CN102663347B - Students ' Learning behavior gather and analysis system and method thereof - Google Patents

Abstract

The invention belongs to the technical field of wireless communication, and specifically relates to a system and method for collecting and analyzing student learning behaviors. The system includes an external computer and a real-time automatic recording device for learning behavior data connected to the external computer; the real-time automatic recording device for learning behavior data Can collect human body acceleration information, and transmit the human body acceleration information to the computer for analysis; learning behavior data real-time automatic recording device includes: action collection unit, clock unit, storage unit, communication unit, intelligent analysis and processing unit, intelligent analysis and processing unit Through its input and output ports, it is respectively connected with the action acquisition unit, storage unit, clock unit and communication unit, and coordinates the actions among them, which avoids the inconvenience and discomfort of wearing acquisition devices on multiple parts of the human body, and through Analyze students' personality traits and learning behavior habits, provide correct guidance for students' healthy development, and avoid partial subjects.

Description

Student Learning Behavior Acquisition and Analysis System and Method

technical field

The invention belongs to the technical field of wireless communication, in particular to a system and method for collecting and analyzing students' learning behavior based on an acceleration sensor.

Background technique

Students are the main body of teaching activities, and classroom teaching is the most important link in the learning process of students. How well students play the main role in teaching activities directly affects the quality of teaching. Monitoring students' learning behavior is the most effective and direct way to analyze students' learning process. Therefore, it is the key to improve the quality of talents to study and analyze students' learning status by collecting and recording students' learning behaviors, and to solve them in a targeted manner. Existing human behavior feature recognition devices and methods are mainly used in medical fields, security fields, sports fields, and the like. For example, in the field of medicine, the symptoms of patients can be understood by analyzing the characteristic behaviors of patients; in the field of safety, it can be judged whether there is danger by analyzing the abnormal behavior of the human body in highly sensitive areas; in the field of sports, various pedometers can count the steps of the human body To analyze the amount of exercise of the human body, etc. However, none of these disclosed devices and methods can collect and analyze students' learning behaviors.

According to the patent query of the Retrieval Center of the National Patent Office, there is an acceleration sensor-based human motion recognition system and method for human motion recognition. The application number is: 200910184850.0, and the publication number is: CN101694693. It is mainly used in the real-time recognition of human body movements. By wearing a sub-device on the limbs and neck of the human body, the acceleration information of the human body is collected and transmitted to the computer for analysis, thereby recognizing human body movements. However, it is mainly used to identify the types of actions currently occurring on the human body, and it cannot record and analyze students' learning behaviors for a long time. Therefore, it is not its function to collect and analyze students' learning behaviors, and then to obtain students' learning conditions; moreover, it needs People wear acquisition devices on their necks, upper limbs and lower limbs, which is not suitable for long-term wear, especially for students, and is inconvenient to use; at the same time, the analysis process requires combining data from multiple acquisition devices, which is cumbersome and difficult accomplish. This undoubtedly limits the scope of application and occasions of the invention based on the acceleration sensor-based human motion recognition system and method.

Contents of the invention

A technical problem to be solved by the present invention is to overcome the shortcomings of the above-mentioned accelerometer-based human motion recognition system and method, and provide a method that can automatically analyze and record learning behaviors, help parents and teachers to grasp students' learning situations in time, and analyze students' learning behaviors. Performance in the classroom, does not affect the health of students, easy to carry student learning behavior collection and analysis system.

Another technical problem to be solved by the present invention is to provide a method for analyzing student behavior using a student learning behavior collection and analysis system.

The technical solution adopted to solve the above technical problems is: comprise an external computer and a real-time automatic recording device for learning behavior data connected with the external computer; the real-time automatic recording device for learning behavior data of the present invention can collect human body acceleration information, and transmit the human body acceleration information to computer for analysis.

