WO2022068435A1 - Method for measuring knowledge point mastering state - Google Patents
Method for measuring knowledge point mastering state Download PDFInfo
- Publication number
- WO2022068435A1 WO2022068435A1 PCT/CN2021/112675 CN2021112675W WO2022068435A1 WO 2022068435 A1 WO2022068435 A1 WO 2022068435A1 CN 2021112675 W CN2021112675 W CN 2021112675W WO 2022068435 A1 WO2022068435 A1 WO 2022068435A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- knowledge
- answering
- state
- question
- student
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 64
- 230000008859 change Effects 0.000 claims description 4
- 238000009826 distribution Methods 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 abstract description 5
- 230000006870 function Effects 0.000 description 33
- 238000004590 computer program Methods 0.000 description 23
- 230000008569 process Effects 0.000 description 17
- 238000003860 storage Methods 0.000 description 16
- 238000012545 processing Methods 0.000 description 8
- 238000012360 testing method Methods 0.000 description 8
- 208000025174 PANDAS Diseases 0.000 description 6
- 208000021155 Paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection Diseases 0.000 description 6
- 240000004718 Panda Species 0.000 description 6
- 235000016496 Panda oleosa Nutrition 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 230000008676 import Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000005314 correlation function Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
Definitions
- the present application relates to the application field of artificial intelligence and adaptive technology in education, for example, a method for measuring the mastery state of knowledge points.
- the process of students learning knowledge points is actually a process of continuously improving their mastery of knowledge points.
- students are generally allowed to practice in the order of easy first and then difficult.
- the difficulty of the questions is continuously increased, or the corresponding difficulty of the questions is provided according to the students' changing mastery state of knowledge points, so it is necessary to evaluate the students' mastery state of knowledge points.
- the students as a whole are usually pushed to the same topic to all students without considering the different learning situations of each student. This method is bound to appear.
- the difficulty of the questions is too low, and simply repeating exercises will result in low learning efficiency and poor effect; for students with poor grasp of knowledge points, Said that the difficulty of the topic is too high, not only does not have the effect of practice, but will dampen the enthusiasm of learning.
- the order of learning the contents of multiple knowledge points is determined in advance by the teacher or the lesson plan, and the different situations of each student are not considered.
- the methods of evaluating students' knowledge point mastery mainly include correct answer rate (that is, correct or incorrect answering questions), ability value (based on Item Response Theory (IRT)) and other indicators to measure students' knowledge point granularity. state of knowledge.
- IRT also known as Item Response Theory and Latent Trait Theory, is a general term for a series of psychostatistical models. IRT is a mathematical model used to analyze test scores or questionnaire data. The goal of these models is to determine whether underlying psychological characteristics can be passed through The test questions are reflected, and the interaction between the test questions and the testee.
- the ability value takes into account the correctness and difficulty and the difficulty, the obtained value is relatively discrete. There will be a large number of students' ability values concentrated on multiple values that are equal or close to each other, and it is difficult to distinguish the mastery state of different students' knowledge points. , the discrimination is poor.
- the above methods are more concerned with the correct answer rate, that is, the knowledge status of students is considered from the perspective of correctness and error, but it is not considered or difficult to comprehensively consider the difficulty of the question and the time to do the question.
- the present application provides a method for measuring the mastery state of knowledge points, which can comprehensively evaluate students' knowledge point mastery state from multiple perspectives, and can better distinguish students, which is conducive to more accurately locating students' ability levels, In this way, personalized learning paths can be recommended to students in a more targeted manner, which solves the problem of not distinguishing, considering, and locating students' knowledge status and ability level from multiple perspectives.
- the present application provides a method for measuring the mastery state of knowledge points, including:
- Fig. 1 is a function curve diagram of a sigmoid ( ⁇ ) provided by an embodiment of the present application
- KS Knowledge State
- FIG. 3 is a schematic flowchart of a method for measuring a knowledge point mastery state provided by an embodiment of the present application.
- the present application provides a method for measuring the mastery state of knowledge points, which may include the following steps:
- Step 1 Collect the answering time and answering results of students answering questions related to knowledge points.
- the answer result may include two states: correct and incorrect, with correct being 1 and being incorrect being 0.
- the answer result may also be represented by a score rate, which is the ratio of the actual score to the total score of the question.
- Step 2 Calculate the student's KS on the question according to the answering time and the answering result.
- Each question corresponds to a question difficulty, and the knowledge status can be calculated according to the answering time, answering result, and question difficulty.
- the question difficulty is calculated according to the distribution of the answering results of all the students who answered the question. The greater the difficulty of the question, the greater the value of the knowledge state. In some embodiments, the value range of the item difficulty is [0, 1], and the higher the value, the more difficult the item is.
- the knowledge state is calculated only according to the difficulty of the question; if the answer result is correct, the knowledge state is calculated according to the answering time and the difficulty of the question. The longer the answering time, the smaller the value of the knowledge state.
- the answering time may also be processed, so that as the answering time increases, the change trend of the value of the knowledge state is more gradual, that is, when the answering time is too long, the change of the answering time will affect the value of the knowledge state. The impact will gradually decrease.
- the knowledge state is calculated according to the multiple of the answering time relative to the estimated question-making time of the question and the answering result, wherein the estimated question-making time is the median answering time of all students who answered the question.
- the estimated question-making time is the median answering time of all students who answered the question.
- the knowledge state may also be calculated according to the reference state value, the answering time, and the answering result, wherein the reference state value represents the knowledge when a question with the lowest difficulty is answered correctly and the answering time for answering the question meets the preset duration condition
- the value of the state can be a preset fixed value, or a value that is calculated and updated and adjusted according to historical data.
- the reference state value is introduced, so that the calculated KS can take all the values in the range of [0, 1] without faults.
- a knowledge state greater than 0 can still be set for the student, because in the adaptive education system, the student can complete the question of higher difficulty, which means that the student is considered to have this problem based on historical performance. The ability to do a question of this difficulty, so even if the student makes a mistake on the question, a knowledge state greater than 0 will be set for the student.
- a student does a question of lower difficulty, it is likely that the system considers the student's ability to be suitable for a more basic question, so even if the student does the question correctly, the answering time is very fast, and the value of the knowledge status It will also be limited to a certain extent, not very high.
- Step 3 Take the average value of the KS of all the questions related to the knowledge point answered by the student as the student's comprehensive knowledge status (comprehensive KS) on the knowledge point, so as to measure the student's mastery of the knowledge point.
- the average value of the KS of all the questions related to the knowledge point completed and answered by the student is a weighted average, wherein each question corresponds to a weight. Questions whose completion date is closer to the current date have greater weights, which strengthens the impact of recent question performance and weakens the impact of long-term historical question performance, so that it can more accurately reflect the latest state and situation of students.
- a weighting sequence may be set according to the number of questions, and each weight in the sequence may be assigned to each question in a front-to-back order according to the order of answering dates from farthest to nearer.
- the weight sequence may be an arithmetic sequence whose sum is 1, and the value in the front is small and the value in the back is large.
- the weight sequence may also be a sequence of increasing difference, and the difference between the later two values is larger, so that the weight and influence of the recent topic can be strengthened.
- the average value of the students' comprehensive knowledge states on all knowledge points may be used as the student's overall knowledge state (Student Knowledge State, SKS), and the initial difficulty of pushing a question to the student is determined according to the SKS.
- SKS Student Knowledge State
- a student's SKS reflects the student's own ability level. When learning a new knowledge point, students can be pushed to the students with questions that match the SKS according to the difficulty of the question.
- the average value of the comprehensive knowledge state of all students under the knowledge point may be taken as the average knowledge state (Knowledge Point Knowledge State, KPKS) of the knowledge point, and the push to the student is determined according to the KPKS of the knowledge point and the SKS of the student order of knowledge points.
- KPKS Knowledge Point Knowledge State
- SKS the student order of knowledge points.
- the granularity of the knowledge points here can be very finely divided, each knowledge point can represent a small range of knowledge content, in a large learning content, there can be multiple such refined knowledge points, and learning these
- the order of refining knowledge points may not be fixed.
- the KS of a student on a question can be calculated according to the difficulty of the question, the answering time, the answering result, and the reference status value.
- the calculation formula is as follows:
- ⁇ represents the difficulty of the question, and the value range is [0,1]. The larger the ⁇ , the more difficult the question is.
- the difficulty of a question may be calculated according to the distribution of correct and incorrect answers under a question.
- ⁇ represents the relative time, that is, the ratio of the actual answering time (real_time) of the student to the estimated answering time (estimates_time) of the question. The smaller the ⁇ , the faster the student will answer the question.
- the estimated time for answering the question may be calculated according to the median duration or the average duration of the answering time of the question in the question-making history data.
- ⁇ is correct or incorrect, 1 is correct and 0 is incorrect.
- ⁇ is the reference state value, which indicates the KS value when the question with the lowest difficulty is done correctly and the time spent is long enough.
- the reason why the student made a wrong question and the student's KS value is still greater than 0 is because in the adaptive system, the student can complete the question of higher difficulty. Even if the student gets the question wrong, a knowledge status greater than 0 will be set for the student.
- KS is not affected by the answering time when the question is wrong. If the student does a question of lower difficulty, it is likely that the system believes that the student's ability is suitable for doing the more basic question. Therefore, even if the student does the question correctly and the answering time is fast, the KS value will be limited to a certain extent. It can be seen that this method is directly related to adaptive learning.
- FIG. 1 is a function curve diagram of a sigmoid ( ⁇ ) provided by an embodiment of the present application.
- the abscissa represents ⁇ (that is, the relative answering time, Time Ratio), and the ordinate represents the value of sigmoid ( ⁇ ).
- ⁇ that is, the relative answering time, Time Ratio
- ⁇ the initial value of sigmoid( ⁇ ) is 0.5; as ⁇ increases, that is, with the increase of answering time, the value of sigmoid( ⁇ ) gradually increases, but the growth rate increases. slowed down and gradually approached 1.
- FIG. 2 is a function curve diagram of a KS value provided by an embodiment of the present application.
- the abscissa represents ⁇ (that is, the relative answering time, Time Ratio), and the ordinate represents the value of KS.
- ⁇ that is, the relative answering time, Time Ratio
- KS the relative answering time
- the student's comprehensive knowledge status (comprehensive KS) on a knowledge point can be calculated to measure the student's knowledge point. Master the state.
- the comprehensive KS of the knowledge point may be obtained by assigning a weight to each question under the knowledge point, and then weighted average of the KS values of each question under the knowledge point of the student.
- a neural network may be used to iteratively converge through back-propagation and gradient descent methods to finally obtain the weights.
- an arithmetic progression with a sum of 1 can also be used as the weight of the topic under the knowledge point, and the calculation method is as follows:
- S k is the sum of the first k items of the arithmetic sequence
- k is a natural number of [1,n]
- n is the number of questions answered under the knowledge point (that is, how many questions need to be weighted)
- d is the arithmetic sequence
- a 1 is the first item of the arithmetic sequence (that is, the weight of the first question)
- a k is the kth item of the arithmetic sequence (that is, the kth item of the question).
- Weights weights
- a 1 0.0278
- the weights of the 8 questions from far to near are [0.0278, 0.0556, 0.0833, 0.1111, 0.1389, 0.1667, 0.1944, 0.2222].
- the student's comprehensive KS on the knowledge point is the sum of the two arrays Dot multiply and add to get a combined KS value of 0.45.
- the weights are assigned by the above arithmetic sequence, and the weight of each question can be directly obtained only by knowing the number of questions n.
- This method also strengthens the influence of the performance of recent items (the date of answering the item is more recent), while weakening the weight of the performance of older historical items (the date of answering the item is smaller, the weight is smaller), which can effectively reflect the students The latest performance and the latest knowledge status.
- this method is simple and convenient, as long as the relevant parameters of the sequence that need to be weighted are input, the required weight sequence can be returned immediately, and this method also naturally strengthens the weight of the latest answers, which is in line with the actual learning situation of the students. .
- the above method of assigning weights through arithmetic progression can also be used in the scheme of assigning the weights of students' multiple unit test scores, that is, according to this method, the weights of students' multiple unit test scores are calculated chronologically, and then A weighted average of multiple unit test scores is calculated to obtain a composite score that reflects a student's recent performance.
- FIG. 3 is a schematic flowchart of a method for measuring a knowledge point mastery state provided by an embodiment of the present application, and an example is used for description.
