CN116342335A - Course recommendation method and device - Google Patents
Course recommendation method and device Download PDFInfo
- Publication number
- CN116342335A CN116342335A CN202310099795.5A CN202310099795A CN116342335A CN 116342335 A CN116342335 A CN 116342335A CN 202310099795 A CN202310099795 A CN 202310099795A CN 116342335 A CN116342335 A CN 116342335A
- Authority
- CN
- China
- Prior art keywords
- target
- vector
- course
- learning
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 239000013598 vector Substances 0.000 claims abstract description 196
- 230000015654 memory Effects 0.000 claims abstract description 36
- 238000004088 simulation Methods 0.000 claims abstract description 14
- 238000013507 mapping Methods 0.000 claims description 24
- 230000007787 long-term memory Effects 0.000 claims description 23
- 230000006403 short-term memory Effects 0.000 claims description 23
- 230000002159 abnormal effect Effects 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000002474 experimental method Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 abstract description 12
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 6
- 230000004913 activation Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000007786 learning performance Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000003442 weekly effect Effects 0.000 description 1
Images
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Primary Health Care (AREA)
- Marketing (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Electrically Operated Instructional Devices (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a course recommendation method and device, wherein the method comprises the following steps: obtaining the scientific achievements and the target learning achievements of a plurality of time points of a user, inputting the scientific achievements and the target learning achievements into each long-short-period memory network, and obtaining indication parameters so as to construct simulation achievements, and recommending relevant courses for the user to learn according to the learning target vectors. The invention has the beneficial effects that: the learning target of the user, namely the target learning score, can be reached as soon as possible by the user according to the recommended course, the learning time of the user can be greatly reduced, and the phenomenon of deviant of the user in the learning process is avoided.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to a course recommendation method and device.
Background
In the prior course recommendation, related courses are generally recommended to users according to a learning plan established by the users or learning targets of the users, however, in the learning process, the user receives different courses and learning ability, and each score can not view the recommended course score to reach the learning targets, so that the learning targets of the users cannot be reached as soon as possible, and therefore, a course recommendation method is needed.
Disclosure of Invention
The invention mainly aims to provide a course recommending method and device, and aims to solve the problem that the existing course recommending method cannot reach the learning target of a user as soon as possible.
The invention provides a course recommendation method, which comprises the following steps:
acquiring various scientific achievements and target learning achievements of a plurality of time points in a user set time period;
vectorizing each scientific achievement in each time point to obtain achievement vectors corresponding to each time point;
sequentially inputting the score vectors into each circulation unit in the long-term and short-term memory network according to the time sequence to obtain a target state; the long-term and short-term memory network comprises a plurality of circulation units, the number of the circulation units is the same as that of the achievement vectors, one circulation unit comprises a forgetting gate, an input gate, newly-added information, an output gate and a unit state, and the unit state is respectively connected with the forgetting gate, the input gate, the newly-added information, the output gate and the unit state of the last circulation unit;
inputting the target state into a full connection layer of the long-short-term memory network to be converted into a designated dimension, so as to obtain a target state vector;
performing sigmoid nonlinear mapping on the target state vector to a (0, 1) interval, and solving a linear distance between the target state vector and the 1 to obtain an indication parameter;
comparing the indication parameter with a set threshold; when the linear distance is higher than a set threshold, judging whether the indication parameter is abnormal or not, wherein the set threshold is obtained according to each scientific achievement and target learning achievement at a first time point in the user set time period;
when the indication parameters are abnormal, adding one or more circulation units to the long-short-period memory network, inputting a simulation score vector to the newly added circulation unit, and continuously adjusting the simulation score vector until the indication parameters are normal, so as to obtain a final adjusted target score vector;
obtaining a latest achievement vector of a time point closest to the current time point in the set time period;
calculating the difference value between the latest target score vector and the target score vector, thereby obtaining a learning target vector;
and recommending relevant courses for the user to learn according to the learning target vector.
