KR102558969B1 - Method for reinforced online learning method using problem solving measuring fuzzy mental state - Google Patents

Abstract

The enhanced online learning method according to the present invention includes a problem presenting step (S100) of presenting a learning problem to which at least one learning ability for solving the problem is assigned to the learner using an output unit, an input step (S300) of receiving an answer to the learning problem from the learner through an input unit (S300), a psychological state determination step (S400) of determining the learner's psychological state for the learning problem by a psychological decision unit, and a correct answer scoring unit for the problem based on the input value for the learning problem in the input step. A scoring step (S500) of scoring whether or not the answer is given, an achievement analysis step (S600) of analyzing the achievement by learning ability of the learner by analyzing the results scored in the scoring step by the learning state determination unit (S600), and a learning target determining step (S700) of determining the learning target by considering the achievement by learning ability based on the result analyzed in the analysis step by the problem setting unit.

Description

Enhanced online learning method through problem solving to measure fuzzy psychological state { METHOD FOR REINFORCED ONLINE LEARNING METHOD USING PROBLEM SOLVING MEASURING FUZZY MENTAL STATE}

The present invention relates to an enhanced online learning method through problem solving that analyzes achievement by reflecting a user's psychological state.

Although many online learning methods are being developed, there is a lack of advanced technology in learning assessment methods. In many cases, it is simply determined whether the answer is correct or not to determine whether the learner understands the problem or not, and then proceeds with reinforcement learning.

The learner sometimes understands the problem in fragments, sometimes does not understand the whole thing at all, and sometimes learns the correct answer by mistake.

In the case of determining a learner's learning achievement simply based on correct and incorrect answers, it is difficult to determine whether the learner has achieved an accurate level of achievement.

Considering these various possible situations, it is necessary to evaluate the learner's achievement and prepare a method to efficiently proceed reinforcement learning accordingly.

Patent Registration No. 10-2431604

In order to solve the above problems, an object of the present invention is to provide an enhanced online learning method through problem solving that analyzes achievement by reflecting the learner's psychological state including the fuzzy state.

The enhanced online learning method according to the present invention includes a problem presenting step (S100) of presenting a learning problem to which at least one learning ability for solving the problem is assigned to the learner using an output unit, an input step (S300) of receiving an answer to the learning problem from the learner through an input unit (S300), a psychological state determination step (S400) of determining the learner's psychological state for the learning problem by a psychological decision unit, and a correct answer scoring unit for the problem based on the input value for the learning problem in the input step. A scoring step (S500) of scoring whether or not the answer is given, an achievement analysis step (S600) of analyzing the achievement by learning ability of the learner by analyzing the results scored in the scoring step by the learning state determination unit (S600), and a learning target determining step (S700) of determining the learning target by considering the achievement by learning ability based on the result analyzed in the analysis step by the problem setting unit.

In one embodiment, after the learning target determining step (S700), a reinforced problem presentation step (S800) of submitting a new problem to the learner based on the determined learning target so that the learner can take a retest It can be characterized by further including.

In one embodiment, in the mental state determination step (S400), the mental state of the user may be characterized in that it is directly input from the user.

In one embodiment, the psychological state of the learner may include a certain state and a fuzzy state.

In one embodiment, the certainty state may be characterized in that it is divided into a correct answer certainty state and an incorrect answer certainty state.

In one embodiment, the achievement analysis step (S600) includes a case where the learner's input value for the learning problem is not correct when the learner's psychological state is in a correct answer certainty state, includes regardless of whether the learner's input value for the learning problem is correct when the learner's psychological state is in a fuzzy state, and includes whether or not the learner's input value for the learning problem is correct in an incorrect answer state, the learner evaluates the learning ability related to the learning problem as an unaccomplished target requiring re-learning can be done with

In one embodiment, the achievement analysis step (S600) is characterized in that the learner analyzes the learning utilization of the learning ability related to the learning problem as a fuzzy area, regardless of whether the learner's input value for the learning problem is correct, when the learner's psychological state is in a correct answer certainty state, the learner's input value for the learning problem is not correct, and the learner's psychological state is in a fuzzy state, regardless of whether the learner's input value for the learning problem is correct or not. It may be characterized by analyzing the learning utilization of the related learning ability as an unlearned area.