- Step 1 as shown in Table 1, pull the correctness (is_right), difficulty (difficulty), estimated time (estimates_time), actual error (is_right), difficulty (difficulty), estimated time (estimates_time), actual Time (cost_time), start time (create_time) and other fields, and the start time (the date of the start of the topic + time) is sorted in ascending order (order by) as a standard to ensure that the topic records are displayed from old to new , and then export to csv file.
- Step 2 import the above csv file using pandas' read csv file function (read_csv) in Python, and obtain the data frame (DataFrame) format in pandas as shown in Table 2.
- read_csv pandas' read csv file function
- DataFrame data frame format in pandas as shown in Table 2.
- Map mapping function
- user_id tag_code question_id is_right difficulty estimates_time cost_time create_time knowledge_state 33261 3093 26375 0 0.5 160 39 2019/10/26 13:53:45 0.1000 33261 3093 27171 1 0.8 180 158 2019/10/27 9:56:49 0.8841 33261 3093 21342 1 0.2 30 6 2019/11/1 13:12:43 0.5702 33261 3093 21793 1 0.3 120 twenty four 2019/11/2 16:26:13 0.6367 33261 3093 17221 0 0.6 120 112 2019/11/9 12:20:42 0.1200 33261 3093 15157 1 0.5 120 34 2019/12/1 11:03:06 0.7472 33261 3093 19738 0 0.5 160 28 2019/12/6 19:48:10 0.1000 33261 3093 20306 1 0.2 90 30 2019/12/7 16:25:51 0.5515 47162 7884 18358 1 0.4 150 54 2019/12/14 11:03:37 0.6803 47162 7884
- Step 3 as shown in Table 3, use pandas' grouping statistical function (groupby) to group the data table by student account number (user_id) and knowledge point number (tag_code), and use the aggregation function (agg) to group people + knowledge point granularity.
- groupby pandas' grouping statistical function
- agg aggregation function
- the following series of KS values are combined into the form of a list (list), and then use the reset index function (reset_index) to reset the fields and index of the table.
- Step 4 use a self-defined weighted knowledge state function (weighted_knowledge_state) to weight and aggregate KS at the granularity of people + knowledge points according to the arithmetic sequence.
- the custom function only takes the sequence of the KS list form on the topic shown in step 3, such as [0.83, 0.72, 0.69, 0.23, 0.8] as the input parameter, to obtain the n value of the arithmetic sequence, thereby generating the arithmetic difference
- the KS derived from this step is updated every 24 hours to capture the student's latest ability level.
- Step 5 Take the mean value of the KS of all knowledge points of each student as the student's SKS. As described in step 4, update the SKS every 24 hours, remove the old data to obtain new data, and obtain the student's latest ability level, thereby correspondingly. Update the initial difficulty value of the student's subsequent questions. For example, the updated overall SKS values of the three students are 0.82, 0.57, and 0.35, respectively, then the initial difficulty of pushing the questions to the students on the knowledge points on the second day will be different according to a certain mapping method (such as 0.8, 0.5, 0.2) to adapt to The level of different students reflects the adaptability of the system (while the initial difficulty in the traditional method is the same, such as 0.5).
- a certain mapping method such as 0.8, 0.5, 0.2
- Step 6 Take the mean value of students' KS under each knowledge point as the KPKS of the knowledge point, update the KPKS every 24 hours, and get the latest KPKS of the knowledge point.
- the knowledge points are pushed from small to large according to the absolute value of the difference between KPKS and SKS, that is, those with similar KS values are pushed first.
- a student whose SKS is 0.67 has four knowledge points A, B, C, and D with KPKS of 0.82, 0.7, 0.45, and 0.58 to learn, then the learning order of knowledge points is B, D, A, C ( The traditional method is to push according to certain rules or order, which cannot reflect the adaptability).
- Python is used as a tool, and the Python libraries used are: Pandas, Numpy.
- Numpy is a library that provides a variety of complex mathematical operations in Python.
- Pandas is a library for data analysis and processing through data frames. The data is pulled from the PostgreSQL database.
- part of the code is as follows:
- the sigmoid function is a nonlinear function, and is placed in the exponential position rather than simply multiplied by ( ⁇ +(1- ⁇ )* ⁇ ). Using this feature, the value of KS can not always be greatly reduced due to the increase of relative time, but can be gradually reduced gradually. When the relative time is large to a certain extent, the impact on KS is also very small.
- the sigmoid function can effectively control the value range of the relative time between (0.5, 1), and adjust KS appropriately. That is, if the students do the questions fast, they can raise the KS value by taking the base value to the power of 0.5; if the students do the questions slowly, the base value is basically not raised.
- the KS is obtained by weighting the arithmetic sequence weight.
- the weighted_knowledge_state function is simple, convenient and quick to input parameters, which can play a role in strengthening the recent performance of students.
- Other sequences can also be used, and the weight sequence is a sequence in which the sum is equal to 1, the weight value in the front of the two adjacent weights is small, and the weight value in the back is large, and the difference between the two adjacent weights is gradually increasing or equal.
- this indicator can be used to deduce questions and help students find knowledge loopholes and plan learning paths in a targeted manner. If the average KS of the students on the knowledge points is very high (eg 0.8), it means that the students have done quickly and accurately on the knowledge points, especially the last few questions, and the ability is very strong. Then the difficulty of the subsequent push questions will become larger to adapt to the latest ability level of students. Subsequent push knowledge points will also give priority to knowledge points with similar KS values.
- the technical solutions provided in this application may be systems, methods, apparatuses and/or computer program products.
- the computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement aspects of the present application.
- the present application also provides a computer apparatus, device or terminal.
- the computer device, device or terminal includes a processor, memory, network interface, display screen and input device connected by a system bus.
- the processor is configured to provide computing and control capabilities
- the memory includes a non-volatile storage medium and an internal memory.
- the nonvolatile storage medium stores an operating system and a computer program.
- Internal memory provides an environment for the execution of the operating system and computer programs in the non-volatile storage medium.
- the network interface is configured to communicate with external terminals through a network connection.
- the display screen can be a liquid crystal display screen or an electronic ink display screen
- the input device can be a touch layer covered on the display screen, a button, a trackball or a touchpad set on the casing, or an external keyboard, touch board or mouse, etc.
- a computer program can be divided into one or more modules or units, and these modules or units are stored in a memory and executed by a processor to implement the technical solutions of the present application.
- These modules or units may be a series of computer program instruction segments capable of accomplishing specific functions, and the instruction segments are used to describe the execution process of the computer program in an apparatus, device or terminal.
- the above-mentioned apparatus, device or terminal may be a desktop computer, a notebook, a mobile electronic device, a palmtop computer, a cloud server and other computing devices.
- the structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the device, equipment or terminal to which the solution of the present application is applied. shown in more or less components, or in combination of some components, or with different component arrangements.
- the processor can be a central processing unit (CPU), or other general-purpose or special-purpose processors, microprocessors, digital signal processors (DSPs), application specific integrated circuits (Application Specific Integrated Circuits) , ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
- the processor is the control center of the above-mentioned apparatus, equipment or terminal, and uses various interfaces and lines to connect various parts of the apparatus, equipment or terminal.
- the memory can be configured to store computer programs, modules and data, and the processor implements various functions of the apparatus, device or terminal by running or executing the computer programs and/or modules stored in the memory and calling the data stored in the memory.
- the memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; Created various types of data (such as multimedia data, documents, operation history, etc.) and so on.
- the memory may include high-speed random access memory, and may also include non-volatile memory such as hard disk, internal memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card , Flash Card, magnetic disk storage device, flash memory device, or other volatile solid-state storage device.
- non-volatile memory such as hard disk, internal memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card , Flash Card, magnetic disk storage device, flash memory device, or other volatile solid-state storage device.
- the present application also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above method are implemented. All or part of the process in the method of the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium, and the computer program is stored in a non-volatile computer-readable storage medium. When executed, it may include the flow of the embodiment of the above-mentioned method. Wherein, any reference to memory, storage, database or other medium used in the embodiments provided in this application may include non-volatile and/or volatile memory.
- Non-volatile memory may include read-only memory (Read-Only Memory, ROM), programmable ROM (Programmable ROM, PROM), electrically programmable ROM (Electrical PROM, EPROM), electrically erasable programmable ROM (Electrically Erasable ROM) PROM, EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory.
- RAM is available in various forms, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (Double Data) Rate SDRAM, DDR SDRAM), enhanced SDRAM (Enhanced SDRAM, ESDRAM), synchronous link DRAM (Synchlink DRAM, SLDRAM), memory bus direct RAM (Rambus Direct RAM, RDRAM), direct memory bus dynamic RAM (Direct Rambus DRAM, DRDRAM) ), and memory bus dynamic RAM (Rambus DRAM, RDRAM), etc.
- the integrated modules and units of the apparatus or terminal equipment are implemented in the form of software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium.
- the implementation of all or part of the processes in the various methods disclosed in the present application can also be completed by instructing relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program is in When executed by a processor, the steps of the various methods described above may be implemented.
- the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate forms, and the like.
- Computer readable media may include: any entity or device capable of carrying computer program code, recording media, USB flash drives, removable hard disks, magnetic disks, optical discs, computer memory, ROM, RAM, electrical carrier signals, telecommunication signals, and software distribution media, etc. .
- Computer readable media may contain suitable additions or deletions as required by legislation and patent practice in jurisdictions.
- the various methods, processes, modules, apparatus, devices, or systems disclosed herein may be implemented in one or more processing devices (eg, digital processors, analog processors, devices designed and arranged to process information) implemented or implemented in digital circuits, analog circuits designed to process information, state machines, computing devices, computers, and/or other mechanisms configured to process information electronically).
- the one or more processing devices may include one or more devices that perform some or all of the operations of the method in response to instructions electronically stored on an electronic storage medium.
- the one or more processing means may comprise one or more means specially designed for carrying out one or more operations of the method configured by hardware, firmware and/or software.
- Embodiments of the present application may be implemented in hardware, firmware, software, or various combinations thereof, and may also be implemented as instructions stored on a machine-readable medium that may be read and executed using one or more processing devices.
- a machine-readable medium can include various mechanisms for storing and/or transmitting information in a form readable by a machine (eg, a computing device).
- machine-readable storage media may include read-only memory, random-access memory, magnetic disk storage media, optical storage media, flash memory devices, and other media configured to store information
- machine-readable transmission media may include a variety of forms Propagated signals (including carrier waves, infrared signals, digital signals) and other media configured to transmit information.
- firmware, software, routines, or instructions may be described in the above disclosure in terms of certain exemplary aspects and implementations that perform some actions, it will be apparent that such descriptions are for convenience only and that such actions are practical generated by a machine device, computing device, processing device, processor, controller, or other device or machine executing firmware, software, routines or instructions.
- a module for performing a specified function is intended to encompass any means capable of performing the function, such as a combination of circuit elements that perform the function, a module used to perform or implement the function Software, hardware, and a combination of software and hardware, or any form of software, firmware, code, and their combination with suitable circuitry or other means.
- the functions provided by the various modules are combined together in the manner claimed herein, and it is therefore intended that any module, component, element that provides the functions is equivalent or equivalent to the modules defined in the claims.
- the circuit structure of some embodiments in this application can also be changed or modified, for example, the current source is transformed into a voltage source, the series structure is transformed into a parallel structure, etc., so as to obtain more diversified Examples, but these changes and modifications belong to the scope of the disclosure of the present application.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Disclosed is a method for measuring the knowledge point mastering state, comprising: acquiring the answering time and the answering results of titles relevant to knowledge points and answered by a student; calculating the knowledge state of the student on the titles according to the answering time and the answering results; and using the average value of the knowledge state of all the titles relevant to the knowledge points and completed answering by the student as the comprehensive knowledge state of the student on the knowledge points for measuring the knowledge point mastering state of the student. In this way, personalized adaptive education can be achieved.
Description
本申请要求在2020年09月30日提交中国专利局、申请号为202011059182.1的中国专利申请的优先权,该申请的全部内容通过引用结合在本申请中。This application claims the priority of the Chinese patent application with application number 202011059182.1 filed with the China Patent Office on September 30, 2020, the entire contents of which are incorporated into this application by reference.
本申请涉及人工智能、自适应技术在教育上的应用领域,例如涉及一种对知识点掌握状态进行测量的方法。The present application relates to the application field of artificial intelligence and adaptive technology in education, for example, a method for measuring the mastery state of knowledge points.
