Memory grading training algorithm and system

Abstract

The invention relates to the technical field of memory training, in particular to a memory grading training algorithm and system, and the algorithm comprises the steps: carrying out the memory test of a to-be-trained person, and obtaining a first memory curve; matching the analysis information, the training target and a plurality of algorithm containers to obtain a first algorithm container; performing memory training on the to-be-trained person according to the first algorithm container; testing the trained to-be-trained person to obtain a second memory curve; and updating the first algorithm container according to a comparison result of the first memory curve and the second memory curve. According to the invention, the memory basis and characteristics of each to-be-trained person can be accurately grasped by performing memory analysis on the to-be-trained person. On the basis, the analysis information is matched with the training target and the algorithm container, the first algorithm container is selected for each to-be-trained person, personalized memory training is achieved, the training is more suitable for the personal actual situation, and the pertinence and effectiveness of the training are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of memory training, and in particular to a memory classification training algorithm and system. Background Art

[0002] In today's era of information explosion, memory, as an important part of human cognitive ability, has a profound impact on personal learning, work and life. Good memory can help students absorb knowledge more efficiently and achieve excellent academic results; for professionals, strong memory can help improve work efficiency and better cope with complex and changing work tasks.

[0003] At present, there are many memory training methods on the market, such as the traditional memory palace method, associative memory method, etc. However, these methods often lack systematicity and personalization. Most of them adopt a "one-size-fits-all" training model, failing to fully consider the differences between different individuals in terms of memory foundation, learning ability, and training goals, resulting in uneven training results. In addition, the existing training methods are difficult to accurately evaluate the training effect, and it is impossible to adjust the training strategy in time according to the actual situation of the individual.

[0004] Therefore, how to conduct memory training based on individual circumstances is a current research direction. Summary of the invention

[0005] 1. Purpose of the invention The purpose of the present invention is to provide a memory classification training algorithm and system which can improve the pertinence and effectiveness of training according to individual conditions.

[0006] (II) Technical solution To solve the above problems, the present invention provides a memory hierarchical training algorithm, comprising: Conduct memory tests on trainees to obtain the first memory curve; Analyzing the first memory curve to obtain analysis information; Matching the analysis information, the training target and a plurality of algorithm containers to obtain a first algorithm container; Performing memory training on the person to be trained according to the first algorithm container; Testing the trained personnel to obtain the second memory curve; The first algorithm container is updated according to a comparison result between the first memory curve and the second memory curve.

[0007] In another aspect of the present invention, preferably, analyzing the first memory curve to obtain analysis information includes: Acquire a plurality of descending gradient segments of the first memory curve; The first memory curve is segmented according to a plurality of descending gradient segments to obtain memory curve segment information.

[0008] In another aspect of the present invention, preferably, the acquiring of the plurality of descending gradient segments of the first memory curve comprises: Performing mathematical modeling on the first memory curve to obtain a function expression of the memory curve; calculating a derivative of the first memory curve to obtain a rate of change of the first memory curve; According to the change rate of the first memory curve, a descending gradient of the memory curve is determined, and the descending gradient is divided into a plurality of descending gradient segments.

[0009] In another aspect of the present invention, preferably, the algorithm container includes an algorithm drawer and an algorithm capacity; the algorithm drawer includes at least one memory training algorithm, and the algorithm capacity corresponds to the difficulty coefficient and / or training duration of the memory training algorithm.

[0010] In another aspect of the present invention, preferably, matching the analysis information and the training target with a plurality of algorithm containers to obtain a first algorithm container includes: Identifying peak memory periods and low memory periods in the analysis information; Determine a training strategy according to the peak memory period, the trough memory period and the training goal; The training strategy is matched with a plurality of algorithm containers to obtain a first algorithm container.

[0011] In another aspect of the present invention, preferably, matching the analysis information and the training target with a plurality of algorithm containers to obtain a first algorithm container further comprises: Monitoring the training results of the first algorithm container in real time; The algorithm container, algorithm drawer and algorithm capacity are dynamically adjusted according to the training results.