Further, the step of inputting each score vector to each circulation unit in the long-term and short-term memory network in sequence according to time sequence to obtain a target state includes:
sequentially inputting the score vectors into a forgetting gate, an input gate, newly-added information and an output gate in each circulation unit in a long-term and short-term memory network according to a time sequence;
according to formula F t ＝sigmoid(W xf |X t |+W hf H t-1 +b f )
I t ＝sigmoid(W xi |X t |+W hi H t-1 +b i )
O t ＝sigmoid(W xo |X t |+W ho H t-1 +b o )
N t ＝tanh(W xn |X t |+W hn H t-1 +b n )
C t ＝F t ⊙C t-1 +I t ⊙N t
Sequentially calculating the unit states of all the circulating units so as to obtain the target state output by the last circulating unit; wherein F is t Represents the t-th forget gate, I t Represents the t-th input gate, N t Represents the t-th newly added information, O t Represents the t-th output gate, C t Represents the t-th cell state, X t Represents the t-th score vector, W xf、 W hf、 b f 、W xi 、W hi 、b i 、W xo 、W ho 、b o 、W xn 、W hn 、b n Are all preset parameters.
Further, the step of recommending relevant courses for the user to learn according to the learning target vector includes:
determining a target course associated with the learning target vector, the target course including at least one base course, each base course including at least one of teaching resources, course content, job content, experiment content, the target course including target courses corresponding to all or part of the base courses;
determining the node number of each target course according to the target course and the learning target vector;
determining a curriculum schedule for the user based on the node numbers of each target curriculum;
and recommending relevant courses for the user to learn according to the curriculum schedule.
Further, after the step of recommending relevant courses for the user to learn according to the learning target vector, the method further includes:
obtaining the scientific achievement after the user learns the relevant course setting time and carrying out vectorization to obtain an actual achievement vector;
adding a circulation unit for the long-short-period memory network, and inputting the target state vector and the actual achievement vector to obtain an actual state vector;
performing sigmoid nonlinear mapping on the actual state vector to a (0, 1) interval, and solving a linear distance between the actual state vector and the 1 to obtain an actual indication parameter;
comparing the actual indication parameter with the set threshold;
if the indication parameters are normal, learning according to a course preset by recommendation; the preset course is a course of a learning plan of the user in a set time period.
Further, before the step of recommending relevant courses for the user to learn according to the curriculum schedule, the method further includes:
acquiring dimension values of each subject according to the latest achievement vector;
obtaining the dimension level of each subject according to the dimension value of each subject;
and acquiring one or more relevant courses corresponding to the dimension levels of the subjects.
The invention also provides a course recommending device, which comprises:
the acquisition module is used for acquiring the scientific achievements and the target learning achievements of a plurality of time points in a user set time period;
the vectorization module is used for vectorizing the scientific achievements in each time point to obtain score vectors corresponding to each time point;
the first input module is used for sequentially inputting each achievement vector into each circulation unit in the long-term and short-term memory network according to a time sequence to obtain a target state; the long-term and short-term memory network comprises a plurality of circulation units, the number of the circulation units is the same as that of the achievement vectors, one circulation unit comprises a forgetting gate, an input gate, newly-added information, an output gate and a unit state, and the unit state is respectively connected with the forgetting gate, the input gate, the newly-added information, the output gate and the unit state of the last circulation unit;
the second input module is used for inputting the target state into the full connection layer of the long-short-term memory network so as to be converted into a designated dimension to obtain a target state vector;
the mapping module is used for performing sigmoid nonlinear mapping on the target state vector to a (0, 1) interval, and solving a linear distance between the target state vector and the 1 to obtain an indication parameter;
the comparison module is used for comparing the indication parameter with a set threshold value; when the linear distance is higher than a set threshold, judging whether the indication parameter is abnormal or not, wherein the set threshold is obtained according to each scientific achievement and target learning achievement at a first time point in the user set time period;
the adding module is used for adding one or more circulation units to the long-short-period memory network when the indication parameters are abnormal, inputting a simulation score vector to the newly added circulation unit, and continuously adjusting the simulation score vector until the indication parameters are normal, so that a final adjusted target score vector is obtained;
the third acquisition module is used for acquiring a latest achievement vector of a time point closest to the current time point in the set time period;
the calculation module is used for calculating the difference value between the latest target score vector and the target score vector so as to obtain a learning target vector;
and the recommending module is used for recommending relevant courses for the user to learn according to the learning target vector.