学生学习知识点的过程,实际上是对知识点的掌握状态不断提升的过程。在学习的过程中,例如通过做练习题来进行学习的过程中,一般都遵循先易后难的顺序让学生进行练习,实际上就是将练习题的难度与学生对知识点的掌握状态相匹配,随着掌握状态的提升不断地提高题目的难度,或者说根据学生针对知识点的不断变化的掌握状态提供相对应难度的题目,因此就需要评估学生对知识点的掌握状态。The process of students learning knowledge points is actually a process of continuously improving their mastery of knowledge points. In the process of learning, for example, in the process of learning by doing practice questions, students are generally allowed to practice in the order of easy first and then difficult. , with the improvement of the mastery state, the difficulty of the questions is continuously increased, or the corresponding difficulty of the questions is provided according to the students' changing mastery state of knowledge points, so it is necessary to evaluate the students' mastery state of knowledge points.
在传统的教学过程中,通常是将学生作为一个整体而向所有的学生推送相同的题目,而不会考虑每一个学生的不同学习情况。这种方式势必会出现,对于知识点的掌握状态较好的学生来说,题目难度过低,简单地重复练习,学习效率较低、效果较差;对于知识点的掌握状态较差的学生来说,题目难度过高,不仅没有练习效果,反而会挫伤学习的积极性。并且,在传统的教学过程中,学习多个知识点内容的顺序是由老师或者教案事先确定的,也不会考虑每一个学生的不同情况。In the traditional teaching process, the students as a whole are usually pushed to the same topic to all students without considering the different learning situations of each student. This method is bound to appear. For students who have a good grasp of knowledge points, the difficulty of the questions is too low, and simply repeating exercises will result in low learning efficiency and poor effect; for students with poor grasp of knowledge points, Said that the difficulty of the topic is too high, not only does not have the effect of practice, but will dampen the enthusiasm of learning. Moreover, in the traditional teaching process, the order of learning the contents of multiple knowledge points is determined in advance by the teacher or the lesson plan, and the different situations of each student are not considered.
评估学生的知识点掌握状态的方式主要有正答率(即作答题目正确或错误的情况)、能力值(基于项目反应理论(Item Response Theory,IRT))等指标来衡量学生在知识点粒度上的知识掌握状态。IRT又称题目反应理论、潜在特质理论,是一系列心理统计学模型的总称,IRT是用来分析考试成绩或者问卷调查数据的数学模型,这些模型的目标是来确定潜在的心理特征是否可以通过测试题被反应出来,以及测试题和被测试者之间的互动关系。The methods of evaluating students' knowledge point mastery mainly include correct answer rate (that is, correct or incorrect answering questions), ability value (based on Item Response Theory (IRT)) and other indicators to measure students' knowledge point granularity. state of knowledge. IRT, also known as Item Response Theory and Latent Trait Theory, is a general term for a series of psychostatistical models. IRT is a mathematical model used to analyze test scores or questionnaire data. The goal of these models is to determine whether underlying psychological characteristics can be passed through The test questions are reflected, and the interaction between the test questions and the testee.
与学生的知识点掌握状态相关的衡量指标存在以下问题:There are the following problems in the measurement indicators related to students' knowledge point mastery status:
(1)正答率仅仅考虑正误情况,考虑的维度太少,不够全面。(1) The correct answer rate only considers correct and incorrect situations, and considers too few dimensions and is not comprehensive enough.
(2)能力值虽然考虑了正误和难度,但其得出的值比较离散,会有大量的学生的能力值集中在相等或接近的多个值上,难以区分开不同学生的知识点掌 握状态,区分度较差。(2) Although the ability value takes into account the correctness and difficulty and the difficulty, the obtained value is relatively discrete. There will be a large number of students' ability values concentrated on multiple values that are equal or close to each other, and it is difficult to distinguish the mastery state of different students' knowledge points. , the discrimination is poor.
(3)虽然可以根据正答率来调整之后向学生推送的题目的难度,但给学生推送题目的初始难度都是一样的(尚未作答该知识点的题目,正答率指标为空),难以个性化定制学生的初始推题难度。(3) Although the difficulty of the questions pushed to the students can be adjusted according to the correct answer rate, the initial difficulty of the questions pushed to the students is the same (the questions of the knowledge point have not yet been answered, and the correct answer rate indicator is empty), which is difficult to personalize. Customize the difficulty of the initial push question for students.
(4)当需要向学生推送新的知识点进行学习时,以上方法都是按照事先确定的顺序或规则进行推送,难以适应学生的个性化的学习要求。(4) When it is necessary to push new knowledge points to students for learning, the above methods are all pushed according to the predetermined order or rules, which are difficult to adapt to the individual learning requirements of students.
以上方法更多考虑的是正答率,即从正误角度考虑学生的知识状态,但未考虑到或难以综合地考虑题目难度、做题时间等角度。The above methods are more concerned with the correct answer rate, that is, the knowledge status of students is considered from the perspective of correctness and error, but it is not considered or difficult to comprehensively consider the difficulty of the question and the time to do the question.
发明内容SUMMARY OF THE INVENTION
本申请提供一种对知识点掌握状态进行测量的方法,能够从多角度全面地评估学生的知识点掌握状态,并且能够较好地对学生进行区分,有利于更准确地定位学生的能力水平,从而能够更有针对性地向学生推荐个性化的学习路径,解决了没有从多个角度区分、考虑、定位学生的知识状态和能力水平的问题。The present application provides a method for measuring the mastery state of knowledge points, which can comprehensively evaluate students' knowledge point mastery state from multiple perspectives, and can better distinguish students, which is conducive to more accurately locating students' ability levels, In this way, personalized learning paths can be recommended to students in a more targeted manner, which solves the problem of not distinguishing, considering, and locating students' knowledge status and ability level from multiple perspectives.
本申请提供一种对知识点掌握状态进行测量的方法,包括:The present application provides a method for measuring the mastery state of knowledge points, including:
采集学生作答与知识点相关的题目的作答时间和答题结果;根据作答时间和答题结果,计算学生在题目上的知识状态;将学生完成作答的与知识点相关的所有题目的知识状态的平均值,作为学生在知识点上的综合知识状态,以实现测量学生对知识点的掌握状态。Collect the answering time and answering results of students answering questions related to knowledge points; calculate the students' knowledge status on the questions according to the answering time and answering results; take the average value of the knowledge statuses of all questions related to knowledge points answered by students , as the comprehensive knowledge status of students on knowledge points, in order to measure students' mastery status of knowledge points.
图1是本申请实施例提供的一种sigmoid(β)的函数曲线图;Fig. 1 is a function curve diagram of a sigmoid (β) provided by an embodiment of the present application;
图2是本申请实施例提供的一种知识状态(Knowledge State,KS)值的函数曲线图;2 is a function curve diagram of a knowledge state (Knowledge State, KS) value provided by an embodiment of the present application;
图3是本申请实施例提供的一种对知识点掌握状态进行测量的方法的流程示意图。FIG. 3 is a schematic flowchart of a method for measuring a knowledge point mastery state provided by an embodiment of the present application.
下面将对本申请实施例中的技术方案进行描述。本申请可以通过许多不同形式的实施例来得以体现,本申请的保护范围并非仅限于文中提到的实施例。The technical solutions in the embodiments of the present application will be described below. The present application can be embodied in many different forms of embodiments, and the protection scope of the present application is not limited to the embodiments mentioned herein.
本申请中所使用的单数形式的“一种”、“所述”和“该”等描述旨在同时包括单数或多数的形式,除非上下文表示其它含义。“多种”或“多个”等一般包含至少两种或至少两个。本申请中使用的术语“和/或”仅仅是一种描述关联对象的关联关 系,表示可以存在三种关系,例如,A和/或B,可以表示:单独存在A、同时存在A和B、单独存在B这三种情况。另外,本申请中字符“/”,一般表示前后关联对象是一种“或”的关系。As used in this application, the singular forms "a", "the" and "the" are intended to include both the singular and the plural unless the context dictates otherwise. "Plurality" or "plurality" and the like generally include at least two or at least two. The term "and/or" used in this application is only an association relationship to describe associated objects, indicating that there may be three kinds of relationships, for example, A and/or B, which may indicate: the existence of A alone, the existence of A and B at the same time, There are three cases of B alone. In addition, the character "/" in this application generally indicates that the related objects are an "or" relationship.
术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素,而且还包括没有列出的其他要素,或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下,由语句“包括一个……”所限定的要素,并不排除在包括该要素的过程、方法、物品或者设备中还存在另外的相同要素。The terms "comprising", "comprising" or any other variation thereof are intended to encompass a non-exclusive inclusion such that a process, method, article or device comprising a list of elements includes not only those elements, but also other elements not listed , or elements inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.
在本申请的技术方案中,提出了KS的概念,以测量和区分学生对知识点的掌握状态。In the technical solution of this application, the concept of KS is proposed to measure and distinguish students' mastery of knowledge points.
本申请提供一种对知识点掌握状态进行测量的方法,可以包括以下步骤:The present application provides a method for measuring the mastery state of knowledge points, which may include the following steps:
步骤一,采集学生作答与知识点相关的题目的作答时间和答题结果。在一些实施例中,答题结果可以包括正确和错误两种状态,正确为1,错误为0。在另一些实施例中,答题结果也可以通过得分率来表示,得分率为实际得分占该题目总分的比例。Step 1: Collect the answering time and answering results of students answering questions related to knowledge points. In some embodiments, the answer result may include two states: correct and incorrect, with correct being 1 and being incorrect being 0. In other embodiments, the answer result may also be represented by a score rate, which is the ratio of the actual score to the total score of the question.
步骤二,根据作答时间和答题结果,计算学生在题目上的KS。Step 2: Calculate the student's KS on the question according to the answering time and the answering result.
每一个题目对应一个题目难度,可以根据作答时间、答题结果以及题目难度计算知识状态,其中,题目难度是根据所有作答该题目的学生的答题结果的分布统计出来的。题目难度越大,则知识状态的值越大。在一些实施例中,题目难度的值域范围为[0,1],值越高题目越难。Each question corresponds to a question difficulty, and the knowledge status can be calculated according to the answering time, answering result, and question difficulty. The question difficulty is calculated according to the distribution of the answering results of all the students who answered the question. The greater the difficulty of the question, the greater the value of the knowledge state. In some embodiments, the value range of the item difficulty is [0, 1], and the higher the value, the more difficult the item is.
在一些实施例中,如果答题结果为错误,则仅根据题目难度计算知识状态;如果答题结果为正确,则根据作答时间和题目难度计算知识状态。作答时间越长,则知识状态的值越小。In some embodiments, if the answer result is wrong, the knowledge state is calculated only according to the difficulty of the question; if the answer result is correct, the knowledge state is calculated according to the answering time and the difficulty of the question. The longer the answering time, the smaller the value of the knowledge state.
在一些实施例中,还可以对作答时间进行处理,以使得随着作答时间的增大,知识状态的值的变化趋势越平缓,即作答时间过长时,作答时间变化而对知识状态的值的所产生的影响会逐渐减小。In some embodiments, the answering time may also be processed, so that as the answering time increases, the change trend of the value of the knowledge state is more gradual, that is, when the answering time is too long, the change of the answering time will affect the value of the knowledge state. The impact will gradually decrease.
在一些实施例中,根据作答时间相对于题目的预估做题时间的倍数以及答题结果,计算知识状态,其中,预估做题时间为作答该题目的所有学生的作答时间的中位数。在实际作答时间接近预估做题时间的范围时,知识状态的值的变化幅度较大,区分度较高;而在实际作答时间远远超过预估做题时间的范围时,知识状态的值的变化幅度较小。In some embodiments, the knowledge state is calculated according to the multiple of the answering time relative to the estimated question-making time of the question and the answering result, wherein the estimated question-making time is the median answering time of all students who answered the question. When the actual answering time is close to the range of the estimated test time, the value of the knowledge state varies greatly and the degree of discrimination is high; and when the actual answer time is far beyond the range of the estimated test time, the value of the knowledge state changes are small.
在一些实施例中,还可以根据参考状态值、作答时间以及答题结果计算知识状态,其中,参考状态值表示做对一道难度最低的题目且作答该题目的作答 时间满足预设时长条件时的知识状态的值。参考状态值可以是预先设定的固定值,也可以是根据历史做题数据统计并更新调整的值。在答题结果正确和错误时,都引入参考状态值,可以使得计算得到的KS能够取到值域[0,1]范围的所有值,不会出现断层。In some embodiments, the knowledge state may also be calculated according to the reference state value, the answering time, and the answering result, wherein the reference state value represents the knowledge when a question with the lowest difficulty is answered correctly and the answering time for answering the question meets the preset duration condition The value of the state. The reference state value can be a preset fixed value, or a value that is calculated and updated and adjusted according to historical data. When the answer result is correct or incorrect, the reference state value is introduced, so that the calculated KS can take all the values in the range of [0, 1] without faults.