[0012] In another aspect of the present invention, preferably, dynamically adjusting the algorithm container, the algorithm drawer and the algorithm capacity according to the training results comprises: Selecting an initial number of algorithm drawers from the first algorithm container for training; If the algorithm capacity is insufficient, increase the number of algorithm drawers; If the number of algorithm drawers exceeds the maximum value, other algorithm containers are selected to continue training.

[0013] In another aspect of the present invention, preferably, the dynamically adjusting the algorithm container, the algorithm drawer and the algorithm capacity according to the training results further includes: Selecting the maximum capacity of an algorithm drawer from the first algorithm container for training; If the training duration or training results cannot reach the set value under the maximum capacity, an algorithm container with a larger capacity will be selected to continue training.

[0014] In another aspect of the present invention, preferably, updating the first algorithm container according to the comparison result between the first memory curve and the second memory curve includes: If the comparison result shows that the rising amplitude of the second memory curve relative to the first memory curve is greater than or equal to the rising amplitude threshold, the training difficulty coefficient and training duration of the first algorithm container are increased; If the comparison result shows that the increase of the second memory curve relative to the first memory curve is less than the increase threshold, the training difficulty coefficient and training duration of the first algorithm container are reduced; If the comparison result shows that the second memory curve shows a downward trend relative to the first memory curve, the first algorithm container is replaced and retrained.

[0015] Another aspect of the present invention is preferably a memory hierarchical training system, comprising: The first acquisition module: conducts a memory test on the trainee and obtains the first memory curve; Analysis module: analyzing the first memory curve to obtain analysis information; Matching module: matching the analysis information, training objectives and multiple algorithm containers to obtain a first algorithm container; Training module: performing memory training on the trainee according to the first algorithm container; The second acquisition module: testing the trained personnel to obtain the second memory curve; An updating module is configured to update the first algorithm container according to a comparison result between the first memory curve and the second memory curve.

[0016] (III) Beneficial effects The above technical solution of the present invention has the following beneficial technical effects: The present invention obtains the first memory curve by conducting a memory test on the trainees, and analyzes the analysis information to accurately grasp the memory foundation and characteristics of each trainee. Based on this, the analysis information is matched with the training goal and the algorithm container, and the first algorithm container is selected for each trainee to achieve personalized memory training, avoiding the drawbacks of the traditional "one size fits all" training mode, making the training more in line with the actual situation of the individual, and significantly improving the pertinence and effectiveness of the training. After the training, the second memory curve is tested again, and the algorithm container is updated according to the comparison results of the first and second memory curves. This process realizes the dynamic adjustment of the training strategy, and the algorithm can be optimized in real time according to the effect of each training, ensuring that the training method always adapts to the ability improvement and changes of the trainees, continuously providing the most suitable training plan for the trainees, and constantly tapping the memory improvement potential of the trainees. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the overall process of an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present invention. In addition, in the following description, the description of well-known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present invention.

[0019] Obviously, the described embodiments are only some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] In the description of the present invention, it should be noted that the terms “first” and “second” are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0021] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0022] Embodiment 1 A memory hierarchical training algorithm, Figure 1 FIG. 1 shows a schematic diagram of the overall process of an embodiment of the present invention. Figure 1 As shown, including: Conduct a memory test on the trainees to obtain the first memory curve; the memory test tool can be combined with a variety of methods such as digit span test, vocabulary memory test, graphic memory test, etc. to comprehensively examine the trainees' instantaneous memory, short-term memory and long-term memory abilities. During the test, control the test environment to ensure that the environment is quiet and free of interference to obtain accurate test data. By recording and analyzing the test data, draw the first memory curve of the trainees.

[0023] The first memory curve is analyzed to obtain analysis information, and data analysis technology can be used to deeply analyze the first memory curve. From the ups and downs, peaks, valleys and other characteristics of the curve, the performance of the trainee in different memory durations (instant memory, ability to convert short-term memory into long-term memory) and other aspects can be analyzed. In this embodiment, the analysis of the first memory curve to obtain analysis information includes: obtaining a plurality of descending gradient segments of the first memory curve; The first memory curve is segmented according to a plurality of descending gradient segments to obtain memory curve segment information.