In one embodiment, after the problem presentation step (S100), the psychological state information measuring step (S200) is further included, and the psychological state information measuring step is characterized in that the problem option recognition time for recognizing each option of the problem is individually measured. In the area allocated to each option expressed on the display device, the relative time value is calculated and measured by integrating the movement trajectory where the focus point of the learner moves and the instantaneous dwell time for each trajectory. that can be characterized.

In one embodiment, the individual problem option recognition time may be characterized in that it is calculated by the following formula.

Ta: final value of relative time to that object

m: the number of times the focus point entered the target

x _in , y _in : the first time of entry into the target in the nth entry attempt

x _out, y _out : Last time to exit the target on the nth entry attempt

ts: the instantaneous stay time of the focus point at the x, y coordinates included in the target and moving

In one embodiment, the learner's focus point may be characterized in that the learner's gaze is tracked using an eye-tracking unit to calculate a movement trajectory and an instantaneous stay time for each trajectory.

According to the present invention, since the evaluation of the learning ability is performed by reflecting the psychological state of the test taker, more accurate achievement evaluation is possible. Compared to the prior art, which simply checks the correct answer and reflects the achievement level, a three-dimensional learning achievement evaluation can be performed.

On the other hand, by identifying achievement evaluation and learning utilization, retraining and retesting, learners can utilize a learning system that can gradually raise their achievements step by step.

In addition, when the user inputs the psychological state of the test taker, subjectivity may intervene or intentional errors may occur. Since this can be objectively evaluated based on the input data, a more accurate psychological state evaluation is possible, and through this, accurate achievement evaluation and achievement improvement through re-learning can be effectively progressed.

1 is a flowchart of an online learning method according to an embodiment of the present invention.
FIG. 2 is a configuration diagram of a system using the online learning method according to the embodiment of FIG. 1 .
3A to 3F are diagrams for explaining the online learning method according to the embodiment of FIG. 1 .
4 is a flowchart of an online learning method according to another embodiment of the present invention.
5A and 5B are conceptual diagrams illustrating a part of the online learning method according to the embodiment of FIG. 4 .

Hereinafter, with reference to the accompanying drawings, a preferred embodiment will be described with reference to the accompanying drawings. In this process, the thickness of lines or the size of components shown in the drawings may be exaggerated for clarity and convenience of description. In addition, terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to the intention or custom of a user or operator. Therefore, definitions of these terms will have to be made based on the content throughout this specification.

In this specification, a module may mean a functional and structural combination of hardware for implementing the technical idea of the present invention and software for driving the hardware. For example, the module may refer to a predetermined code and a logical unit of hardware resources for executing the predetermined code, and that it does not necessarily mean a physically connected code or a type of hardware. An average expert in the technical field of the present invention can easily infer.

In addition, when a component is referred to as being "connected" or "connected" to another component, it may be directly connected or connected to the other component, but it should be understood that other components may exist in the middle. On the other hand, when an element is referred to as “directly connected” or “directly connected” to another element, it should be understood that no other element exists in the middle.

Singular expressions include plural expressions unless the context clearly dictates otherwise. In addition, throughout the present specification, when a part "includes" a certain component, it means that it may further include other components without excluding other components unless otherwise stated. It is apparent to those skilled in the art that the present invention can be embodied in other specific forms without departing from the essential characteristics of the present invention.

In addition, the following examples do not limit the scope of the present invention, but are presented as examples only, and there may be various embodiments implemented through the technical idea. Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Like reference numerals designate like elements in the drawings.

Overall process of online learning method

1 is a flowchart of an online learning method according to an embodiment of the present invention. FIG. 2 is a configuration diagram of a system using the online learning method according to the embodiment of FIG. 1 .