Further, the first input module includes:
the input sub-module is used for sequentially inputting each score vector into a forgetting gate, an input gate, newly-added information and an output gate in each circulation unit in the long-term and short-term memory network according to a time sequence;
a calculation sub-module for calculating according to formula F t ＝sigmoid(W xf |X t |+W hf H t-1 +b f )
I t ＝sigmoid(W xi |X t |+W hi H t-1 +b i )
O t ＝sigmoid(W xo |X t |+W ho H t-1 +b o )
N t ＝tanh(W xn |X t |+W hn H t-1 +b n )
C t ＝F t ⊙C t-1 +I t ⊙N t
Sequentially calculating the unit states of all the circulating units so as to obtain the target state output by the last circulating unit; wherein F is t Represents the t-th forget gate, I t Represents the t-th input gate, N t Represents the t-th newly added information, O t Represents the t-th output gate, C t Represents the t-th cell state, X t Represents the t-th score vector, W xf 、W hf、 b f、 W xi 、W hi 、b i 、W xo 、W ho 、b o 、W xn 、W hn 、b n Are all preset parameters.
Further, the recommendation module includes:
a first determining submodule, configured to determine a target course associated with the learning target vector, where the target course includes at least one basic lesson, each basic lesson includes at least one of teaching resources, lesson content, job content, and experiment content, and the target course includes a target lesson corresponding to all or part of the basic lessons;
the second determining submodule is used for determining the node number of each target course according to the target course and the learning target vector;
a third determining sub-module for determining a curriculum schedule for the user based on the number of nodes of each target curriculum;
and the recommending sub-module is used for recommending relevant courses for the user to learn according to the course schedule.
Further, the course recommending device further includes:
the vector acquisition module is used for acquiring the scientific achievement after the user learns the related course setting time and vectorizing the scientific achievement to obtain an actual achievement vector;
the unit adding module is used for adding a circulation unit for the long-period and short-period memory network, inputting the target state vector and the actual achievement vector, and obtaining an actual state vector;
the vector mapping module is used for performing sigmoid nonlinear mapping on the actual state vector to a (0, 1) interval, and solving a linear distance between the actual state vector and the 1 to obtain an actual indication parameter;
a parameter comparison module for comparing the actual indication parameter with the set threshold;
the course learning module is used for learning according to recommended preset courses if the indication parameters are normal; the preset course is a course of a learning plan of the user in a set time period.
Further, the recommendation module further includes:
the dimension value acquisition sub-module is used for acquiring the dimension value of each subject according to the latest achievement vector;
the dimension level computing sub-module is used for obtaining the dimension level of each subject according to the dimension value of each subject;
and the related course obtaining sub-module is used for obtaining one or more related courses corresponding to the dimension levels of the subjects respectively.
The invention has the beneficial effects that: the method comprises the steps of obtaining each scientific learning score and target learning score of a plurality of time points of a user, inputting the obtained results into each long-term and short-term memory network, and obtaining indication parameters to construct simulation scores, so that relevant courses are recommended to the user according to the learning target vectors to learn, the user can reach the learning target of the user as soon as possible according to the recommended courses, namely the target learning score, the learning time of the user can be greatly reduced, and the phenomenon of deviational appearance of the user in the learning process is avoided.
Drawings
FIG. 1 is a flow chart of a course recommendation method according to an embodiment of the present invention;
FIG. 2 is a block diagram schematically illustrating a course recommendation apparatus according to an embodiment of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that, in the embodiments of the present invention, all directional indicators (such as up, down, left, right, front, and back) are merely used to explain the relative positional relationship, movement conditions, and the like between the components in a specific posture (as shown in the drawings), if the specific posture is changed, the directional indicators correspondingly change, and the connection may be a direct connection or an indirect connection.
The term "and/or" is herein merely an association relation describing an associated object, meaning that there may be three relations, e.g., a and B, may represent: a exists alone, A and B exist together, and B exists alone.
Furthermore, descriptions such as those referred to as "first," "second," and the like, are provided for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implying an order of magnitude of the indicated technical features in the present disclosure. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In addition, the technical solutions of the embodiments may be combined with each other, but it is necessary to base that the technical solutions can be realized by those skilled in the art, and when the technical solutions are contradictory or cannot be realized, the combination of the technical solutions should be considered to be absent and not within the scope of protection claimed in the present invention.