The real-time automatic recording device for learning behavior data of the present invention includes: an action acquisition unit for collecting human body acceleration information and outputting an action signal, and the action acquisition unit is an acceleration sensor; a clock unit provides time information for the entire device and provides an acceleration signal. The behavior data collected by the sensor provides a time stamp; the storage unit is used to store the system configuration information, the learning behavior classification feature library, and the learning behavior feature analysis results; the communication unit is connected to an external computer and is used to combine intelligent analysis and processing with the processing unit The results are transmitted to the external computer, and receive the instructions sent by the external computer to realize the setting of the configuration information and the updating and upgrading of the learning behavior classification library; the intelligent analysis and processing unit communicates with the action acquisition unit, storage unit, The clock unit and the communication unit are connected and coordinate the actions between them.

The system configuration information of the present invention includes student's name, student number, curriculum schedule, class start and end time of each class, device automatic opening time and automatic closing time; learning behavior classification library includes running, jumping, walking, shaking under initial condition , standing up, sitting down, static behaviors; the communication unit of the present invention is connected with an external computer through a USB interface, bluetooth, wifi, GPRS, Zigbee.

The present invention also provides a method for using the above-mentioned student learning behavior collection and analysis system, which consists of the following steps:

1) Start the system, judge whether to connect to an external computer, if yes, proceed to step 2), if not, proceed to step 3).

2) System configuration, set student name, student number, class schedule, system automatic opening time and automatic closing time, and synchronize system time.

3) Initialize the system and clear the data analysis flag.

4) Start the data analysis process.

5) Set the data acquisition interrupt processing program.

6) Judging whether the system is finished, if it is finished, exit the system, otherwise continue to judge.

The method of the present invention also includes step 7) an external analysis program, which sends the stored data records to an external computer for further analysis, and the analysis process is combined with the curriculum arrangement to analyze different courses and breaks.

Above-mentioned step 7) specifically is:

7.1) Synchronize the system configuration information and time information to the real-time automatic recording device for learning behavior data.

7.2) After the data records are obtained, the data records are grouped according to the curriculum and the start and end time of each class, and the data records of the same type of courses are grouped into one group.

7.3) The data records after grouping are stored in the computer database;

7.4) Analyze each set of data with intelligent data processing methods for courses;

7.5) Store the analysis results in the database and end the program.

Step 4 of the present invention) specifically is:

4.1) Determine whether the data analysis flag is set, if it is set, go to step 4.2), otherwise, repeat 4.1).

4.2) Divide the data into frames according to the time tags in the data units, and calculate the characteristics of each data frame.

4.3) Compare the frame sequence features with the classification feature library in the storage unit to determine the action category.

4.4) After the time stamp is added, a data record is formed and stored in the storage unit.

4.5) Clear the data analysis flag and return to step 4.1).

The characteristics of the data frame in step 4.2) of the present invention refer to action frequency, intensity, direction and duration. The action category in step 4.3) of the present invention refers to running, jumping, walking, shaking, standing up, sitting down and standing still. The data record in step 4.4) of the present invention includes action time, action category, action direction, average intensity, frequency and duration.

Step 5 of the present invention) specifically is:

5.1) Read clock information.

5.2) Obtain the sensor data, add a time stamp to form a data unit, and store it in the current data collection buffer where the pointer points to the storage location.

5.3) Determine whether the buffer is full, if the buffer is full, proceed to steps 5.4) to 5.6); otherwise, proceed to step 5.7).

5.4) Copy the data from the data acquisition buffer to the data analysis buffer.

5.5) Point the data collection buffer pointer to the head of the buffer.

5.6) Set the data analysis flag and end the program.

5.7) Point the data acquisition buffer pointer to the next storage location, and end the program.

The data unit in step 5.2) of the present invention includes time stamp, x acceleration, y acceleration and z acceleration.

The present invention has the following advantages:

1. The present invention fixes the learning behavior data real-time automatic recording device on the waist of the human body, automatically obtains the human body behavior data through the acceleration sensor in the acquisition unit, and realizes the classification of the learning behavior through the data analysis process, and automatically reports the analysis results Stored in the memory according to the category, avoiding the inconvenience and discomfort of wearing acquisition devices on multiple parts of the human body, and by extracting multiple feature values and comparing them with the feature library, accurate analysis results can be obtained, avoiding the need for multiple parts of the human body. Acceleration data analysis of the site results in difficulties in data combination, heavy workload and other defects.