[0024] The step of obtaining a plurality of descending gradient segments of the first memory curve comprises: Mathematically model the first memory curve to obtain a function expression of the memory curve; use curve fitting technology to approximate these points with mathematical functions according to the data points on the curve. For example, a polynomial function may be used so that the polynomial function can represent the memory curve as accurately as possible. Of course, it is also possible to select other suitable function types such as exponential function, logarithmic function, etc. for fitting according to the actual shape of the curve to obtain a function expression that best fits the actual curve.

[0025] Calculate the derivative of the first memory curve to obtain the rate of change of the first memory curve. According to the derivative formula, derive the function expression obtained above. If the function is a polynomial function, follow the derivative rule to derive each term in turn to obtain the derivative function. The value of this derivative function at each point represents the rate of change of the original memory curve at that point, that is, the slope. A positive slope indicates that the curve is rising; a negative slope indicates that the curve is falling; the larger the absolute value of the slope, the faster the curve rises or falls.

[0026] According to the rate of change of the first memory curve, the descending gradient of the memory curve is determined, and the descending gradient is divided into several descending gradient segments. A reasonable gradient judgment threshold is set. When the derivative (i.e., the rate of change) is less than the threshold, it is determined to be in the descending stage. By scanning the value of the derivative function, the starting point and the ending point of the descending stage are determined, and these continuous descending parts are divided into different descending gradient segments. For example, in a continuous descent process, if the derivative is continuously less than the threshold in a certain interval, and the derivative changes from less than the threshold to greater than or equal to the threshold at both ends of the interval, then the original memory curve portion corresponding to this interval is a descending gradient segment. Different descending gradient segments may have different descending speeds, durations, etc., and this information is crucial for the subsequent analysis of the characteristics of memory changes. The starting and ending points of each descending gradient segment are the key nodes for curve segmentation. Through these nodes, the entire curve is divided into multiple different curve segments. Each curve segment contains rich memory information. For example, a curve segment with an obvious and gentle upward trend means that the trainee's memory is steadily improving during this stage, which may be due to the use of some effective memory method or being more proficient in specific types of memory tasks. A curve segment with a larger downward gradient indicates that the trainee's memory ability has declined significantly during this period.

[0027] Match the analysis information and training objectives with multiple algorithm containers to obtain a first algorithm container; the algorithm container includes an algorithm drawer and an algorithm capacity; the algorithm drawer includes at least one memory training algorithm, and the algorithm capacity corresponds to the difficulty coefficient and / or training duration of the memory training algorithm. The algorithm container is a framework for managing and scheduling algorithms, while the algorithm drawer is a specific algorithm implementation. The algorithm container selects a suitable algorithm drawer to execute according to specific conditions. The memory training algorithms constructed in the algorithm drawer may include: memory enhancement algorithm, forgetting curve adaptation algorithm, multivariate memory stimulation algorithm, adaptive challenge algorithm, etc.

[0028] The weights of different algorithm drawers in the algorithm container are determined according to the memory curve characteristics of the trainee, and the algorithm drawers in the algorithm container are sorted according to different algorithm weight values.

[0029] For example, for trainees with strong short-term memory, the memory enhancement algorithm is assigned a weight value of 0.4, the forgetting curve adaptation algorithm is assigned a weight value of 0.3, the multi-memory stimulation algorithm is assigned a weight value of 0.2, and the adaptive challenge algorithm is assigned a weight value of 0.1.

[0030] Specifically, the algorithm drawers in the algorithm container are sorted according to different algorithm weight values: they can be sorted from large to small, or they can be sorted in sections from small to large and then to small.

[0031] Furthermore, in this embodiment, matching the analysis information, the training target and the plurality of algorithm containers to obtain a first algorithm container includes: Identifying peak memory periods and low memory periods in the analysis information; Determine the training strategy according to the peak memory period, the trough memory period and the training goal; the training strategy may include formulating training strategies and specific parameters for the peak memory period and the trough memory period, such as formulating training strategies for the peak memory period, including rapid recall exercises, associative memory games, high-intensity repetitive training, etc., and setting specific parameters for training, such as training duration, training frequency, etc.; formulating training strategies for the trough period, including simple repetitive exercises, easy review, relaxing memory games, etc., and setting specific parameters for training; The training strategy is matched with a plurality of algorithm containers to obtain a first algorithm container.