Referring to FIG. 1, the online learning method according to this embodiment includes a problem presentation step (S100), an input step (S300), a psychological state determination step (S400), a scoring step (S500), an achievement analysis step (S600), and a learning target determination step (S700). The online learning method may further include a reinforced problem presentation step (S800) after the learning target determining step (S700).

On the other hand, referring to FIG. 2, the online learning system according to this embodiment includes an output unit 110, an input unit 120, a psychological determination unit 130, a correct answer scoring unit 140, an achievement analysis unit 150, a learning target determination unit 160, and a problem question unit 17. Each configuration can be composed of function-specific functions or modules that implement separate functions within a single software, and separate configuration display for each configuration is to help understanding of driving and operation. It is composed of overall coding, and since it is common for each functional configuration to be responsible for performing functions in an operation unit such as a central processing unit according to input coding, each configuration does not necessarily mean that it is composed of separate software, and is physically operated in a separate device. It is not limited to functioning in a separate area. Hereinafter, with reference to each configuration, details of each step mentioned in FIG. 1 will be described.

(a) Structure of training problem data

3A to 3F are diagrams for explaining the online learning method according to the embodiment of FIG. 1 .

FIG. 3A is a diagram illustrating a learning ability information dataset for an initial test question applied to the online learning method according to the embodiment of FIG. 1 .

Referring to FIG. 3A , data of a learning problem used in this embodiment includes learning ability as data subordinate to a problem ID. This means the learning items required to solve the problem. For example, for a learning problem with a problem ID of 1, the correct answer to the problem can be answered only when the learning ability corresponding to the basic computational ability is possessed. This can be used as a basis for feedback on the improvement of achievement for future learning problems. A learning capability subordinate to the learning problem data does not include a single capability, but may include a plurality of learning capabilities, and the number is not limited.

(b) Problem presentation step (S100)

In the problem presentation step (S100), a learning problem assigned with at least one learning ability for solving the problem is presented to the learner using the output unit 110. Typically, questions can be presented through a PC terminal, tablet, or smart phone, and the above-mentioned output unit 110 refers to the image output unit of the above-mentioned devices. As described above, each learning problem is assigned at least one learning ability required to solve the problem, and although this is not clearly displayed to the examiner, the necessary learning ability may be indicated in some cases.

(c) Input step (S300)

In the input step (S300), an answer to the learning problem is input from the learner through an input unit. The input unit can also receive input using a PC terminal, tablet, or smart phone in general, and the mentioned input unit may include various input devices such as a keyboard, mouse, touch, touch pen, and the like of each terminal. The psychological state information measurement step (S200), which is performed in parallel with the input step (S300), will be described later as a separate embodiment.

(d) Psychological State Determination Step (S400)

FIG. 3B is a diagram illustrating a test taker's psychological dataset applied to the online learning method according to the embodiment of FIG. 1 .

Referring to FIG. 3B , in the psychological state determination step (S400), the psychological state of the learner for the learning problem is determined by the psychological determination unit 130. The psychological state of the learner includes a certain state and a fuzzy state. The certainty state can be subdivided into correct answer certainty state and incorrect answer certainty state.

The psychological state of the learner can be directly inputted, or a method of inferring it by analyzing recognizable data from the learner can be used. In the direct input method, learners receive input by selecting correct answer certainty state, fuzzy state, and wrong answer certainty state for the learning problem. The method of analyzing the data that can be obtained from the learner is determined to utilize various context information that can be obtained from the terminal, such as the learner's focus change that can be received while solving the learning problem and the time spent on the problem. One embodiment of this approach is described in detail below.

(e) Scoring step (S500)

FIG. 3C is a diagram showing incorrect answer and achievement data sets applied to the online learning method according to the embodiment of FIG. 1 .