由于在做错题目的情况下,影响作答时间长短的原因较为复杂且通常不影响最终的分析判断,故在一些实施例中,答题结果为错误时,可以不考虑作答时间对KS的影响。In the case of wrong answers, the reasons affecting the answering time are complex and usually do not affect the final analysis and judgment. Therefore, in some embodiments, when the answering result is wrong, the impact of answering time on KS may not be considered.
在一些实施例中,即使答题结果为错误,也仍然可以给学生设置大于0的知识状态,因为在自适应教育系统中,学生能做到较高难度题目,可以说明基于历史表现认为学生有这个能力做这种难度的题目,所以即使学生将该题目做错,也会给学生设置大于0的知识状态。In some embodiments, even if the answer result is wrong, a knowledge state greater than 0 can still be set for the student, because in the adaptive education system, the student can complete the question of higher difficulty, which means that the student is considered to have this problem based on historical performance. The ability to do a question of this difficulty, so even if the student makes a mistake on the question, a knowledge state greater than 0 will be set for the student.
在一些实施例中,如果学生做到难度较低的题,很大可能说明系统认为学生的能力适合做较基础的题,因此即使学生将该题做对,作答时间很快,知识状态的值也会受到一定限制,不会很高。In some embodiments, if a student does a question of lower difficulty, it is likely that the system considers the student's ability to be suitable for a more basic question, so even if the student does the question correctly, the answering time is very fast, and the value of the knowledge status It will also be limited to a certain extent, not very high.
步骤三,将学生完成作答的与知识点相关的所有题目的KS的平均值,作为学生在知识点上的综合知识状态(综合KS),以实现测量学生对知识点的掌握状态。Step 3: Take the average value of the KS of all the questions related to the knowledge point answered by the student as the student's comprehensive knowledge status (comprehensive KS) on the knowledge point, so as to measure the student's mastery of the knowledge point.
在一些实施例中,学生完成作答的与知识点相关的所有题目的KS的平均值为加权平均值,其中,每一个题目对应一个权重。完成作答的日期越靠近当前日期的题目所对应的权重越大,即加强了最近题目表现的影响,而削弱了久远历史题目表现的影响,从而能够更准确地反映学生最新的状态和情况。In some embodiments, the average value of the KS of all the questions related to the knowledge point completed and answered by the student is a weighted average, wherein each question corresponds to a weight. Questions whose completion date is closer to the current date have greater weights, which strengthens the impact of recent question performance and weakens the impact of long-term historical question performance, so that it can more accurately reflect the latest state and situation of students.
在一些实施例中,可以根据题目的数量设定权重序列,并根据答题日期从远到近的顺序,将序列中的每一个权重按照从前向后的顺序赋给每一个题目。作为一个实施例中,权重序列可以是一个和为1的等差数列,前面值小后面值大。在另一些实施例中,权重序列也可以是差递增的数列,越后面的两个值之间的差越大,从而能够加强近期题目的权重和影响。In some embodiments, a weighting sequence may be set according to the number of questions, and each weight in the sequence may be assigned to each question in a front-to-back order according to the order of answering dates from farthest to nearer. As an embodiment, the weight sequence may be an arithmetic sequence whose sum is 1, and the value in the front is small and the value in the back is large. In some other embodiments, the weight sequence may also be a sequence of increasing difference, and the difference between the later two values is larger, so that the weight and influence of the recent topic can be strengthened.
在一些实施例中,可以将学生在所有知识点上的综合知识状态的平均值,作为学生的总体知识状态(Student Knowledge State,SKS),根据SKS确定向学生推送题目的初始难度。学生的SKS反映了学生本身的能力水平,当学习一个新的知识点时,可以根据题目的难度向学生推送与SKS相匹配的题目。In some embodiments, the average value of the students' comprehensive knowledge states on all knowledge points may be used as the student's overall knowledge state (Student Knowledge State, SKS), and the initial difficulty of pushing a question to the student is determined according to the SKS. A student's SKS reflects the student's own ability level. When learning a new knowledge point, students can be pushed to the students with questions that match the SKS according to the difficulty of the question.
在一些实施例中,可以将知识点下所有学生的综合知识状态的平均值,作为知识点的平均知识状态(Knowledge Point Knowledge State,KPKS),根据知识点的KPKS和学生的SKS确定向学生推送知识点的顺序。当有多个相关的知 识点需要学生学习时,可以根据KPKS和SKS,确定知识点的学习顺序,先学习与SKS最接近的知识点。此处的知识点的粒度可以划分得很细,每一个知识点可以代表一个很小范围的知识内容,在一个大的学习内容中,可以存在多个这类细化的知识点,而学习这些细化知识点的顺序可以是不固定的,通过根据不同学生的不同知识状态确定不同的知识点学习顺序,可以有效地提升学习效率和学习效果,实现自适应的个性化学习。In some embodiments, the average value of the comprehensive knowledge state of all students under the knowledge point may be taken as the average knowledge state (Knowledge Point Knowledge State, KPKS) of the knowledge point, and the push to the student is determined according to the KPKS of the knowledge point and the SKS of the student order of knowledge points. When there are multiple related knowledge points that students need to learn, they can determine the learning order of knowledge points according to KPKS and SKS, and learn the knowledge points closest to SKS first. The granularity of the knowledge points here can be very finely divided, each knowledge point can represent a small range of knowledge content, in a large learning content, there can be multiple such refined knowledge points, and learning these The order of refining knowledge points may not be fixed. By determining different learning orders of knowledge points according to different knowledge states of different students, learning efficiency and learning effect can be effectively improved, and self-adaptive personalized learning can be realized.
根据上述的思路和方法,可以设计不同的计算方式来计算KS。在一些实施例中,可以根据题目难度、作答时间、答题结果、参考状态值来计算学生在一个题目上的KS,计算公式如下:According to the above ideas and methods, different calculation methods can be designed to calculate KS. In some embodiments, the KS of a student on a question can be calculated according to the difficulty of the question, the answering time, the answering result, and the reference status value. The calculation formula is as follows:
其中,α代表题目难度,值域区间为[0,1],α越大则题目越难。在一些实施例中,题目难度可以根据一道题目下的做题正误分布统计出来。Among them, α represents the difficulty of the question, and the value range is [0,1]. The larger the α, the more difficult the question is. In some embodiments, the difficulty of a question may be calculated according to the distribution of correct and incorrect answers under a question.
β代表相对时间,即学生做题的实际作答时间(real_time)与该题目的预估做题时间(estimates_time)的比值,β越小表示学生做题越快。在一些实施例中,预估做题时间可以根据做题历史数据中该题的作答时间的中位数时长或平均时长统计出来。β represents the relative time, that is, the ratio of the actual answering time (real_time) of the student to the estimated answering time (estimates_time) of the question. The smaller the β, the faster the student will answer the question. In some embodiments, the estimated time for answering the question may be calculated according to the median duration or the average duration of the answering time of the question in the question-making history data.
σ即做题正误,1为正确,0为错误。σ is correct or incorrect, 1 is correct and 0 is incorrect.
θ为参考状态值,表示做对一道难度最低的题,且所花时间足够长时的KS值。σ=1或σ=0时都取θ相关函数,可使KS能取到[0,1]间所有值,不会出现断层。θ is the reference state value, which indicates the KS value when the question with the lowest difficulty is done correctly and the time spent is long enough. When σ=1 or σ=0, the θ correlation function is taken, so that KS can take all the values between [0, 1], and no fault will appear.
学生题目做错,学生的KS值仍然大于0的原因是因为在自适应系统中,学生能做到较高难度题目,说明系统基于历史表现,认为学生有这个能力做这种难度的题目,所以即使学生将该题目做错,也会给学生设置大于0的知识状态。The reason why the student made a wrong question and the student's KS value is still greater than 0 is because in the adaptive system, the student can complete the question of higher difficulty. Even if the student gets the question wrong, a knowledge status greater than 0 will be set for the student.
由于在题目做错的情况下,影响作答时间长短的原因较复杂,故在题目做错时KS不受作答时间的影响。学生做到难度较低的题,很大可能说明系统认为学生的能力适合做较基础的题,因此即使学生将该题做对,作答时间很快,KS值也会受到一定限制。由此可以看出,该方法与自适应学习是有直接关系的。Because of the complicated reasons that affect the answering time when the question is wrong, KS is not affected by the answering time when the question is wrong. If the student does a question of lower difficulty, it is likely that the system believes that the student's ability is suitable for doing the more basic question. Therefore, even if the student does the question correctly and the answering time is fast, the KS value will be limited to a certain extent. It can be seen that this method is directly related to adaptive learning.
对于做对的题目,会根据其做题相对时间用sigmoid函数作为指数对其分值 进行上调。sigmoid(β)(指数)在(0,+∞)区间内的取值为(0.5,1),而底数(θ+(1-θ)*α)在(0,1)之间,故指数sigmoid(β)越小,KS值越大。学生做对题目且相对答题时间β也较小的情况下,sigmoid(β)的值接近0.5,则KS值上调幅度较大;相对答题时间β较大时,sigmoid(β)的值接近1,则上调幅度较小。For questions that are done right, their scores will be adjusted upwards according to the relative time of answering questions using the sigmoid function as an index. The value of sigmoid(β) (exponent) in the (0,+∞) interval is (0.5,1), and the base (θ+(1-θ)*α) is between (0,1), so the exponent The smaller the sigmoid(β), the larger the KS value. When the students do the right questions and the relative answering time β is also small, the value of sigmoid(β) is close to 0.5, and the KS value increases greatly; when the relative answering time β is large, the value of sigmoid(β) is close to 1, The increase is smaller.
图1是本申请实施例提供的一种sigmoid(β)的函数曲线图。如图1所示,横坐标表示β(即相对答题时间,Time Ratio),纵坐标表示sigmoid(β)的值。从图中可以看到,β为0时,sigmoid(β)的初始值为0.5;随着β的增大,即随着作答时间的增长,sigmoid(β)的值逐渐增大,但增速放缓,并逐渐趋近于1。FIG. 1 is a function curve diagram of a sigmoid (β) provided by an embodiment of the present application. As shown in Figure 1, the abscissa represents β (that is, the relative answering time, Time Ratio), and the ordinate represents the value of sigmoid (β). As can be seen from the figure, when β is 0, the initial value of sigmoid(β) is 0.5; as β increases, that is, with the increase of answering time, the value of sigmoid(β) gradually increases, but the growth rate increases. slowed down and gradually approached 1.
图2是本申请实施例提供的一种KS值的函数曲线图。如图2所示,横坐标表示β(即相对答题时间,Time Ratio),纵坐标表示KS的值。从图中可以看到,当β较小时,即作答时间较短,KS值较大;随着β的增大,即随着作答时间的增长,KS值逐渐减小,但减速放缓。例如:正常情况下,大部分学生是在10分钟之内做对一道题目,则在这个β的范围内,由于函数曲线斜率较大,因此区分度较好,即使不同的学生的作答时间比较接近,也能够有效地区分不同的学生;而对于超过10分钟之外的时间范围,比如30分钟甚至60分钟,即使最终能够做对题目,也说明学生的掌握状态都不太好,因此在这个β的范围内,也就不再需要做更细致的区分了,函数曲线斜率较小,变化幅度趋近于平缓。通过sigmoid(β)进行处理,既能够对大部分学生进行有效地区分,也能够兼顾处理趋近于正无穷的作答时间。FIG. 2 is a function curve diagram of a KS value provided by an embodiment of the present application. As shown in Figure 2, the abscissa represents β (that is, the relative answering time, Time Ratio), and the ordinate represents the value of KS. It can be seen from the figure that when β is small, that is, the answering time is short, and the KS value is large; with the increase of β, that is, as the answering time increases, the KS value gradually decreases, but the deceleration slows down. For example: under normal circumstances, most students get a question right within 10 minutes, then within the range of β, due to the larger slope of the function curve, the degree of discrimination is better, even if the answering time of different students is relatively close , can also effectively distinguish different students; and for the time range beyond 10 minutes, such as 30 minutes or even 60 minutes, even if they can finally get the question right, it means that the students' mastery state is not very good, so in this β Within the range of , there is no need to make a more detailed distinction, the slope of the function curve is small, and the change range tends to be gentle. Processing by sigmoid(β) can not only effectively distinguish most students, but also deal with the answering time approaching positive infinity.