[0032] Furthermore, in this embodiment, matching the analysis information and the training target with a plurality of algorithm containers to obtain a first algorithm container further includes: Monitoring the training results of the first algorithm container in real time; The algorithm container, algorithm drawer and algorithm capacity are dynamically adjusted according to the training results.

[0033] Further, the dynamically adjusting the algorithm container, the algorithm drawer and the algorithm capacity according to the training result includes: An initial number of algorithm drawers is selected from the first algorithm container for training; the algorithm container can be regarded as a collection of multiple algorithm drawers, each of which stores a different training algorithm. At the beginning of training, a suitable initial number of algorithm drawers is selected from the first algorithm container based on the initial situation and experience of the trainee.

[0034] If the algorithm capacity is insufficient, increase the number of algorithm drawers; during the training process, through the analysis of the training results, if it is found that the training content and intensity provided by the currently selected algorithm drawer cannot meet the training needs of the trainees, that is, the algorithm capacity is insufficient, the number of algorithm drawers needs to be increased. For example, after completing a certain stage of training, the trainees complete the current training task too easily, and the accuracy and speed are far beyond expectations, indicating that the training difficulty provided by the existing algorithm drawers is not enough. At this time, select more algorithm drawers from the first algorithm container to join the training to increase the diversity and difficulty of the training.

[0035] If the number of algorithm drawers exceeds the maximum value, select another algorithm container to continue training. Each algorithm container has its preset maximum number of algorithm drawers, which is based on the balance between system resources and training effect. Once the number of algorithm drawers added to meet training needs exceeds this maximum value during training, it means that the current first algorithm container can no longer effectively support training. At this time, you need to reselect a more suitable algorithm container from multiple algorithm containers to continue training.

[0036] Furthermore, the dynamically adjusting the algorithm container, the algorithm drawer and the algorithm capacity according to the training results further includes: Selecting the maximum capacity of an algorithm drawer from the first algorithm container for training; If the training duration or training results cannot reach the set value under the maximum capacity, an algorithm container with a larger capacity will be selected to continue training.

[0037] Performing memory training on the person to be trained according to the first algorithm container; Testing the trained personnel to obtain the second memory curve; Updating the algorithm container according to the comparison result between the first memory curve and the second memory curve includes: If the comparison result shows that the increase of the second memory curve relative to the first memory curve is greater than or equal to the increase threshold, the training difficulty coefficient and training duration of the first algorithm container are increased; here, the upward trend is obvious. If it is found that the curve of a certain memory type (such as short-term memory or long-term memory) has an obvious increase after training, and the increase is large, it means that the current training difficulty may be low and there is room for improvement. The algorithm difficulty and training duration can be appropriately increased; If the comparison result shows that the second memory curve has a smaller increase than the first memory curve than the increase threshold, the training difficulty coefficient and training duration of the first algorithm container are reduced; here, the upward trend is slow or flat. If the upward curve has a small increase or is basically flat, it means that the training difficulty may be high, and the progress is not obvious under the current difficulty. The algorithm difficulty should be appropriately reduced; If the comparison result shows that the second memory curve is decreasing relative to the first memory curve, the first algorithm container is replaced and retrained. If the curve decreases, it may be that the training intensity is too high or the training method is not suitable. It is necessary to significantly reduce the difficulty and re-evaluate the training plan.

[0038] Specific practical applications include: Trainees: Category: college students; Age: 20 years old; Gender: female; Major: law; Memory requirements: preparing for final exams, which requires memorizing a lot of legal provisions, case analysis, etc.

[0039] Initial Assessment, First Memory Curve: Short-term memory: In the initial assessment, users could only accurately recall 6 of 10 randomly displayed legal terms.

