screen capture event information Behavior analysis system. delete 3. The method of claim 2,
The event collection agent is configured to collect all input information including left-click, right-click, and click position of the mouse as the mouse input event information according to an input time. 3. The method of claim 2,
The event collection agent is configured to collect screen capture event information by capturing a screen of a display connected to the user terminal according to the occurrence of a preset specific event by the keyboard or the mouse or a set time period system. The method of claim 1,
The event analysis server collects an event set that is a set of event information collected during a preset unit time or a time up to a preset event occurrence time, and connects at least one event set having a continuous collection time, A user behavior analysis system configured to detect and classify a unit event by classifying the connected event set according to a set identifier. delete The method of claim 1,
The delimiter is a user behavior analysis system configured to include at least one of an enter key input event, a space bar input event, a comma input event, a punctuation input event, and a language conversion key input event. The method of claim 1,
The event analysis server is configured to classify the input string of the word unit according to characteristics including the frequency of use, the number of times of key input, and the input required time for each input string of the word unit. 10. The method of claim 9
The event analysis server detects a feedback target word unit input string based on the frequency of use for each word unit input string, the number of key input, and the input required time, and transmits feedback information for input of the feedback target word unit input string to the user terminal. A user behavior analysis system configured to provide. delete delete delete delete delete An event collection agent provided in the user terminal to detect a process state change, a key input event, a mouse input event, and a screen capture event of the user terminal to generate event information, and an event analysis for receiving the event information from the event collection agent As a user behavior analysis method in a user behavior analysis system including a server,
executing the event collection agent as the user terminal is activated;
generating event information by collecting all operation information for the program by the user over time from the execution time to the end time of the program including the operating system and the application executed in the user terminal; and
building the event analysis server into a database by receiving the event information; and
analyzing the event information based on the database;
including,
The analyzing may include: receiving, by the event analysis server, input signals through all keys included in a keyboard connected to the user terminal from the event collection agent as key input event information according to an input time;
connecting at least one key input set having a continuous collection time to a key input set that is a set of key input signals collected during a set unit time or a time up to a preset event occurrence time;
classifying an input string in units of words from the connected input set according to a set delimiter; and
classifying the input string of the divided word unit;
User behavior analysis method configured to include. delete 17. The method of claim 16,
The generating of the event information includes, by the event collection agent, collecting all input information including left clicks, right clicks, and click positions of a mouse connected to the user terminal as the mouse input event according to an input time. A method of analyzing user behavior that is configured to: 17. The method of claim 16,
In the step of generating the event information, the event collection agent captures the screen of the display connected to the user terminal according to the occurrence of a preset specific event by the keyboard or the mouse or a set time period to capture the screen capture event. A user behavior analysis method configured to include collecting. 17. The method of claim 16,
The analyzing may include: collecting, by the event analysis server, an event set, which is a set of event information collected during a preset unit time or a time up to a preset event occurrence time, from the event collection agent;
concatenating at least one set of events having consecutive collection times;
detecting a unit event by classifying the connected event set according to a set identifier;
classifying the detected unit event;
User behavior analysis method configured to include. delete 17. The method of claim 16,
The delimiter is a user behavior analysis method configured to include at least one of an enter key input event, a space bar input event, a comma input event, a punctuation input event, and a language conversion key input event. 17. The method of claim 16,
In the classifying step, the event analysis server is configured to classify the input string of the word unit according to characteristics including the frequency of use for each input string of the word unit, the number of times of key input, and the input required time. 24. The method of claim 23
The classifying may include: detecting, by the event analysis server, a feedback target word unit input string based on a frequency of use for each word unit input string, the number of key input, and an input required time; and
providing feedback information for input of the feedback target word unit input string to the user terminal;
User behavior analysis method configured to include.

Images