在得到学生在一个题目上的KS(即人+知识点粒度上的知识状态)的基础上,可以计算该学生在一个知识点上的综合知识状态(综合KS),以测量学生对知识点的掌握状态。在一些实施例中,可以通过对该知识点下每一道题目赋一个权重,然后将该学生在该知识点下每一道题目的KS值加权平均来得到该知识点的综合KS。On the basis of obtaining the student's KS on a topic (that is, the knowledge status of the person + knowledge point granularity), the student's comprehensive knowledge status (comprehensive KS) on a knowledge point can be calculated to measure the student's knowledge point. Master the state. In some embodiments, the comprehensive KS of the knowledge point may be obtained by assigning a weight to each question under the knowledge point, and then weighted average of the KS values of each question under the knowledge point of the student.
在一些实施例中,可以利用神经网络,通过反向传播和梯度下降方法反复迭代收敛而最终得到权重。在另一些实施例中,也可以使用和为1的等差数列作为该知识点下题目的权重,计算方式如下:In some embodiments, a neural network may be used to iteratively converge through back-propagation and gradient descent methods to finally obtain the weights. In other embodiments, an arithmetic progression with a sum of 1 can also be used as the weight of the topic under the knowledge point, and the calculation method is as follows:
a
k=a
1+(k-1)*d
a k =a 1 +(k-1)*d
其中,S
k为等差数列前k项的和,k取[1,n]的自然数,n为该知识点下作答的题目数量(即需要为多少道题目赋予权重),d为等差数列中两个相邻值之间的 差,a
1为等差数列的第一项(即为第一道题目的权重),a
k为等差数列的第k项(即为第k道题目的权重)。
Among them, S k is the sum of the first k items of the arithmetic sequence, k is a natural number of [1,n], n is the number of questions answered under the knowledge point (that is, how many questions need to be weighted), d is the arithmetic sequence The difference between two adjacent values in , a 1 is the first item of the arithmetic sequence (that is, the weight of the first question), and a k is the kth item of the arithmetic sequence (that is, the kth item of the question). Weights).
取S
n=1(等差数列的和为1,即所有题目的权重之和为1)以及a
1=d(为便于计算,并且使最新作答的题目的权重足够大),则
Taking Sn = 1 (the sum of the arithmetic sequence is 1, that is, the sum of the weights of all questions is 1) and a 1 =d (for the convenience of calculation, and the weight of the newly answered question is large enough), then
例如:n=8(即学生在该知识点下作答了8道题目),则a
1=0.0278,且8道题目按日期顺序由远到近的权重分别是[0.0278,0.0556,0.0833,0.1111,0.1389,0.1667,0.1944,0.2222]。如果学生在该知识点上做的8道题的KS分别是[0.1,0.88,0.57,0.64,0.12,0.74,0.1,0.55],则该学生在该知识点上的综合KS为两个数组的点乘并相加,得到综合KS值为0.45。
For example: n=8 (that is, the student answered 8 questions under this knowledge point), then a 1 =0.0278, and the weights of the 8 questions from far to near are [0.0278, 0.0556, 0.0833, 0.1111, 0.1389, 0.1667, 0.1944, 0.2222]. If the KS of the 8 questions that the student does on the knowledge point is [0.1, 0.88, 0.57, 0.64, 0.12, 0.74, 0.1, 0.55], then the student's comprehensive KS on the knowledge point is the sum of the two arrays Dot multiply and add to get a combined KS value of 0.45.
通过上述等差数列的方式为权重赋值,只需知道题目数n,即可直接得到每道题的权重。这种方式还加强了最近题目表现的影响(作答题目的日期越近权重越大),而削弱了久远历史题目表现的权重(作答题目的日期越远权重越小),从而能够有效地反映学生最新的做题表现和最新的知识状态。并且,该方法简易便捷,只要输入需要赋权重的序列的相关参数,就能立即返回所需的权重序列,并且这种方法还自然地加强了最新作答的题目的权重,符合学生的实际学习情况。The weights are assigned by the above arithmetic sequence, and the weight of each question can be directly obtained only by knowing the number of questions n. This method also strengthens the influence of the performance of recent items (the date of answering the item is more recent), while weakening the weight of the performance of older historical items (the date of answering the item is smaller, the weight is smaller), which can effectively reflect the students The latest performance and the latest knowledge status. In addition, this method is simple and convenient, as long as the relevant parameters of the sequence that need to be weighted are input, the required weight sequence can be returned immediately, and this method also naturally strengthens the weight of the latest answers, which is in line with the actual learning situation of the students. .
上述通过等差数列为权重赋值的方法,还可以用于对学生的多次单元考试成绩的权重进行赋值的方案中,即依照此方法按时间先后计算学生的多次单元考试成绩的权重,进而计算多次单元考试成绩的加权平均值,以得到反映学生近期表现的综合成绩。The above method of assigning weights through arithmetic progression can also be used in the scheme of assigning the weights of students' multiple unit test scores, that is, according to this method, the weights of students' multiple unit test scores are calculated chronologically, and then A weighted average of multiple unit test scores is calculated to obtain a composite score that reflects a student's recent performance.
图3是本申请实施例提供的一种对知识点掌握状态进行测量的方法的流程示意图,以一个例子来进行描述。FIG. 3 is a schematic flowchart of a method for measuring a knowledge point mastery state provided by an embodiment of the present application, and an example is used for description.
步骤一,如表1所示,从PostgreSQL数据库中拉取user_record(做题记录表)中多个学生在多个题目上的正误(is_right)、难度(difficulty)、预估时间(estimates_time)、实际时间(cost_time)、开始时刻(create_time)等字段,并以开始时刻(题目开始的日期+时间)进行以一个字段为标准的升序排列(order by),以确保题目记录是从旧到新展现出来的,然后导出成csv文件。 Step 1, as shown in Table 1, pull the correctness (is_right), difficulty (difficulty), estimated time (estimates_time), actual error (is_right), difficulty (difficulty), estimated time (estimates_time), actual Time (cost_time), start time (create_time) and other fields, and the start time (the date of the start of the topic + time) is sorted in ascending order (order by) as a standard to ensure that the topic records are displayed from old to new , and then export to csv file.
表1Table 1
步骤二,如表2所示,在Python里用pandas的读取csv文件函数(read_csv)将上述csv文件导入,得出如表2所示的pandas里的数据框架(DataFrame)格式。对它新加一个字段knowledge_state,然后通过技术方案所述公式用Python里的映射函数(map)对每行的is_right,difficulty,estimates_time,cost_time进行相应的映射,得到每行的knowledge_state值。 Step 2, as shown in Table 2, import the above csv file using pandas' read csv file function (read_csv) in Python, and obtain the data frame (DataFrame) format in pandas as shown in Table 2. Add a new field knowledge_state to it, and then use the mapping function (map) in Python to map the is_right, difficulties, estimates_time, and cost_time of each row according to the formula described in the technical solution, and obtain the knowledge_state value of each row.
表2Table 2
user_iduser_id | tag_codetag_code | question_idquestion_id | is_rightis_right | difficultydifficulty | estimates_timeestimates_time | cost_timecost_time | create_timecreate_time | knowledge_stateknowledge_state | |
3326133261 | 30933093 | 2637526375 | 00 | 0.50.5 | 160160 | 3939 | 2019/10/262019/10/26 | 13:53:4513:53:45 | 0.10000.1000 |
3326133261 | 30933093 | 2717127171 | 11 | 0.80.8 | 180180 | 158158 | 2019/10/272019/10/27 | 9:56:499:56:49 | 0.88410.8841 |
3326133261 | 30933093 | 2134221342 | 11 | 0.20.2 | 3030 | 66 | 2019/11/12019/11/1 | 13:12:4313:12:43 | 0.57020.5702 |
3326133261 | 30933093 | 2179321793 | 11 | 0.30.3 | 120120 | 24twenty four | 2019/11/22019/11/2 | 16:26:1316:26:13 | 0.63670.6367 |
3326133261 | 30933093 | 1722117221 | 00 | 0.60.6 | 120120 | 112112 | 2019/11/92019/11/9 | 12:20:4212:20:42 | 0.12000.1200 |
3326133261 | 30933093 | 1515715157 | 11 | 0.50.5 | 120120 | 3434 | 2019/12/12019/12/1 | 11:03:0611:03:06 | 0.74720.7472 |
3326133261 | 30933093 | 1973819738 | 00 | 0.50.5 | 160160 | 2828 | 2019/12/62019/12/6 | 19:48:1019:48:10 | 0.10000.1000 |
3326133261 | 30933093 | 2030620306 | 11 | 0.20.2 | 9090 | 3030 | 2019/12/72019/12/7 | 16:25:5116:25:51 | 0.55150.5515 |
4716247162 | 78847884 | 1835818358 | 11 | 0.40.4 | 150150 | 5454 | 2019/12/142019/12/14 | 11:03:3711:03:37 | 0.68030.6803 |
4716247162 | 78847884 | 2669826698 | 00 | 0.20.2 | 3030 | 33 | 2019/12/152019/12/15 | 9:58:039:58:03 | 0.04000.0400 |
4716247162 | 78847884 | 1694116941 | 00 | 0.80.8 | 180180 | 1717 | 2019/12/192019/12/19 | 19:54:1219:54:12 | 0.16000.1600 |
4716247162 | 78847884 | 2066320663 | 11 | 0.20.2 | 100100 | 7575 | 2019/12/212019/12/21 | 8:46:408:46:40 | 0.49960.4996 |
4716247162 | 78847884 | 1708117081 | 00 | 0.50.5 | 180180 | 135135 | 2019/12/212019/12/21 | 9:46:569:46:56 | 0.10000.1000 |
4716247162 | 78847884 | 1157111571 | 00 | 0.50.5 | 180180 | 2626 | 2019/12/212019/12/21 | 14:23:3414:23:34 | 0.10000.1000 |
4716247162 | 78847884 | 2818228182 | 00 | 0.80.8 | 210210 | 5858 | 2019/12/212019/12/21 | 14:30:0014:30:00 | 0.16000.1600 |
4716247162 | 78847884 | 1007410074 | 11 | 0.80.8 | 240240 | 315315 | 2019/12/212019/12/21 | 21:59:3221:59:32 | 0.87160.8716 |
4716247162 | 78847884 | 1190311903 | 00 | 0.50.5 | 120120 | 2727 | 2019/12/272019/12/27 | 20:46:3720:46:37 | 0.10000.1000 |
4716247162 | 78847884 | 2775827758 | 11 | 0.80.8 | 120120 | 8787 | 2019/12/282019/12/28 | 9:54:249:54:24 | 0.88920.8892 |
4058640586 | 65216521 | 2048320483 | 00 | 0.80.8 | 240240 | 1818 | 2019/12/292019/12/29 | 9:56:449:56:44 | 0.16000.1600 |
4058640586 | 65216521 | 1729417294 | 11 | 0.20.2 | 9090 | 3232 | 2020/1/42020/1/4 | 11:14:1911:14:19 | 0.54840.5484 |
4058640586 | 65216521 | 2535825358 | 11 | 0.50.5 | 130130 | 4343 | 2020/1/42020/1/4 | 11:18:1411:18:14 | 0.74280.7428 |
步骤三,如表3所示,用pandas的分组统计函数(groupby)将数据表以学生账号(user_id)、知识点编号(tag_code)来进行分组,以聚合函数(agg)将人+知识点粒度下的一系列KS值组合成列表(list)的形式,再用重置角标函数(reset_index)重置表格的字段和角标。 Step 3, as shown in Table 3, use pandas' grouping statistical function (groupby) to group the data table by student account number (user_id) and knowledge point number (tag_code), and use the aggregation function (agg) to group people + knowledge point granularity. The following series of KS values are combined into the form of a list (list), and then use the reset index function (reset_index) to reset the fields and index of the table.