2. The system of the present invention adopts an interrupt mode to process data collection. When the waist produces motion, that is, when the acceleration sensor has data output, an interruption is induced, and the CPU of the intelligent analysis and processing unit automatically starts data collection after receiving the interrupt signal, so as to achieve power saving. Effect.

3. The system of the present invention is automatically opened and closed according to the set opening and closing time, and can also be opened and closed manually, which is convenient to use; and the learning behavior classification feature library stored in the storage unit is obtained by analyzing the learning samples through machine learning methods, The feature library can be updated to facilitate system upgrades.

4. The external analysis program of the present invention adopts the intelligent data processing method for courses to analyze the data records of different groups respectively, and obtains the performance of the student in different classrooms and after class and can send the analysis results to the parents of the students, and the analysis results Stored in the database, teachers and students themselves can view it at any time, helping parents and teachers to grasp students' learning situation in a timely manner, analyze students' personality characteristics and learning behavior habits, and provide correct guidance for students' healthy development, so as to avoid partial subjects, morality The goal of all-round development of intelligence and body provides support for science education.

5. The present invention does not affect the normal study and life of students, and automatically analyzes and records the learning behavior, which is less affected by the surrounding environment and has strong adaptability.

Description of drawings

FIG. 1 is a schematic diagram of a logic block of the learning behavior collection and analysis system of the present invention.

Fig. 2 is a working flowchart of the learning behavior collection and analysis method of the present invention.

Fig. 3 is a working flow diagram of the data analysis process of the present invention.

Fig. 4 is a working flow chart of the data acquisition interruption processing program of the present invention.

Fig. 5 is a flow chart of the work of the external analysis program of the present invention.

Detailed ways

The present invention will be described in further detail below in conjunction with the accompanying drawings and various embodiments, but the present invention is not limited to these embodiments.

As can be seen from Figure 1, the student learning behavior collection and analysis system of the present invention is composed of an external computer connected to a learning behavior data real-time automatic recording device.

The external computer can set the correct date, time, class schedule, student name, student number and other information for the system; when the system is initialized, the curriculum, the class time and get out of class end time of each class, student name and student number are synchronized to the learning Behavioral data automatic recording device.

The real-time automatic recording device for learning behavior data is fixed on the waist of the human body, which can automatically record the acceleration changes and the time of occurrence of the waist movement of the human body. At the same time, through the intelligent analysis of the recorded data, the action category is determined, and the analysis results are transmitted to an external computer for further analysis. further analysis. The real-time automatic recording device for learning behavior data of the present embodiment is composed of an action acquisition unit, an intelligent analysis and processing unit, a storage unit, a clock unit, and a communication unit. unit, clock unit and communication unit are connected.

The motion acquisition unit adopts an acceleration sensor. Different types of motion of the human body will produce different acceleration changes corresponding to the waist. When the device is turned on, the acceleration sensor is in the acquisition state to sense whether the waist of the human body is moving; if there is motion, then The acceleration sensor outputs an operation signal.

The clock unit provides year, month, day, hour, minute, and second time information for the entire device, and provides time stamps for the behavior data collected by the acceleration sensor, which can maintain normal work when the system power is interrupted.

The storage unit stores system configuration information, learning behavior classification feature library and learning behavior feature analysis results through non-volatile memory. The system configuration information includes the student’s name, student number, class schedule, class start and end time of each class, device automatic opening time and automatic closing time; the learning behavior classification library includes running, jumping, walking, shaking, standing up under the initial conditions 8 types of behaviors, such as , sitting down, stillness and other actions, are an updateable and upgradable behavioral information feature classification library, which is obtained by analyzing learning samples through machine learning methods. First, collect some sample data of behavioral actions After analyzing these samples and extracting feature values by machine learning methods, a learning behavior classification feature library can be formed. After the collected behavior data is divided into frames and feature values are calculated, it can be compared with the learning behavior classification feature library. category of the behavior.