[0040] Long-term memory: For the legal provisions learned the day before, 40% of the total content can be accurately recalled 24 hours later.

[0041] Distraction: In the distraction task test, users were distracted an average of 3 times in 5 minutes, with each distraction lasting about 20 seconds.

[0042] First algorithm container: Algorithm drawer selection: According to the user's memory curve and training needs, the high-intensity training module in the memory enhancement algorithm is matched, especially for training during the user's peak memory period.

[0043] Algorithm capacity selection: High-intensity training: 45 minutes of quick recall exercises and associative memory games every morning from 8 to 10 (peak memory period for users), focusing on the memory of legal terms and clauses. Low-intensity training: 30 minutes of simple repetitive exercises and easy review every evening from 7 to 8 to consolidate what was learned during the day.

[0044] Training process: First week training: Users train according to the training plan and record their training performance and feelings every day. The system monitors the user's training data in real time, such as reaction time, accuracy, etc., and dynamically adjusts the training difficulty based on the data.

[0045] Second week training: Continue to train according to the training plan, and the system begins to collect user feedback, such as the difficulty of the training content, the training duration, etc. Based on user feedback and training data, the system fine-tunes the training plan, such as appropriately increasing the training duration and reducing the difficulty of some training tasks.

[0046] The training effect data is shown in Table 1. Table 1 Training effect data The memory training algorithm of this embodiment improves the short-term memory of trainees. After two weeks of training, the number of 10 randomly displayed legal terms that users can accurately recall has increased to 8, and short-term memory has improved by about 33%. The long-term memory of trainees has improved: for the legal provisions learned the day before, the content that can be accurately recalled after 24 hours accounts for 65% of the total content, and long-term memory has improved by about 50%. The distraction of trainees has improved: in the distraction task test, the average number of distractions of users in 5 minutes is reduced to 2 times, and the distraction time is shortened to 15 seconds each time, and the distraction is significantly improved.

[0047] The present invention obtains the first memory curve by conducting a memory test on the trainees, and analyzes the analysis information to accurately grasp the memory foundation and characteristics of each trainee. Based on this, the analysis information is matched with the training goal and the algorithm container, and the first algorithm container is selected for each trainee to achieve personalized memory training, avoiding the drawbacks of the traditional "one size fits all" training mode, making the training more in line with the actual situation of the individual, and significantly improving the pertinence and effectiveness of the training. After the training, the second memory curve is tested again, and the algorithm container is updated according to the comparison results of the first and second memory curves. This process realizes the dynamic adjustment of the training strategy, and the algorithm can be optimized in real time according to the effect of each training, ensuring that the training method always adapts to the ability improvement and changes of the trainees, continuously providing the most suitable training plan for the trainees, and constantly tapping the memory improvement potential of the trainees.

[0048] Embodiment 2 A memory graded training system, comprising: The first acquisition module: conducts a memory test on the trainee and obtains the first memory curve; Analysis module: analyzing the first memory curve to obtain analysis information; Matching module: matching the analysis information, training target and algorithm container to obtain a first algorithm container; Training module: performing memory training on the trainee according to the first algorithm container; The second acquisition module: testing the trained personnel to obtain the second memory curve; An updating module is configured to update the algorithm container according to a comparison result between the first memory curve and the second memory curve.

[0049] It should be understood that the above specific embodiments of the present invention are only used to illustrate or explain the principles of the present invention, and do not constitute a limitation of the present invention. Therefore, any modifications, equivalent substitutions, improvements, etc. made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. In addition, the appended claims of the present invention are intended to cover all changes and modifications that fall within the scope and boundaries of the appended claims, or the equivalent forms of such scope and boundaries.

[0050] The present invention has been described above with reference to the embodiments of the present invention. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention. The scope of the present invention is defined by the appended claims and their equivalents. Without departing from the scope of the present invention, a person skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

[0051] Although the embodiments of the present invention have been described in detail, it should be understood that the various changes, substitutions, and alterations could be made hereto without departing from the spirit and scope of the invention.