Referring to FIG. 3C , in the scoring step (S500), the answer scoring unit 140 scores whether or not the correct answer is correct for the problem based on the value input for the learning problem in the input step (S400). Since information on correct answers is included when the data of the learning problem is created, it is possible to perform calculations immediately after input of information on the answer to the learning problem is completed by utilizing this database.

(f) Achievement analysis step (S600)

Referring back to FIG. 3C , in the achievement analysis step (S600), the achievement analysis unit 150 analyzes the learner's achievement by learning ability by analyzing the result scored in the scoring step (S500). This can be judged by whether or not individual learning abilities that can be inferred from learning problems have been achieved based on the test taker's psychology of each learning problem and whether or not they answered correctly or not.

In the achievement analysis step (S600), the learner's achievement is classified into achievement and non-achievement, and the case of non-achievement is as follows. (i) When the learner's psychological state is in the correct answer certainty state and the learner's input value for the learning problem is not correct, (ii) in the fuzzy state, the learner's psychological state is included in the non-achievement regardless of whether the learner's input value for the learning problem is correct, (iii) in the learner's psychological state is in the wrong answer certainty state, regardless of whether the learner's input value for the learning problem is correct or not.

FIG. 3D is a diagram illustrating learning ability and achievement datasets applied to the online learning method according to the embodiment of FIG. 1 .

While the dataset in Fig. 3c measures achievement and non-achievement for individual problems, the dataset in Fig. 3d determines achievement based on learning ability, and this is the final method for determining achievement for the corresponding learning ability.

Instead, when various result values are derived for one learning ability, the result values for the corresponding learning ability may be summed by setting a ratio of achievement and non-achievement. For example, if 10 learning problems are asked for one learning ability and 9 achievements and 1 non-achievement are measured, it can be considered an achievement, and if 5 achievements and 5 non-achievements are measured, it can be regarded as an achievement.

If achievement is included as unachieved, it means that the learner's learning ability related to the learning problem is evaluated as an object that needs re-learning.

(g) Learning target decision step (S700)

FIG. 3E is a diagram showing a learning utilization data set for each learning ability applied to the online learning method according to the embodiment of FIG. 1 .

Referring to FIG. 3E , learning utilization data can be constructed in parallel with determining learning achievement, which allows more gradual classification than the previous achievement determination. This is done by the learning object determination unit.

In terms of learning utilization, what is determined as a fuzzy region is (i) when the learner's psychological state is in a certain correct answer state and the learner's input value for the learning problem is not correct, and (ii) when the learner's psychological state is in a fuzzy state and the learner's input value for the learning problem is in a correct answer or not.

In terms of learning utilization, what is determined as an unlearned area is (i) a case where the learner's psychological state is in a state of being sure of an incorrect answer, regardless of whether the learner's input value for the learning problem is correct.

In learning utilization, the fuzzy area is determined as having knowledge of the corresponding area, but not fully understanding it, and the unlearned area can be determined as a case of not having knowledge of the corresponding area. For example, when a separate incorrect answer note is created for a corresponding learning ability, an incorrect answer note can be created in an unlearned area, and a fuzzy note can be created in a fuzzy area as a concept different from incorrect answer notes.

In terms of the determined utilization of learning, necessary learning information before re-evaluation can be provided to learners. This provides a distinction between the fuzzy area and the unlearned area in terms of learning utilization, and through this, the learner's re-learning can proceed efficiently.

(h) Enhanced problem presentation step (S800)

FIG. 3F is a diagram illustrating a learning ability information dataset for a secondary test applied to the online learning method according to the embodiment of FIG. 1 .

Referring to FIG. 3F , the question setting unit submits a new question to the learner based on the determined learning target to take a retest. Questions in the retest can be calculated based on previous achievements or utilization of learning.

In FIG. 3F, the problem of basic computational ability, which was first presented in FIG. 3a, is omitted, and problems related to learning ability for the achievement standard unfulfilled learning ability, the fuzzy area of the learning utilization standard, and the unlearned area are presented.