The intelligent analysis and processing unit receives the interrupt signal generated by the acceleration sensor and responds to it, and saves the obtained motion data to the storage unit after adding time tags as required; at the same time, it conducts short-term analysis on the obtained motion data and calculates the action Frequency of occurrence, duration of each action, and action intensity and other characteristic parameters; then it uses pattern classification intelligent algorithm or other algorithms to compare the analysis results with the samples in the learning behavior classification feature library in the storage unit to determine the behavior category.

The communication unit is also connected to the external computer, which includes USB interface, Bluetooth, wifi, GPRS, Zigbee and other wired or wireless communication modules, which can transmit the processing results to the external computer and receive instructions from the external computer. At the same time, it can realize The setting of the configuration information in the storage unit and the updating and upgrading of the learning behavior taxonomy library.

The method of using the student learning behavior collection and analysis system is shown in Figure 2, including the following implementation steps:

Step 1: After the system starts, judge whether it is connected to an external computer, if yes, go to step 2, otherwise go to step 3; the automatic opening and closing time of the system can be set, such as opening at 8:00 am and closing at 9:00 pm; the system follows The set opening time and closing time are automatically opened and closed, and can also be switched manually.

Step 2: Configure the system, including setting the student name, student number, class schedule, system automatic opening time and automatic closing time, etc., and synchronize the system time.

Step 3: Initialize the hardware system and software system, and clear the data analysis flag.

Step 4: start the data analysis process; see Figure 3, the data analysis process is specifically implemented by the following steps.

Step 4.1: Determine whether the data analysis flag is set, if it is set, go to step 4.2, otherwise, repeat step 4.1.

Step 4.2: According to the time tag in the data unit, divide the data into frames, and calculate the characteristics of each data frame, including action frequency, intensity, direction, duration, etc.

Step 4.3: Compare the frame sequence features with the classification feature library in the storage unit to determine the action category, including 8 categories of running, jumping, walking, shaking, standing up, sitting down, still and other actions.

Step 4.4: After adding the time tag, form a data record, namely action time, action category, action direction, average intensity, frequency and duration, and store it in the storage unit.

Step 4.5: Clear the data analysis flag, and then return to step 4.1 to start data analysis again.

Step 5: Set up the data acquisition interrupt processing program; apply a data acquisition buffer and a data analysis buffer, first add a time tag to the obtained sensor data, that is, record the time when the action occurs, and then compose the sensor data and the time tag A data unit is stored in the data acquisition buffer; when the acquisition buffer is full, all data is copied to the analysis buffer for analysis. At the same time, the acquisition buffer continues to store new data units, and the previously stored data units are re-stored data unit coverage. Referring to Fig. 4, it is specifically implemented by the following steps.

5.1) Read clock information.

5.2) Obtain the sensor data, add a time tag to form a data unit, and store it in the storage location pointed to by the current data acquisition buffer pointer; the data unit includes a time tag, x acceleration, y acceleration and z acceleration, etc.;

5.3) Determine whether the buffer is full, if the buffer is full, proceed to steps 5.4) to 5.6); otherwise, proceed to step 5.7).

5.4) Copy the data from the data acquisition buffer to the data analysis buffer.

5.5) Point the data collection buffer pointer to the head of the buffer.

5.6) Set the data analysis flag and end the program.

5.7) Point the data acquisition buffer pointer to the next storage location, and end the program.

Step 6: Judging whether the system is finished, if it is finished, exit the system, otherwise continue to judge.

Step 7: The external analysis program sends the data records in the storage unit to an external computer for further analysis. The analysis process is combined with the curriculum arrangement to analyze different courses and breaks. Referring to Figure 5, it is specifically implemented by the following steps:

Step 7.1: Synchronize system configuration information and time information to the real-time automatic recording device for learning behavior data.