Referring to fig. 1, the present invention proposes a course recommendation method, including:
s1: acquiring various scientific achievements and target learning achievements of a plurality of time points in a user set time period;
s2: vectorizing each scientific achievement in each time point to obtain achievement vectors corresponding to each time point;
s3: sequentially inputting the score vectors into each circulation unit in the long-term and short-term memory network according to the time sequence to obtain a target state; the long-term and short-term memory network comprises a plurality of circulation units, the number of the circulation units is the same as that of the achievement vectors, one circulation unit comprises a forgetting gate, an input gate, newly-added information, an output gate and a unit state, and the unit state is respectively connected with the forgetting gate, the input gate, the newly-added information, the output gate and the unit state of the last circulation unit;
s4: inputting the target state into a full connection layer of the long-short-term memory network to be converted into a designated dimension, so as to obtain a target state vector;
s5: performing sigmoid nonlinear mapping on the target state vector to a (0, 1) interval, and solving a linear distance between the target state vector and the 1 to obtain an indication parameter;
s6, comparing the indication parameter with a set threshold value; when the linear distance is higher than a set threshold, judging whether the indication parameter is abnormal or not, wherein the set threshold is obtained according to each scientific achievement and target learning achievement at a first time point in the user set time period;
s7: when the indication parameters are abnormal, adding one or more circulation units to the long-short-period memory network, inputting a simulation score vector to the newly added circulation unit, and continuously adjusting the simulation score vector until the indication parameters are normal, so as to obtain a final adjusted target score vector;
s8, obtaining a latest score vector of a time point closest to the current time point in the set time period;
s9: calculating the difference value between the latest target score vector and the target score vector, thereby obtaining a learning target vector;
s10: and recommending relevant courses for the user to learn according to the learning target vector.
The invention is applicable to any viable user, for example to students in primary and middle school students, college students, other types of schools.
As described in step S1, the respective learning results and the target learning results at a plurality of time points in the user-set time period are obtained. The method of obtaining may be that the user directly inputs or obtains from a database or a table in which the scientific achievements and the target learning achievements are stored, and specifically, the multiple time points may be weekly or monthly test achievements.
As described in the above step S2, vectorizing each of the learning achievements in each time point to obtain a achievement vector corresponding to each time point, where the vectorizing may be performed by ordering each of the learning achievements according to a preset discipline sequence, and the learning achievements of each discipline are one dimension of the vector, so as to obtain the achievement vector corresponding to each time point.
And (3) inputting the score vectors into the circulation units in the long-term and short-term memory network in sequence according to the time sequence to obtain the target state as described in the steps S3-S6. The long-period memory network mainly comprises a plurality of circulating units, a full-connection layer and a softmax layer, the circulating units are sequentially connected, the last circulating unit is connected with the full-connection layer, the full-connection layer is connected with the softmax layer, the circulating units mainly comprise forgetting doors, input doors, newly-added information, output doors and unit states, the unit states are respectively connected with the forgetting doors, the input doors, the newly-added information, the output doors and the unit states of the last circulating unit, wherein the activation function of the newly-added information layer is a tanh function, the activation functions of the output doors, the input doors and the forgetting doors are all sigmoid functions, and deviation conditions of a user to each scientific result in a learning process can be effectively counted through the long-period memory network, so that whether each scientific result of the user progresses simultaneously according to a target learning result in the learning process can be judged. Specifically, each achievement vector is sequentially input into each circulation unit in the long-short-period memory network according to a time sequence to obtain a target state, namely a value output by the unit state of the last circulation unit, then the target state is input into a full-connection layer of the long-short-period memory network to be converted into a specified dimension to obtain a target state vector, wherein the main function of the full-connection layer is to convert the input into the specified dimension, then obtain the target state vector, at the moment, the target state vector is subjected to sigmoid nonlinear mapping to a (0, 1) interval, the linear distance between the target state vector and the target state vector is obtained, the indication parameter is obtained, then the mapping value of each first similarity value in the (0, 1) interval is calculated through a sigmoid function, and finally the linear distance between the target state and the target state vector is obtained according to the mapping value in a mode of subtracting the mapping value from 1.
As described in step S7, when the indication parameter is abnormal, one or more circulation units are added to the long-short-term memory network, and a simulated score vector is input to the newly added circulation unit, and is continuously adjusted until the indication parameter is normal, so as to obtain a final adjusted target score vector. The simulation vector is continuously adjusted until the indication parameters are normal, namely the next or next few achievements of the user need to meet certain requirements, the achievements of the user can be forced to progress simultaneously, and the user can continuously adjust in the learning process.
As described in the above steps S8-S10, the latest achievement vector of the closest point in time from the current set time period is obtained. The latest target score vector can reflect the learning condition of the user, so that the difference value between the latest target score vector and the target score vector can be calculated, the learning target vector is obtained, and finally, related courses are recommended for the user to learn according to the learning target vector. Therefore, the user can reach the learning target of the user as soon as possible according to the recommended course, namely the target learning score, the learning time of the user can be greatly reduced, and the phenomenon of deviant of the user in the learning process is avoided.