By repeating this process, the learner can improve the overall learner's achievement by proceeding with reeducation and re-evaluation, including unfulfilled areas and vaguely learned areas.

Embodiment of Automatically Determining Purge State

4 is a flowchart of an online learning method according to another embodiment of the present invention.

Referring to FIG. 4 , the embodiment of FIG. 1 includes the same process except for further including a psychological state information measuring step (S200).

The psychological state information measurement step (S200) is a step of collecting data that can be obtained in the course of the learner inputting an answer to the corresponding problem in the input step (S300) after the problem presentation step (S100) in order to recognize whether the candidate's psychological state for the corresponding learning problem is correct answer check state, fuzzy state, or incorrect answer check state.

A variety of information can be acquired from terminals such as PCs, tablets, and mobile phones until an answer is input. First of all, the time spent on the problem can be measured, and this can be used as a method of detecting that the learner is in a fuzzy state depending on how much time they stayed in each correct answer option.

In the step of measuring mental state information (S200) of this embodiment, the problem fingerprint recognition time in which the learner recognizes the fingerprint of the problem or the problem option recognition time in which the learner recognizes each option of the problem is measured.

In this method of measuring recognition time, a relative time value is calculated and measured by integrating the trajectory of the learner's focus point and the instantaneous stay time for each trajectory in the area allocated to the problem text and each option.

Equation 1

Ta: final value of relative time to that object

m: the number of times the focus point entered the target

x _in , y _in : Initial coordinates that entered the target at the nth entry attempt

x _out, y _out : The final coordinates of exiting the target at the nth entry attempt

ts: the instantaneous stay time of the focus point at the x, y coordinates included in the target and moving

Unlike the method of simply measuring only the length of the trajectory, it is possible to accurately measure how long the focus point actually stayed in the corresponding area. A stay time in a corresponding region, which is a group of specific coordinates, may be calculated inversely by measuring and integrating how long the focus point has stayed at individual coordinates of each region. The advantage of this method is that it is possible to recalculate the stay time in reverse when the corresponding area is designated, and it is possible to review the subdivided area within the area to be measured.

This can be implemented in other ways as shown below.

Equation 2

Ta: final value of relative time to that object

m: the number of times the focus point entered the target

t _in : The first time of entering the target in the nth entry attempt

t _out : The final time of exiting the target in the nth entry attempt

In this case, it is a method of calculating the stay time from the time of entering the designated area to the time of leaving the designated area, and the measurement is simple. However, this has the disadvantage that a designated area must always be defined, and it is impossible to perform other analyzes within that area.

Unlike the method of simply measuring only the length of the trajectory, it is possible to accurately measure how long the focus point actually stayed in the corresponding area.

The learner's focus point is calculated by tracking the movement trajectory of the mouse cursor and the instantaneous dwell time for each trajectory, or by using a separate eye tracking unit, the learner's gaze is tracked to calculate the movement trajectory and the instantaneous dwell time for each trajectory.

5A and 5B are conceptual diagrams illustrating a part of the online learning method according to the embodiment of FIG. 4 .

Referring to FIG. 5A first, when one learning problem is displayed, a learning fingerprint (QM) and options (Q1, Q2, Q3, Q4) may be displayed respectively. Each learning question (QM) and options (Q1, Q2, Q3, Q4) can be allocated a designated area, and if a focus point is allocated within this area, it is assumed that the learner is reviewing the corresponding fingerprint or option.

The focus point can be tracked through equipment such as the PC's mouse, the user's eye gaze captured by the camera, and the smart pen used on the tablet.

Referring to FIG. 5B , the stay time may be calculated differently for each region. The figure does not show the dwell time at which the focus point stays for each coordinate, but assuming that it is calculated by integrating it, it can be seen that the most dwell time occurs in the Q3 area. In this case, the test taker's psychological state is regarded as a correct answer certainty state.