Step 7.2: After obtaining the data records, group the data records according to the class schedule and the start and end time of each class, and group the data records of the same type of courses into one group.

Step 7.3: The grouped data records are stored in the computer database.

Step 7.4: Each set of data is analyzed with curriculum-specific smart data processing methods.

Step 7.5: Store the analysis results in the database and end the program.

The learning behavior collection and analysis method of the present invention utilizes the data automatic collection interruption processing program and the data analysis process to automatically collect the learning behavior data of the students and analyze and store them, and further classify and analyze the learning behavior habits of the students by using the external analysis process. Or pattern analysis provides a new approach. The learning behavior collection and analysis method of the present invention is not limited to the above-mentioned embodiments.

Claims (3)

1. utilize Students ' Learning behavior gather and analysis system to carry out a method for behavior gather and analysis, wherein, this system comprises:

Action collecting unit, for gathering human body acceleration information, and output action signal, described action collecting unit is acceleration transducer;

Clock unit, for whole device provides temporal information, and the behavioral data collected for acceleration transducer provides time mark;

Storage unit, for storage system configuration information, learning behavior characteristic of division storehouse and learning behavior signature analysis result;

Communication unit, is connected with outer computer, for the result of intellectual analysis and processing unit is transferred to outer computer, and receives the instruction that outer computer sends, and realizes the setting of configuration information and the renewal of learning behavior class library and upgrading;

Intellectual analysis and processing unit, be connected with action collecting unit, storage unit, clock unit and communication unit respectively by its input/output port, thus coordinate the action between them, and process data;

The method comprises the following steps:

1) start up system, judges whether to connect outer computer, if so, then carry out step 2), if not, then carry out step 3);

2) system configuration, arranges student name, student number, curriculum schedule, system automatic opening time and automatic shut-in time, and carries out system time synchronization;

3) initialization is carried out to system, data analysis mark is removed;

4) data analysis process is started

4.1) judge the whether set of data analysis mark, if set, then enter step 4.2), otherwise, repeat 4.1);

4.2) according to the time tag in data cell, to data framing, and the feature of each Frame is calculated;

4.3) is compared in the characteristic of division storehouse in frame sequence feature and storage unit, determine action classification;

4.4), after adding time tag, data record is formed, and stored in storage unit;

4.5) clear data analysis mark, returns step 4.1);

5) setting data gathers interrupt handling routine

5.1) temporal information is read;

5.2) obtain sensing data, add time tag, form a data cell, data cell comprises time tag, x acceleration, y acceleration and z acceleration, points to memory location stored in Current data acquisition buffer pointer;

5.3) judge that whether buffer zone is full, if buffer zone is full, then carry out step 5.4) to 5.6); Otherwise, then carry out step 5.7);

5.4) data acquisition buffer data is copied to data analysis buffer zone;

5.5) data acquisition buffer pointer is pointed to stem;

5.6) put data analysis zone bit, terminate program;

5.7) data acquisition buffer pointer is pointed to next memory location, terminate program;

6) judge whether system terminates, if terminate, then log off, otherwise continue to judge;

7) external analysis program, is sent to outer computer by the data record of storage and is further analyzed, and analytic process, in conjunction with timetable arrangement, is analyzed for different courses and break.

2. method according to claim 1, is characterized in that: described step 7) specifically:

7.1) by the synchronizing information set by system configuration to learning behavior data real-time automatic recording device;

7.2), after obtaining data records, divide into groups to data record according to the attending class of curriculum schedule and every class, play time, the data record with kind course is divided into one group;

7.3) the data record after step 7.2 being divided into groups is stored in Computer Database;

7.4) by the data intelligence processing method for course to often organizing data analysis;

7.5) by analysis result stored in database, terminate program.

3. method according to claim 1, is characterized in that: described step 4.2) in the feature of Frame refer to operating frequency, intensity, direction and duration; Described step 4.3) in action classification refer to race, jumping, walk, rock, stand up, to sit down and static; Described step 4.4) in data record comprise actuation time, action classification, direction of action, mean intensity, operating frequency and duration.