表3table 3
user_iduser_id | tag_codetag_code | knowledge_state_listknowledge_state_list |
3471634716 | 70127012 | [0.34,0.78,0.12,0.26,0.53][0.34, 0.78, 0.12, 0.26, 0.53] |
3495534955 | 67066706 | [0.26,0.83,0.1,0.23,0.34][0.26, 0.83, 0.1, 0.23, 0.34] |
3496934969 | 35383538 | [0.83,0.78,0.53,0.87,0.12,0.34][0.83, 0.78, 0.53, 0.87, 0.12, 0.34] |
3530835308 | 36313631 | [0.15,0.84,0.29,0.39,0.88,0.47][0.15, 0.84, 0.29, 0.39, 0.88, 0.47] |
3558635586 | 77947794 | [0.52,0.7,0.84,0.18,0.3,0.51][0.52, 0.7, 0.84, 0.18, 0.3, 0.51] |
3578935789 | 47454745 | [0.27,0.12,0.79,0.17,0.67,0.28][0.27, 0.12, 0.79, 0.17, 0.67, 0.28] |
3629136291 | 30383038 | [0.44,0.65,0.06,0.61,0.37,0.61,0.29][0.44, 0.65, 0.06, 0.61, 0.37, 0.61, 0.29] |
3642036420 | 42564256 | [0.64,0.82,0.33,0.29,0.03,0.53,0.15][0.64, 0.82, 0.33, 0.29, 0.03, 0.53, 0.15] |
3662036620 | 52855285 | [0.73,0.28,0.82,0.25,0.24,0.74,0.17][0.73, 0.28, 0.82, 0.25, 0.24, 0.74, 0.17] |
3675036750 | 71857185 | [0.84,0.66,0.04,0.8,0.1,0.58][0.84, 0.66, 0.04, 0.8, 0.1, 0.58] |
步骤四,如表4所示,用自定义的加权知识状态函数(weighted_knowledge_state),根据等差数列加权聚合出人+知识点粒度上的KS。该自定义函数仅以步骤三所示题目上的KS的list形式的序列,如[0.83,0.72,0.69,0.23,0.8]为输入参数,即可获得等差数列的n值,从而生成等差数列权重,并与入参进行点乘,加权求和(中间用到了numpy中求和函数(sum)),得到最终KS值作为出参,方便快捷。这一步得出的KS每24小时更新一次,以捕捉到学生最新的能力水平。 Step 4, as shown in Table 4, use a self-defined weighted knowledge state function (weighted_knowledge_state) to weight and aggregate KS at the granularity of people + knowledge points according to the arithmetic sequence. The custom function only takes the sequence of the KS list form on the topic shown in step 3, such as [0.83, 0.72, 0.69, 0.23, 0.8] as the input parameter, to obtain the n value of the arithmetic sequence, thereby generating the arithmetic difference The weight of the sequence, and dot multiplication with the input parameters, weighted summation (the summation function (sum) in numpy is used in the middle), and the final KS value is obtained as the output parameter, which is convenient and fast. The KS derived from this step is updated every 24 hours to capture the student's latest ability level.
表4Table 4
步骤五,以每个学生所有知识点的KS的均值作为学生的SKS,如步骤四所述,每24小时更新一次SKS,剔除老数据获得新数据,可得到学生最新的能力水平,从而对应地更新学生的后续推题初始难度值。如3个学生更新后的总体SKS值分别为0.82,0.57,0.35,则第二天给学生在知识点上推题的初始难度根据一定映射方法会不同(如0.8,0.5,0.2),以适应不同学生的水平,体现系统的自适应性(而传统方法中的初始难度都相同,比如都是0.5)。Step 5: Take the mean value of the KS of all knowledge points of each student as the student's SKS. As described in step 4, update the SKS every 24 hours, remove the old data to obtain new data, and obtain the student's latest ability level, thereby correspondingly. Update the initial difficulty value of the student's subsequent questions. For example, the updated overall SKS values of the three students are 0.82, 0.57, and 0.35, respectively, then the initial difficulty of pushing the questions to the students on the knowledge points on the second day will be different according to a certain mapping method (such as 0.8, 0.5, 0.2) to adapt to The level of different students reflects the adaptability of the system (while the initial difficulty in the traditional method is the same, such as 0.5).
步骤六,以每个知识点下学生们的KS的均值作为知识点的KPKS,每24小时更新一次KPKS,可得到知识点最新的KPKS。当一个学生面对N个知识点要学的时候,按照KPKS与SKS的差的绝对值由小到大推送知识点,即KS值相近的优先推送。例如:一名SKS为0.67的学生,有KPKS分别为0.82,0.7,0.45,0.58的A、B、C、D四个知识点要学,则知识点学习顺序为B、D、A、C(传统方法是按照一定的规则或顺序推送,无法体现自适应性)。Step 6: Take the mean value of students' KS under each knowledge point as the KPKS of the knowledge point, update the KPKS every 24 hours, and get the latest KPKS of the knowledge point. When a student faces N knowledge points to learn, the knowledge points are pushed from small to large according to the absolute value of the difference between KPKS and SKS, that is, those with similar KS values are pushed first. For example: a student whose SKS is 0.67 has four knowledge points A, B, C, and D with KPKS of 0.82, 0.7, 0.45, and 0.58 to learn, then the learning order of knowledge points is B, D, A, C ( The traditional method is to push according to certain rules or order, which cannot reflect the adaptability).
在该实施例中,使用Python作为工具,用到的Python库有:Pandas,Numpy。Numpy是在Python里提供多种复杂数学运算的库。Pandas则是通过数据框架以便进行数据分析处理的库。数据是从PostgreSQL数据库里拉取出来的。作为一个实施例,部分代码如下:In this embodiment, Python is used as a tool, and the Python libraries used are: Pandas, Numpy. Numpy is a library that provides a variety of complex mathematical operations in Python. Pandas is a library for data analysis and processing through data frames. The data is pulled from the PostgreSQL database. As an example, part of the code is as follows:
import pandas as pdimport pandas as pd
import numpy as npimport numpy as np
from sklearn.metrics.pairwise import cosine_similarityfrom sklearn.metrics.pairwise import cosine_similarity
本申请提供的测量知识点掌握状态的方法,相对于相关技术至少具有以下优点:The method for measuring the state of mastery of knowledge points provided by the present application has at least the following advantages over related technologies:
(1)通过正误、时间、难度综合考虑学生在知识点上的状态,能够较好地区分不同学生的水平,摆脱了光从正误和难度看学生的知识状态导致相同值太多,无法分别出不同水平的学生的问题。(1) By comprehensively considering the state of students' knowledge points through correctness, time, and difficulty, it is possible to better distinguish the level of different students, and get rid of the knowledge state of students only from correctness and difficulty and difficulty, which leads to too many identical values, which cannot be distinguished. Questions for students of different levels.
(2)sigmoid函数是非线性函数,并且放在指数位置上而不是简单地与(θ+(1-θ)*α)相乘。利用此特点,可以使KS值不总是因相对时间增大而大幅减小,能使其逐渐平缓减小。因相对时间大到一定程度时,对KS影响也很小了。sigmoid函数能有效地控制相对时间的值域在(0.5,1)之间,并且恰如其分地对KS进行合适的调整。即学生做题的速度快的话,可对底数值取0.5次方左右来调高KS值;若学生做题的速度较慢,则底数值基本没有调高。(2) The sigmoid function is a nonlinear function, and is placed in the exponential position rather than simply multiplied by (θ+(1-θ)*α). Using this feature, the value of KS can not always be greatly reduced due to the increase of relative time, but can be gradually reduced gradually. When the relative time is large to a certain extent, the impact on KS is also very small. The sigmoid function can effectively control the value range of the relative time between (0.5, 1), and adjust KS appropriately. That is, if the students do the questions fast, they can raise the KS value by taking the base value to the power of 0.5; if the students do the questions slowly, the base value is basically not raised.
(3)用Python中map函数生成knowledge_state值,用groupby,agg,reset_index函数生成list形式的KS序列,简单快捷,消耗时间少。(3) Use the map function in Python to generate the knowledge_state value, and use the groupby, agg, reset_index functions to generate the KS sequence in the form of a list, which is simple and fast, and consumes less time.
(4)用等差数列权重加权得出KS,weighted_knowledge_state函数入参简单、方便快捷,在效果上能起到加强学生近期表现的作用。也可以使用其他数列,其权重序列是一个总和等于1,相邻两个权重中前面的权重值小后面的权重值大,且相邻两个权重的差逐渐递增或相等的序列。(4) The KS is obtained by weighting the arithmetic sequence weight. The weighted_knowledge_state function is simple, convenient and quick to input parameters, which can play a role in strengthening the recent performance of students. Other sequences can also be used, and the weight sequence is a sequence in which the sum is equal to 1, the weight value in the front of the two adjacent weights is small, and the weight value in the back is large, and the difference between the two adjacent weights is gradually increasing or equal.
(5)通过对题目、知识点等粒度的学生的KS的定位,能够用这个指标来推题和有针对性地帮助学生寻找知识漏洞,规划学习路径。如学生在知识点上的平均KS很高(如0.8),说明学生在知识点上,尤其是最近几道题做得又快 又准,能力很强。则其后续推送题目的难度将会变大,以适应学生最新的能力水平。后续推知识点也会优先推KS值相近的知识点。(5) By locating students' KS at the granularity of topics, knowledge points, etc., this indicator can be used to deduce questions and help students find knowledge loopholes and plan learning paths in a targeted manner. If the average KS of the students on the knowledge points is very high (eg 0.8), it means that the students have done quickly and accurately on the knowledge points, especially the last few questions, and the ability is very strong. Then the difficulty of the subsequent push questions will become larger to adapt to the latest ability level of students. Subsequent push knowledge points will also give priority to knowledge points with similar KS values.
本申请提供的技术方案可以是系统、方法、装置和/或计算机程序产品。计算机程序产品可以包括计算机可读存储介质,其上载有用于使处理器实现本申请的多个方面的计算机可读程序指令。The technical solutions provided in this application may be systems, methods, apparatuses and/or computer program products. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement aspects of the present application.
在一些实施例中,本申请还提供一种计算机装置、设备或终端。该计算机装置、设备或终端包括通过系统总线连接的处理器、存储器、网络接口、显示屏和输入装置。其中,处理器设置为提供计算和控制能力,存储器包括非易失性存储介质、内存储器。非易失性存储介质存储有操作系统和计算机程序。内存储器为非易失性存储介质中的操作系统和计算机程序的运行提供环境。网络接口设置为与外部的终端通过网络连接通信。计算机程序被处理器执行时以实现本申请公开的多种方法、流程、步骤,或者处理器执行计算机程序时实现本申请公开的实施例中多个模块或单元的功能。显示屏可以是液晶显示屏或者电子墨水显示屏,输入装置可以是显示屏上覆盖的触摸层,也可以是外壳上设置的按键、轨迹球或触控板,还可以是外接的键盘、触控板或鼠标等。In some embodiments, the present application also provides a computer apparatus, device or terminal. The computer device, device or terminal includes a processor, memory, network interface, display screen and input device connected by a system bus. The processor is configured to provide computing and control capabilities, and the memory includes a non-volatile storage medium and an internal memory. The nonvolatile storage medium stores an operating system and a computer program. Internal memory provides an environment for the execution of the operating system and computer programs in the non-volatile storage medium. The network interface is configured to communicate with external terminals through a network connection. When the computer program is executed by the processor, it implements various methods, processes, and steps disclosed in the present application, or when the processor executes the computer program, it implements the functions of multiple modules or units in the embodiments disclosed in the present application. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device can be a touch layer covered on the display screen, a button, a trackball or a touchpad set on the casing, or an external keyboard, touch board or mouse, etc.
示例性的,计算机程序可以被分割成一个或多个模块或单元,这些模块或单元被存储在存储器中,并可由处理器执行,以实现本申请的技术方案。这些模块或单元可以是能够完成特定功能的一系列计算机程序指令段,该指令段用于描述计算机程序在装置、设备或终端中的执行过程。Exemplarily, a computer program can be divided into one or more modules or units, and these modules or units are stored in a memory and executed by a processor to implement the technical solutions of the present application. These modules or units may be a series of computer program instruction segments capable of accomplishing specific functions, and the instruction segments are used to describe the execution process of the computer program in an apparatus, device or terminal.
上述的装置、设备或终端可以是桌上型计算机、笔记本、移动电子设备、掌上电脑及云端服务器等计算设备。图中所示出的结构,仅仅是与本申请方案相关的部分结构的框图,并不构成对本申请方案所应用于其上的装置、设备或终端的限定,装置、设备或终端可以包括比图中所示更多或更少的部件,或者组合一些部件,或者具有不同的部件布置。The above-mentioned apparatus, device or terminal may be a desktop computer, a notebook, a mobile electronic device, a palmtop computer, a cloud server and other computing devices. The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the device, equipment or terminal to which the solution of the present application is applied. shown in more or less components, or in combination of some components, or with different component arrangements.