In one embodiment, the step S3 of inputting the score vectors into the circulation units in the long-term and short-term memory network sequentially according to time sequence to obtain the target state includes:
s301: sequentially inputting the score vectors into a forgetting gate, an input gate, newly-added information and an output gate in each circulation unit in a long-term and short-term memory network according to a time sequence;
s302: according to formula F t ＝sigmoid(W xf |X t |+W hf H t-1 +b f )
I t ＝sigmoid(W xi |X t |+W hi H t-1 +b i )
O t ＝sigmoid(W xo |X t |+W ho H t-1 +b o )
N t ＝tanh(W xn |X t |+W hn H t-1 +b n )
C t ＝F t ⊙C t-1 +I t ⊙N t
Sequentially calculating the unit states of all the circulating units so as to obtain the target state output by the last circulating unit; wherein F is t Represents the t-th forget gate, I t Represents the t-th input gate, N t Represents the t-th newly added information, O t Represents the t-th output gate, C t Represents the t-th cell state, X t Represents the t-th score vector, W xf 、W hf 、b f 、W xi 、W hi 、b i 、W xo 、W ho 、b o 、W xn 、W hn 、b n Are all preset parameters.
As described in the above steps S301-S302, calculation of the score vector by each gate in the loop unit is achieved. Specifically, because the calculation of the current unit state combines the unit state of the previous time step and the newly added information, the weights of the current unit state and the newly added information are respectively determined by the input gate and the forgetting gate, and the current unit state can acquire valuable information in an earlier time step through the adjustment of the input gate and the forgetting gate, namely, the deflection problem of a user to each department in the learning process is acquired.
In one embodiment, the step S10 of recommending relevant courses for the user according to the learning objective vector includes:
s1001: determining a target course associated with the learning target vector, the target course including at least one base course, each base course including at least one of teaching resources, course content, job content, experiment content, the target course including target courses corresponding to all or part of the base courses;
s1002: determining the node number of each target course according to the target course and the learning target vector;
s1003: determining a curriculum schedule for the user based on the node numbers of each target curriculum;
s1004: and recommending relevant courses for the user to learn according to the curriculum schedule.
The acquisition of relevant courses is achieved as described in steps S1001-S1004 above. Specifically, a course may refer to: the electronic teaching content consists of a group of lessons, and is used for learning by a learner in a period of time. In one possible implementation, the target course may include any one of a building block programming course, a Python programming course, an artificial intelligence AI programming course, and may also be a high-medium period language number, a Wen Zengli ensemble, etc., and the disclosure is not limited to the specific type of the target course. The target course may include at least one base lesson, each base lesson including at least one of experimental content, lesson content, teaching resources, and job content. Determining the node number of each target course according to the target course and the learning target vector; determining a curriculum schedule for the user based on the node numbers of each target curriculum; and recommending relevant courses for the user to learn according to the curriculum schedule. The relevant course is a relevant course corresponding to the target course, such as a teacher's teaching video.
In one embodiment, after the step S10 of recommending relevant courses for the user according to the learning objective vector, the method further includes:
s1101: obtaining the scientific achievement after the user learns the relevant course setting time and carrying out vectorization to obtain an actual achievement vector;
s1102: adding a circulation unit for the long-short-period memory network, and inputting the target state vector and the actual achievement vector to obtain an actual state vector;
s1103: performing sigmoid nonlinear mapping on the actual state vector to a (0, 1) interval, and solving a linear distance between the actual state vector and the 1 to obtain an actual indication parameter;
s1104: comparing the actual indication parameter with the set threshold;
s1105: if the indication parameters are normal, learning according to a course preset by recommendation; the preset course is a course of a learning plan of the user in a set time period.
As described in the above steps S1101-S1105, the setting of the learning plan of the user is achieved, the user has a learning plan at first, that is, a learning plan, that is, a preset course, is preset for the user according to the performance of the user and the target learning performance of the user, and after the user learns for a period of time, the state of the user has returned to the condition that the performance of each department is advanced synchronously, that is, the progress is the same, at this time, the initial learning plan can be followed, specifically, each learning performance of the user after the setting time of the related course is learned and vectorized, so as to obtain an actual performance vector; and adding a circulation unit for the long-short-period memory network, inputting the target state vector and the actual achievement vector to obtain an actual state vector, performing sigmoid nonlinear mapping on the actual state vector to a (0, 1) interval, and obtaining a linear distance from the actual state vector to 1 to obtain an actual indication parameter. The mapping mode is the same as the above mode, and is not repeated here, if the indication parameter is normal, learning is performed according to a course preset by recommendation; the preset course is a course of a learning plan of the user in a set time period.