As in the case of FIG. 5C , when the stay time is uniformly generated in each area of Q1, Q2, Q3, and Q4, the psychological state of the test taker is regarded as a fuzzy state. Through the trajectory analysis of the focus point, it is possible to measure the user's psychological state for each choice.

As a result, the learner does not have to input the candidate's psychological state for the corresponding learning problem one by one, and in some cases, the function of preventing the learner from intentionally entering an incorrect psychological state can also be performed.

As described above, although the embodiments have been described with limited drawings, those skilled in the art can apply various technical modifications and variations based on the above. For example, even if the described techniques are performed in an order different from the described method, and/or components of the described system, structure, device, circuit, etc. are combined or combined in a different form from the described method, or replaced or substituted by other components or equivalents, appropriate results can be achieved. Therefore, other implementations, other embodiments, and equivalents of the claims are within the scope of the following claims.

110: output unit
120: input unit
130: hearing decision unit
140: correct answer scoring unit
150: achievement analysis unit
160: learning target determination unit
170: problem solving department

Claims (10)

A problem presentation step (S100) of presenting a learning problem to which at least one learning ability for problem solving is assigned to a learner using an output unit;
An input step of receiving an answer to the learning problem from the learner through an input unit (S300);
a psychological state determining step (S400) of determining the learner's mental state with respect to the learning problem by a psychological determination unit;
A scoring step (S500) of scoring, by a correct answer scoring unit, whether or not the correct answer is correct for the problem based on the input value for the learning problem in the input step;
An achievement analysis step (S600) of analyzing, by a learning state determination unit, the achievement level for each learning ability of the learner by analyzing based on the scored result in the scoring step; and
A learning target determining step (S700) of determining a learning target by a problem setting unit in consideration of the achievement level for each learning ability based on the result analyzed in the analysis step (S700); and
Including an enhanced problem presentation step (S800) of submitting a new problem to the learner to take a retest based on the determined learning target,

The psychological state of the learner includes a certainty state and a fuzzy state, and the certainty state is characterized in that it is divided into a correct answer certainty state and an incorrect answer certainty state,

The achievement analysis step (S600) is
When the learner's psychological state is in a correct answer certainty state, the learner's input value for the learning problem is incorrect; and
When the learner's psychological state is in a fuzzy state, regardless of the correct answer of the learner's input value for the learning problem;
Characterized in that the learner analyzes the learning utilization of the learning ability related to the learning problem as a fuzzy area,

When the learner's psychological state is in the wrong answer certainty state, regardless of the correct answer of the learner's input value for the learning problem;
Characterized in that the learner analyzes the learning utilization of the learning ability related to the learning problem as an unlearned area,

After the problem presentation step (S100),
Further comprising a psychological state information measuring step (S200),

The psychological state information measuring step,
It is characterized in that the problem option recognition time for recognizing each option of the problem by the learner is individually measured,

The individual problem option recognition time,
In the area allocated to each of the options expressed on the output unit,
Characterized in that the relative time value is calculated and measured by integrating the movement trajectory along which the learner's focus point moves and the instantaneous stay time for each trajectory,

The individual problem option recognition time is calculated by Equation 1,
Ta is the final value of the time relative to that object,
m is the number of times the focus point entered the target,
x _in and y _in are the coordinates of entering the target at the nth entry attempt,
x _out and y _out are the final coordinates of exiting the target at the nth entry attempt,
ts is characterized in that the instantaneous stay time of the focus point of the x, y coordinates included in the object and moving,
[Equation 1]


When the largest stay time occurs in an area where the problem option recognition time is unique, the psychological state is regarded as the correct answer certainty state,
Reinforced online learning method through problem solving in which the problem option recognition time is regarded as the fuzzy state when an equal stay time occurs in each area.
delete delete delete delete delete delete delete delete According to claim 1,
The learner's focus point is
An enhanced online learning method through problem solving, characterized in that by using a gaze tracking unit, tracking the learner's gaze to calculate the movement trajectory and the instantaneous stay time for each trajectory.