处理器可以是中央处理单元(Central Processing Unit,CPU),也可以是其他通用或专用的处理器、微处理器、数字信号处理器(Digital Signal Processor,DSP)、专用集成电路(Application Specific Integrated Circuit,ASIC)、可编程门阵列(Field-Programmable Gate Array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。处理器是上述的装置、设备或终端的控制中心,利用多种接口和线路连接装置、设备或终端的多个部分。The processor can be a central processing unit (CPU), or other general-purpose or special-purpose processors, microprocessors, digital signal processors (DSPs), application specific integrated circuits (Application Specific Integrated Circuits) , ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor is the control center of the above-mentioned apparatus, equipment or terminal, and uses various interfaces and lines to connect various parts of the apparatus, equipment or terminal.
存储器可设置为存储计算机程序、模块和数据,处理器通过运行或执行存储在存储器内的计算机程序和/或模块,以及调用存储在存储器内的数据,实现装置、设备或终端的多种功能。存储器可主要包括程序存储区和数据存储区,其中,程序存储区可存储操作系统、至少一个功能所需的应用程序(比如声音 播放功能、图像播放功能等)等;数据存储区可存储根据应用所创建的多类数据(比如多媒体数据、文档、操作历史记录等)等。此外,存储器可以包括高速随机存取存储器,还可以包括非易失性存储器,例如硬盘、内存、插接式硬盘、智能存储卡(Smart Media Card,SMC)、安全数字(Secure Digital,SD)卡、闪存卡(Flash Card)、磁盘存储器件、闪存器件、或其他易失性固态存储器件。The memory can be configured to store computer programs, modules and data, and the processor implements various functions of the apparatus, device or terminal by running or executing the computer programs and/or modules stored in the memory and calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; Created various types of data (such as multimedia data, documents, operation history, etc.) and so on. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory such as hard disk, internal memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card , Flash Card, magnetic disk storage device, flash memory device, or other volatile solid-state storage device.
本申请还提供一种计算机可读存储介质,其上存储有计算机程序,计算机程序被处理器执行时实现上述方法的步骤。实现上述实施例方法中的全部或部分流程,是可以通过计算机程序来指令相关的硬件来完成,所述的计算机程序可存储于一非易失性计算机可读取存储介质中,该计算机程序在执行时,可包括如上述方法的实施例的流程。其中,本申请所提供的实施例中所使用的对存储器、存储、数据库或其它介质的任何引用,均可包括非易失性和/或易失性存储器。非易失性存储器可包括只读存储器(Read-Only Memory,ROM)、可编程ROM(Programmable ROM,PROM)、电可编程ROM(Electrical PROM,EPROM)、电可擦除可编程ROM(Electrically Erasable PROM,EEPROM)或闪存。易失性存储器可包括随机存取存储器(Random Access Memory,RAM)或者外部高速缓冲存储器。作为说明而非局限,RAM以多种形式可得,诸如静态RAM(Static RAM,SRAM)、动态RAM(Dynamic RAM,DRAM)、同步动态DRAM(Synchronous DRAM,SDRAM)、双数据率SDRAM(Double Data Rate SDRAM,DDRSDRAM)、增强型SDRAM(Enhanced SDRAM,ESDRAM)、同步链路DRAM(Synchlink DRAM,SLDRAM)、存储器总线直接RAM(Rambus Direct RAM,RDRAM)、直接存储器总线动态RAM(Direct Rambus DRAM,DRDRAM)、以及存储器总线动态RAM(Rambus DRAM,RDRAM)等。The present application also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above method are implemented. All or part of the process in the method of the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium, and the computer program is stored in a non-volatile computer-readable storage medium. When executed, it may include the flow of the embodiment of the above-mentioned method. Wherein, any reference to memory, storage, database or other medium used in the embodiments provided in this application may include non-volatile and/or volatile memory. Non-volatile memory may include read-only memory (Read-Only Memory, ROM), programmable ROM (Programmable ROM, PROM), electrically programmable ROM (Electrical PROM, EPROM), electrically erasable programmable ROM (Electrically Erasable ROM) PROM, EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (Double Data) Rate SDRAM, DDR SDRAM), enhanced SDRAM (Enhanced SDRAM, ESDRAM), synchronous link DRAM (Synchlink DRAM, SLDRAM), memory bus direct RAM (Rambus Direct RAM, RDRAM), direct memory bus dynamic RAM (Direct Rambus DRAM, DRDRAM) ), and memory bus dynamic RAM (Rambus DRAM, RDRAM), etc.
上述的装置或终端设备集成的模块和单元,如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在计算机可读存储介质中。基于这样的理解,本申请实现所公开的多种方法中的全部或部分流程,也可以通过计算机程序来指令相关的硬件来完成,计算机程序可存储于计算机可读存储介质中,该计算机程序在被处理器执行时,可实现上述多个方法的步骤。其中,计算机程序包括计算机程序代码,计算机程序代码可以为源代码形式、对象代码形式、可执行文件或一些中间形式等。计算机可读介质可以包括:能够携带计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、ROM、RAM、电载波信号、电信信号以及软件分发介质等。计算机可读介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减。If the above-mentioned integrated modules and units of the apparatus or terminal equipment are implemented in the form of software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium. Based on this understanding, the implementation of all or part of the processes in the various methods disclosed in the present application can also be completed by instructing relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program is in When executed by a processor, the steps of the various methods described above may be implemented. The computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate forms, and the like. Computer readable media may include: any entity or device capable of carrying computer program code, recording media, USB flash drives, removable hard disks, magnetic disks, optical discs, computer memory, ROM, RAM, electrical carrier signals, telecommunication signals, and software distribution media, etc. . Computer readable media may contain suitable additions or deletions as required by legislation and patent practice in jurisdictions.
在一些实施例中,本申请公开的多种方法、流程、模块、装置、设备或系 统可以在一个或多个处理装置(例如,数字处理器、模拟处理器、被设计成设置为处理信息的数字电路、被设计成设置为处理信息的模拟电路、状态机、计算设备、计算机和/或设置为以电子方式处理信息的其他机构)中被实现或执行。该一个或多个处理装置可以包括响应于以电子方式存储在电子存储介质上的指令来执行方法的一些或所有操作的一个或多个装置。该一个或多个处理装置可以包括通过硬件、固件和/或软件被配置而专门设计成用于执行方法的一项或多项操作的一个或多个装置。In some embodiments, the various methods, processes, modules, apparatus, devices, or systems disclosed herein may be implemented in one or more processing devices (eg, digital processors, analog processors, devices designed and arranged to process information) implemented or implemented in digital circuits, analog circuits designed to process information, state machines, computing devices, computers, and/or other mechanisms configured to process information electronically). The one or more processing devices may include one or more devices that perform some or all of the operations of the method in response to instructions electronically stored on an electronic storage medium. The one or more processing means may comprise one or more means specially designed for carrying out one or more operations of the method configured by hardware, firmware and/or software.
本申请的实施方式可以在硬件、固件、软件或其多种组合中进行,还可以作为存储在机器可读介质上的且可以使用一个或多个处理装置读取和执行的指令来实现。在一些实施方式中,机器可读介质可以包括用于存储和/或传输呈机器(例如,计算装置)可读形式的信息的多种机构。例如,机器可读存储介质可以包括只读存储器、随机存取存储器、磁盘存储介质、光存储介质、快闪存储器装置以及设置为存储信息的其他介质,并且机器可读传输介质可以包括多种形式的传播信号(包括载波、红外信号、数字信号)以及设置为传输信息的其他介质。虽然在执行一些动作的特定示例性方面和实施方式的角度可以在以上公开内容中描述固件、软件、例程或指令,但将明显的是,这类描述仅出于方便目的并且这类动作实际上由机器设备、计算装置、处理装置、处理器、控制器、或执行固件、软件、例程或指令的其他装置或机器产生。Embodiments of the present application may be implemented in hardware, firmware, software, or various combinations thereof, and may also be implemented as instructions stored on a machine-readable medium that may be read and executed using one or more processing devices. In some implementations, a machine-readable medium can include various mechanisms for storing and/or transmitting information in a form readable by a machine (eg, a computing device). For example, machine-readable storage media may include read-only memory, random-access memory, magnetic disk storage media, optical storage media, flash memory devices, and other media configured to store information, and machine-readable transmission media may include a variety of forms Propagated signals (including carrier waves, infrared signals, digital signals) and other media configured to transmit information. Although firmware, software, routines, or instructions may be described in the above disclosure in terms of certain exemplary aspects and implementations that perform some actions, it will be apparent that such descriptions are for convenience only and that such actions are practical generated by a machine device, computing device, processing device, processor, controller, or other device or machine executing firmware, software, routines or instructions.
在本文中,用来执行指定功能的模块或者使用功能性特征描述的模块,意在涵盖能够执行该功能的任何方式,例如:执行该功能的电路元件的组合,用来执行或实现该功能的软件、硬件以及软件和硬件的组合,或者任何形式的软件、固件、代码及其与适当电路或其他装置的组合。由多种模块提供的功能被以本文所主张的方式组合在一起,由此应当认为,是可以提供这些功能的任何模块、部件、元件都等价或等效于权利要求书中限定的模块。根据电路的等效变换的原理,也可以对本申请中一些实施例的电路结构进行变更、修改,例如:将电流源变换为电压源、串联结构变换为并联结构等,从而获得更多样化的实施例,但这些变更和修改均属于本申请公开的范围。As used herein, a module for performing a specified function, or a module described using a functional feature, is intended to encompass any means capable of performing the function, such as a combination of circuit elements that perform the function, a module used to perform or implement the function Software, hardware, and a combination of software and hardware, or any form of software, firmware, code, and their combination with suitable circuitry or other means. The functions provided by the various modules are combined together in the manner claimed herein, and it is therefore intended that any module, component, element that provides the functions is equivalent or equivalent to the modules defined in the claims. According to the principle of the equivalent transformation of the circuit, the circuit structure of some embodiments in this application can also be changed or modified, for example, the current source is transformed into a voltage source, the series structure is transformed into a parallel structure, etc., so as to obtain more diversified Examples, but these changes and modifications belong to the scope of the disclosure of the present application.
本文使用示例来公开本申请,其中的一个或多个示例被描述或者图示于说明书及其附图之中。每个示例都是为了解释本申请而提供,而不是为了限制本申请。The application is disclosed herein using examples, one or more of which are described or illustrated in the specification and its drawings. Each example is provided to explain the application, not to limit the application.
Claims (10)
- 一种对知识点掌握状态进行测量的方法,包括:A method for measuring the mastery state of knowledge points, including:采集学生作答与知识点相关的题目的作答时间和答题结果;Collect the answering time and answering results of students answering questions related to knowledge points;根据所述作答时间和所述答题结果,计算所述学生在所述题目上的知识状态;Calculate the knowledge status of the student on the topic according to the answering time and the answering result;将所述学生完成作答的与所述知识点相关的所有题目的知识状态的平均值,作为所述学生在所述知识点上的综合知识状态,以实现测量所述学生对所述知识点的掌握状态。The average value of the knowledge status of all the questions related to the knowledge point answered by the student is taken as the comprehensive knowledge status of the student on the knowledge point, so as to measure the student's understanding of the knowledge point. Master the state.
- 根据权利要求1所述的方法,其中,每一个题目对应一个题目难度;The method according to claim 1, wherein each question corresponds to a question difficulty;所述根据所述作答时间和所述答题结果,计算所述学生在所述题目上的知识状态,包括:The calculating the knowledge status of the student on the topic according to the answering time and the answering result, including:根据所述作答时间、所述答题结果以及所述题目难度计算所述知识状态,其中,所述题目难度是根据所有作答所述题目的学生的答题结果的分布统计出来的;Calculate the knowledge state according to the answering time, the answering result and the question difficulty, wherein the question difficulty is calculated according to the distribution of answering results of all students who answered the question;其中,所述题目难度越大,则所述知识状态的值越大。Wherein, the greater the difficulty of the question, the greater the value of the knowledge state.
- 根据权利要求2所述的方法,其中,所述根据所述作答时间、所述答题结果以及所述题目难度计算所述知识状态,包括:The method according to claim 2, wherein calculating the knowledge state according to the answering time, the answering result and the question difficulty comprises:在所述答题结果为错误的情况下,仅根据所述题目难度计算所述知识状态;In the case that the answer result is wrong, the knowledge state is only calculated according to the difficulty of the question;在所述答题结果为正确的情况下,根据所述作答时间和所述题目难度计算所述知识状态。When the answering result is correct, the knowledge state is calculated according to the answering time and the difficulty of the question.