In one embodiment, before the step S1004 of recommending relevant courses for the user according to the curriculum schedule for learning, the method further includes:
s10031: acquiring dimension values of each subject according to the latest achievement vector;
s10032: obtaining the dimension level of each subject according to the dimension value of each subject;
s10033: and acquiring one or more relevant courses corresponding to the dimension levels of the subjects.
The acquisition of relevant courses is accomplished as described in steps S10031-S10033 above. Because the basis of each person is different, the corresponding courses are different, and therefore, the dimension values of each subject, namely the scientific achievements, can be obtained according to the latest achievement vector, the dimension level of each subject is obtained according to the dimension values of each subject, and one or more relevant courses corresponding to the dimension levels of each subject respectively are obtained. Wherein, each dimension level and the related course have pre-established association relation, and the corresponding related course can be directly obtained according to the dimension level.
Referring to fig. 2, the present invention further provides a course recommendation device, including:
an acquisition module 10, configured to acquire each of the learning results and the target learning results at a plurality of time points within a user-set time period;
the vectorization module 20 is configured to vectorize each of the training achievements at each time point, so as to obtain a achievement vector corresponding to each time point;
the first input module 30 is configured to sequentially input each score vector into each circulation unit in the long-term and short-term memory network according to a time sequence, so as to obtain a target state; the long-term and short-term memory network comprises a plurality of circulation units, the number of the circulation units is the same as that of the achievement vectors, one circulation unit comprises a forgetting gate, an input gate, newly-added information, an output gate and a unit state, and the unit state is respectively connected with the forgetting gate, the input gate, the newly-added information, the output gate and the unit state of the last circulation unit;
the second input module 40 is configured to input the target state into a full connection layer of the long-short-term memory network, so as to convert the target state into a specified dimension, thereby obtaining a target state vector;
the mapping module 50 is configured to perform sigmoid nonlinear mapping on the target state vector to a (0, 1) interval, and calculate a linear distance from 1 to obtain an indication parameter;
a comparison module 60 for comparing the indication parameter with a set threshold; when the linear distance is higher than a set threshold, judging whether the indication parameter is abnormal or not, wherein the set threshold is obtained according to each scientific achievement and target learning achievement at a first time point in the user set time period;
an adding module 70, configured to add one or more circulation units to the long-short-term memory network when the indication parameter is abnormal, and input a simulated score vector to the newly added circulation unit, and continuously adjust the simulated score vector until the indication parameter is normal, so as to obtain a final adjusted target score vector;
a third obtaining module 80, configured to obtain a latest achievement vector of a point in time closest to a current point in time within the set time period;
a calculating module 90, configured to calculate a difference between the latest target score vector and the target score vector, thereby obtaining a learning target vector;
and the recommending module 100 is used for recommending relevant courses for the user to learn according to the learning target vector.
In one embodiment, the first input module 30 includes:
the input sub-module is used for sequentially inputting each score vector into a forgetting gate, an input gate, newly-added information and an output gate in each circulation unit in the long-term and short-term memory network according to a time sequence;
a calculation sub-module for calculating according to formula F t ＝sigmoid(W xf |X t |+W hf H t-1 +b f )
I t ＝sigmoid(W xi |X t |+W hi H t-1 +b i )
O t ＝sigmoid(W xo |X t |+W ho H t-1 +b o )
N t ＝tanh(W xn |X t |+W hn H t-1 +b n )
C t ＝F t ⊙C t-1 +I t ⊙N t
Sequentially calculating the unit states of all the circulating units so as to obtain the target state output by the last circulating unit; wherein F is t Represents the t-th forget gate, I t Represents the t-th input gate, N t Represents the t-th newly added information, O t Represents the t-th output gate, C t Represents the t-th cell state, X t Represents the t-th score vector, W xf、 W hf、 b f 、W xi 、W hi 、b i 、W xo 、W ho 、b o 、W xn 、W hn 、b n Are all preset parameters.