[0052] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A memory hierarchical training algorithm, characterized in that: include: Conduct memory tests on trainees to obtain the first memory curve; Analyzing the first memory curve to obtain analysis information; Matching the analysis information, the training target and a plurality of algorithm containers to obtain a first algorithm container; Performing memory training on the person to be trained according to the first algorithm container; Testing the trained personnel to obtain the second memory curve; updating the first algorithm container according to a comparison result between the first memory curve and the second memory curve; The algorithm container includes an algorithm drawer and an algorithm capacity.

2. The memory hierarchical training algorithm according to claim 1, characterized in that: The analyzing the first memory curve to obtain analysis information includes: Acquire a plurality of descending gradient segments of the first memory curve; The first memory curve is segmented according to a plurality of descending gradient segments to obtain memory curve segment information.

3. The memory hierarchical training algorithm according to claim 2 is characterized in that: The step of obtaining a plurality of descending gradient segments of the first memory curve comprises: Performing mathematical modeling on the first memory curve to obtain a function expression of the memory curve; calculating a derivative of the first memory curve to obtain a rate of change of the first memory curve; According to the change rate of the first memory curve, a descending gradient of the memory curve is determined, and the descending gradient is divided into a plurality of descending gradient segments.

4. The memory hierarchical training algorithm according to claim 1, characterized in that: The algorithm drawer includes at least one memory training algorithm, and the algorithm capacity corresponds to the difficulty coefficient and / or training duration of the memory training algorithm.

5. The memory hierarchical training algorithm according to claim 4 is characterized in that: Matching the analysis information and the training target with a plurality of algorithm containers to obtain a first algorithm container includes: Identifying peak memory periods and low memory periods in the analysis information; Determine a training strategy according to the peak memory period, the trough memory period and the training goal; The training strategy is matched with a plurality of algorithm containers to obtain a first algorithm container.

6. The memory hierarchical training algorithm according to claim 5 is characterized in that: Matching the analysis information and the training target with a plurality of algorithm containers to obtain a first algorithm container further includes: Monitoring the training results of the first algorithm container in real time; The algorithm container, algorithm drawer and algorithm capacity are dynamically adjusted according to the training results.

7. The memory hierarchical training algorithm according to claim 6, characterized in that: The dynamically adjusting the algorithm container, the algorithm drawer and the algorithm capacity according to the training results includes: Selecting an initial number of algorithm drawers from the first algorithm container for training; If the algorithm capacity is insufficient, increase the number of algorithm drawers; If the number of algorithm drawers exceeds the maximum value, other algorithm containers are selected to continue training.

8. The memory hierarchical training algorithm according to claim 6, characterized in that: The dynamically adjusting the algorithm container, the algorithm drawer and the algorithm capacity according to the training results also includes: Selecting the maximum capacity of an algorithm drawer from the first algorithm container for training; If the training duration or training results cannot reach the set value under the maximum capacity, an algorithm container with a larger capacity will be selected to continue training.

9. The memory hierarchical training algorithm according to claim 1, characterized in that: Updating the first algorithm container according to a comparison result between the first memory curve and the second memory curve includes: If the comparison result shows that the rising amplitude of the second memory curve relative to the first memory curve is greater than or equal to the rising amplitude threshold, the training difficulty coefficient and training duration of the first algorithm container are increased; If the comparison result shows that the increase of the second memory curve relative to the first memory curve is less than the increase threshold, the training difficulty coefficient and training duration of the first algorithm container are reduced; If the comparison result shows that the second memory curve shows a downward trend relative to the first memory curve, the first algorithm container is replaced and retrained.

10. A memory level training system according to the algorithm of claim 1, characterized in that: include: The first acquisition module: conducts a memory test on the trainee and obtains the first memory curve; Analysis module: analyzing the first memory curve to obtain analysis information; Matching module: matching the analysis information, training objectives and multiple algorithm containers to obtain a first algorithm container; Training module: performing memory training on the trainee according to the first algorithm container; The second acquisition module: testing the trained personnel to obtain the second memory curve; An updating module is configured to update the first algorithm container according to a comparison result between the first memory curve and the second memory curve.

Images