- 根据权利要求3所述的方法,其中,The method of claim 3, wherein,所述作答时间越长,则所述知识状态的值越小;且随着所述作答时间的增大,所述知识状态的值的变化趋势越平缓。The longer the answering time, the smaller the value of the knowledge state; and as the answering time increases, the more gentle the change trend of the value of the knowledge state.
- 根据权利要求1所述的方法,其中,所述根据所述作答时间和所述答题结果,计算所述学生在所述题目上的知识状态,包括:The method according to claim 1, wherein calculating the knowledge status of the student on the topic according to the answering time and the answering result, comprising:根据所述作答时间相对于所述题目的预估做题时间的倍数以及所述答题结果,计算所述知识状态;Calculate the knowledge state according to the multiple of the answering time relative to the estimated answering time of the question and the answering result;其中,所述预估做题时间为作答所述题目的所有学生的作答时间的中位数。Wherein, the estimated time for answering the question is the median of the answering time of all the students who answered the question.
- 根据权利要求1所述的方法,其中,The method of claim 1, wherein,所述学生完成作答的与所述知识点相关的所有题目的知识状态的平均值为加权平均值,其中,每一个题目对应一个权重。The average value of the knowledge status of all the questions related to the knowledge point completed by the student is a weighted average, wherein each question corresponds to a weight.
- 根据权利要求6所述的方法,其中,The method of claim 6, wherein,完成作答的日期越靠近当前日期的题目所对应的权重越大。Questions whose answer date is closer to the current date are given greater weight.
- 根据权利要求1所述的方法,其中,所述根据所述作答时间和所述答题结果,计算所述学生在所述题目上的知识状态,包括:The method according to claim 1, wherein calculating the knowledge status of the student on the topic according to the answering time and the answering result, comprising:根据参考状态值、所述作答时间以及所述答题结果计算所述知识状态,其中,所述参考状态值表示在做对一道难度最低的题目且作答所述一道难度最低的题目的作答时间满足预设时长条件的情况下的知识状态的值。The knowledge state is calculated according to the reference state value, the answering time, and the answering result, wherein the reference state value indicates that the answering time for answering a question with the lowest difficulty meets a predetermined requirement. The value of the knowledge state when the duration condition is set.
- 根据权利要求1所述的方法,还包括:The method of claim 1, further comprising:将所述学生在所有知识点上的综合知识状态的平均值,作为所述学生的总体知识状态,根据所述总体知识状态确定向所述学生推送题目的初始难度。The average value of the comprehensive knowledge state of the student on all knowledge points is taken as the overall knowledge state of the student, and the initial difficulty of pushing a question to the student is determined according to the overall knowledge state.
- 根据权利要求9所述的方法,还包括:The method of claim 9, further comprising:将所述知识点下所有学生的综合知识状态的平均值,作为所述知识点的平均知识状态,根据所述知识点的所述平均知识状态和所述学生的所述总体知识状态确定向所述学生推送知识点的顺序。The average value of the comprehensive knowledge state of all students under the knowledge point is taken as the average knowledge state of the knowledge point, and the direction to all students is determined according to the average knowledge state of the knowledge point and the overall knowledge state of the students. Describe the order in which students push knowledge points.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011059182.1 | 2020-09-30 | ||
CN202011059182.1A CN112131349B (en) | 2020-09-30 | 2020-09-30 | Method for measuring knowledge point mastering state |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022068435A1 true WO2022068435A1 (en) | 2022-04-07 |
Family
ID=73844919
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2021/112675 WO2022068435A1 (en) | 2020-09-30 | 2021-08-16 | Method for measuring knowledge point mastering state |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN112131349B (en) |
WO (1) | WO2022068435A1 (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114861916A (en) * | 2022-06-13 | 2022-08-05 | 合肥工业大学 | Knowledge association path fused cognitive tracking method |
CN116089705A (en) * | 2022-12-08 | 2023-05-09 | 读书郎教育科技有限公司 | Learning topic personalized recommendation method and system based on AI |
CN116151242A (en) * | 2023-04-19 | 2023-05-23 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | Intelligent problem recommendation method, system and storage medium for programming learning scene |
CN116383481A (en) * | 2023-02-09 | 2023-07-04 | 四川云数赋智教育科技有限公司 | Personalized test question recommending method and system based on student portrait |
CN116933870A (en) * | 2023-09-15 | 2023-10-24 | 浪潮软件股份有限公司 | Method, system and device for arranging elastic operation based on knowledge graph |
CN117114495A (en) * | 2023-09-11 | 2023-11-24 | 湖南软件职业技术大学 | Professional education quality assessment method for capability generation analysis |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112131349B (en) * | 2020-09-30 | 2021-08-06 | 上海松鼠课堂人工智能科技有限公司 | Method for measuring knowledge point mastering state |
CN113177868A (en) * | 2021-04-29 | 2021-07-27 | 重庆电子工程职业学院 | Teaching display system |
CN113344054B (en) * | 2021-05-29 | 2022-08-12 | 华中师范大学 | Knowledge skill dynamic diagnosis method for space-time evolution |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2002101555A2 (en) * | 2001-06-04 | 2002-12-19 | Honeywell International Inc. | Adaptive knowledge management system for vehicle trend monitoring, health management and preventive maintenance |
CN109800880A (en) * | 2018-12-29 | 2019-05-24 | 上海乂学教育科技有限公司 | Based on the adaptive learning Feature Extraction System of dynamic learning style information and application |
CN109859553A (en) * | 2018-12-04 | 2019-06-07 | 广东宜教通教育有限公司 | A kind of ability evaluating method doing topic situation based on student |
CN110309201A (en) * | 2019-06-28 | 2019-10-08 | 广州云蝶科技有限公司 | Operation method for customizing and system |
CN110377814A (en) * | 2019-05-31 | 2019-10-25 | 平安国际智慧城市科技股份有限公司 | Topic recommended method, device and medium |
CN112131349A (en) * | 2020-09-30 | 2020-12-25 | 上海松鼠课堂人工智能科技有限公司 | Method for measuring knowledge point mastering state |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2005036419A1 (en) * | 2003-10-15 | 2005-04-21 | Dharamdas Gautam Goradia | Interactive wisdom system |
CN104239969A (en) * | 2014-09-04 | 2014-12-24 | 上海合煦信息科技有限公司 | Evaluation and problem recommendation system for individualized education |
CN106776804A (en) * | 2016-11-22 | 2017-05-31 | 广东小天才科技有限公司 | Test question difficulty quantification method and system |
CN107203583A (en) * | 2017-03-27 | 2017-09-26 | 杭州博世数据网络有限公司 | It is a kind of that method is inscribed based on the smart group that big data is analyzed |
CN107292785A (en) * | 2017-06-27 | 2017-10-24 | 北京粉笔蓝天科技有限公司 | One kind is set a question method and system |
CN109634942A (en) * | 2018-11-16 | 2019-04-16 | 许继集团有限公司 | A kind of energy data exception judgment method and device |
CN110399558B (en) * | 2019-07-26 | 2021-09-14 | 江苏曲速教育科技有限公司 | Test question recommendation method and system |
-
2020
- 2020-09-30 CN CN202011059182.1A patent/CN112131349B/en active Active
-
2021
- 2021-08-16 WO PCT/CN2021/112675 patent/WO2022068435A1/en active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2002101555A2 (en) * | 2001-06-04 | 2002-12-19 | Honeywell International Inc. | Adaptive knowledge management system for vehicle trend monitoring, health management and preventive maintenance |
CN109859553A (en) * | 2018-12-04 | 2019-06-07 | 广东宜教通教育有限公司 | A kind of ability evaluating method doing topic situation based on student |
CN109800880A (en) * | 2018-12-29 | 2019-05-24 | 上海乂学教育科技有限公司 | Based on the adaptive learning Feature Extraction System of dynamic learning style information and application |
CN110377814A (en) * | 2019-05-31 | 2019-10-25 | 平安国际智慧城市科技股份有限公司 | Topic recommended method, device and medium |
CN110309201A (en) * | 2019-06-28 | 2019-10-08 | 广州云蝶科技有限公司 | Operation method for customizing and system |
CN112131349A (en) * | 2020-09-30 | 2020-12-25 | 上海松鼠课堂人工智能科技有限公司 | Method for measuring knowledge point mastering state |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114861916A (en) * | 2022-06-13 | 2022-08-05 | 合肥工业大学 | Knowledge association path fused cognitive tracking method |
CN114861916B (en) * | 2022-06-13 | 2024-03-05 | 合肥工业大学 | Knowledge association path-fused cognitive tracking method |
CN116089705A (en) * | 2022-12-08 | 2023-05-09 | 读书郎教育科技有限公司 | Learning topic personalized recommendation method and system based on AI |
CN116089705B (en) * | 2022-12-08 | 2023-11-14 | 读书郎教育科技有限公司 | Learning topic personalized recommendation method and system based on AI |
CN116383481A (en) * | 2023-02-09 | 2023-07-04 | 四川云数赋智教育科技有限公司 | Personalized test question recommending method and system based on student portrait |
CN116383481B (en) * | 2023-02-09 | 2024-03-29 | 四川云数赋智教育科技有限公司 | Personalized test question recommending method and system based on student portrait |
CN116151242A (en) * | 2023-04-19 | 2023-05-23 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | Intelligent problem recommendation method, system and storage medium for programming learning scene |
CN117114495A (en) * | 2023-09-11 | 2023-11-24 | 湖南软件职业技术大学 | Professional education quality assessment method for capability generation analysis |
CN117114495B (en) * | 2023-09-11 | 2024-01-26 | 湖南软件职业技术大学 | Professional education quality assessment method for capability generation analysis |
CN116933870A (en) * | 2023-09-15 | 2023-10-24 | 浪潮软件股份有限公司 | Method, system and device for arranging elastic operation based on knowledge graph |
Also Published As
Publication number | Publication date |
---|---|
CN112131349A (en) | 2020-12-25 |
CN112131349B (en) | 2021-08-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2022068435A1 (en) | Method for measuring knowledge point mastering state | |
Rousselet et al. | An introduction to the bootstrap: a versatile method to make inferences by using data-driven simulations. | |
Alexopoulos | Introduction to multivariate regression analysis | |
Lykourentzou et al. | Early and dynamic student achievement prediction in e‐learning courses using neural networks | |
CN109740048B (en) | Course recommendation method and device | |
JP7052224B2 (en) | Learning support device, learning support system and program | |
Hong et al. | Weighting methods for assessing policy effects mediated by peer change | |
Man et al. | Response time based nonparametric Kullback-Leibler divergence measure for detecting aberrant test-taking behavior | |
CN112116506B (en) | Method for measuring mastery degree of knowledge points | |
He et al. | Can we trust teaching evaluations when response rates are not high? Implications from a Monte Carlo simulation | |
Wang et al. | Design of an adaptive examination system based on artificial intelligence recognition model | |
Wang et al. | Power considerations for generalized estimating equations analyses of four‐level cluster randomized trials | |
WO2020206172A1 (en) | Confidence evaluation to measure trust in behavioral health survey results | |
Georgiadis et al. | On the robustness of stealth assessment | |
Xiong et al. | Data‐driven Q‐matrix learning based on Boolean matrix factorization in cognitive diagnostic assessment | |
Yi | [Retracted] Research on English Teaching Reading Quality Evaluation Method Based on Cognitive Diagnostic Evaluation | |
JP5437211B2 (en) | E-learning system with problem extraction function considering question frequency and learner's weakness | |
US20210124783A1 (en) | Document similarity through reference links | |
CN111080025A (en) | Learning feature data processing method and device and electronic equipment | |
CN108198474A (en) | Learning data generation method, device and equipment | |
Yakir | Introduction to statistical thinking (with r, without calculus) | |
Smith et al. | Seven reliability indices for high‐stakes decision making: Description, selection, and simple calculation | |
Kristensen et al. | Modeling disengaged guessing behavior in a vocabulary learning app using student, item, and session characteristics | |
Xie et al. | Research on personalized exercises and teaching feedback based on big data | |
US20230087817A1 (en) | System and Methods for Educational and Psychological Modeling and Assessment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21874097 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 12.09.2023) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 21874097 Country of ref document: EP Kind code of ref document: A1 |