In one embodiment, the recommendation module 100 includes:
a first determining submodule, configured to determine a target course associated with the learning target vector, where the target course includes at least one basic lesson, each basic lesson includes at least one of teaching resources, lesson content, job content, and experiment content, and the target course includes a target lesson corresponding to all or part of the basic lessons;
the second determining submodule is used for determining the node number of each target course according to the target course and the learning target vector;
a third determining sub-module for determining a curriculum schedule for the user based on the number of nodes of each target curriculum;
and the recommending sub-module is used for recommending relevant courses for the user to learn according to the course schedule.
In one embodiment, the course recommending device further includes:
the vector acquisition module is used for acquiring the scientific achievement after the user learns the related course setting time and vectorizing the scientific achievement to obtain an actual achievement vector;
the unit adding module is used for adding a circulation unit for the long-period and short-period memory network, inputting the target state vector and the actual achievement vector, and obtaining an actual state vector;
the vector mapping module is used for performing sigmoid nonlinear mapping on the actual state vector to a (0, 1) interval, and solving a linear distance between the actual state vector and the 1 to obtain an actual indication parameter;
a parameter comparison module for comparing the actual indication parameter with the set threshold;
the course learning module is used for learning according to recommended preset courses if the indication parameters are normal; the preset course is a course of a learning plan of the user in a set time period.
In one embodiment, the recommendation module 100 further includes:
the dimension value acquisition sub-module is used for acquiring the dimension value of each subject according to the latest achievement vector;
the dimension level computing sub-module is used for obtaining the dimension level of each subject according to the dimension value of each subject;
and the related course obtaining sub-module is used for obtaining one or more related courses corresponding to the dimension levels of the subjects respectively.
The invention has the beneficial effects that: the method comprises the steps of obtaining each scientific learning score and target learning score of a plurality of time points of a user, inputting the obtained results into each long-term and short-term memory network, and obtaining indication parameters to construct simulation scores, so that relevant courses are recommended to the user according to the learning target vectors to learn, the user can reach the learning target of the user as soon as possible according to the recommended courses, namely the target learning score, the learning time of the user can be greatly reduced, and the phenomenon of deviational appearance of the user in the learning process is avoided.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by hardware associated with a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium provided herein and used in embodiments may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, apparatus, article or method that comprises the element.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (10)
1. A method of course recommendation, comprising:
acquiring various scientific achievements and target learning achievements of a plurality of time points in a user set time period;
vectorizing each scientific achievement in each time point to obtain achievement vectors corresponding to each time point;
sequentially inputting the score vectors into each circulation unit in the long-term and short-term memory network according to the time sequence to obtain a target state; the long-term and short-term memory network comprises a plurality of circulation units, the number of the circulation units is the same as that of the achievement vectors, one circulation unit comprises a forgetting gate, an input gate, newly-added information, an output gate and a unit state, and the unit state is respectively connected with the forgetting gate, the input gate, the newly-added information, the output gate and the unit state of the last circulation unit;
inputting the target state into a full connection layer of the long-short-term memory network to be converted into a designated dimension, so as to obtain a target state vector;
performing sigmoid nonlinear mapping on the target state vector to a (0, 1) interval, and solving a linear distance between the target state vector and the 1 to obtain an indication parameter;
comparing the indication parameter with a set threshold; when the linear distance is higher than a set threshold, judging whether the indication parameter is abnormal or not, wherein the set threshold is obtained according to each scientific achievement and target learning achievement at a first time point in the user set time period;
when the indication parameters are abnormal, adding one or more circulation units to the long-short-period memory network, inputting a simulation score vector to the newly added circulation unit, and continuously adjusting the simulation score vector until the indication parameters are normal, so as to obtain a final adjusted target score vector;
obtaining a latest achievement vector of a time point closest to the current time point in the set time period;
calculating the difference value between the latest target score vector and the target score vector, thereby obtaining a learning target vector;
and recommending relevant courses for the user to learn according to the learning target vector.
2. The course recommendation method of claim 1, wherein the step of sequentially inputting each of the score vectors into each of the circulation units in the long-short-term memory network in time sequence to obtain a target state comprises:
sequentially inputting the score vectors into a forgetting gate, an input gate, newly-added information and an output gate in each circulation unit in a long-term and short-term memory network according to a time sequence;
according to formula F t ＝sigmoid(W xf |X t |+W hf H t-1 +b f )
I t ＝sigmoid(W xi |X t |+W hi H t-1 +b i )
O t ＝sigmoid(W xo |X t |+W ho H t-1 +b o )
N t ＝tanh(W xn |X t |+W hn H t-1 +b n )
C t ＝F t ⊙C t-1 +I t ⊙N t
Sequentially calculating the unit states of all the circulating units so as to obtain the target state output by the last circulating unit; wherein F is t Represents the t-th forget gate, I t Represents the t-th input gate, N t Represents the t-th newly added information, O t Represents the t-th output gate, C t Represents the t-th cell state, X t Represents the t-th score vector, W xf 、W hf 、b f 、W xi 、W hi 、b i 、W xo 、W ho 、b o 、W xn 、W hn 、b n Are all preset parameters.
3. The course recommendation method of claim 1, wherein the step of recommending relevant courses for the user based on the learning objective vector comprises:
determining a target course associated with the learning target vector, the target course including at least one base course, each base course including at least one of teaching resources, course content, job content, experiment content, the target course including target courses corresponding to all or part of the base courses;
determining the node number of each target course according to the target course and the learning target vector;
determining a curriculum schedule for the user based on the node numbers of each target curriculum;
and recommending relevant courses for the user to learn according to the curriculum schedule.
4. The course recommendation method of claim 1, further comprising, after the step of recommending relevant courses for the user based on the learning objective vector:
obtaining the scientific achievement after the user learns the relevant course setting time and carrying out vectorization to obtain an actual achievement vector;
adding a circulation unit for the long-short-period memory network, and inputting the target state vector and the actual achievement vector to obtain an actual state vector;
performing sigmoid nonlinear mapping on the actual state vector to a (0, 1) interval, and solving a linear distance between the actual state vector and the 1 to obtain an actual indication parameter;
comparing the actual indication parameter with the set threshold;
if the indication parameters are normal, learning according to a course preset by recommendation; the preset course is a course of a learning plan of the user in a set time period.
5. The method for recommending courses for use in accordance with claim 3, wherein prior to the step of recommending relevant courses for the user based on the curriculum schedule, further comprising:
acquiring dimension values of each subject according to the latest achievement vector;
obtaining the dimension level of each subject according to the dimension value of each subject;
and acquiring one or more relevant courses corresponding to the dimension levels of the subjects.
6. A course recommendation device, comprising:
the acquisition module is used for acquiring the scientific achievements and the target learning achievements of a plurality of time points in a user set time period;
the vectorization module is used for vectorizing the scientific achievements in each time point to obtain score vectors corresponding to each time point;
the first input module is used for sequentially inputting each achievement vector into each circulation unit in the long-term and short-term memory network according to a time sequence to obtain a target state; the long-term and short-term memory network comprises a plurality of circulation units, the number of the circulation units is the same as that of the achievement vectors, one circulation unit comprises a forgetting gate, an input gate, newly-added information, an output gate and a unit state, and the unit state is respectively connected with the forgetting gate, the input gate, the newly-added information, the output gate and the unit state of the last circulation unit;
the second input module is used for inputting the target state into the full connection layer of the long-short-term memory network so as to be converted into a designated dimension to obtain a target state vector;
the mapping module is used for performing sigmoid nonlinear mapping on the target state vector to a (0, 1) interval, and solving a linear distance between the target state vector and the 1 to obtain an indication parameter;
the comparison module is used for comparing the indication parameter with a set threshold value; when the linear distance is higher than a set threshold, judging whether the indication parameter is abnormal or not, wherein the set threshold is obtained according to each scientific achievement and target learning achievement at a first time point in the user set time period;
the adding module is used for adding one or more circulation units to the long-short-period memory network when the indication parameters are abnormal, inputting a simulation score vector to the newly added circulation unit, and continuously adjusting the simulation score vector until the indication parameters are normal, so that a final adjusted target score vector is obtained;
the third acquisition module is used for acquiring a latest achievement vector of a time point closest to the current time point in the set time period;
the calculation module is used for calculating the difference value between the latest target score vector and the target score vector so as to obtain a learning target vector;
and the recommending module is used for recommending relevant courses for the user to learn according to the learning target vector.
7. The curriculum recommendation device of claim 6, wherein said first input module includes:
the input sub-module is used for sequentially inputting each score vector into a forgetting gate, an input gate, newly-added information and an output gate in each circulation unit in the long-term and short-term memory network according to a time sequence;
a calculation sub-module for calculating according to formula F t ＝sigmoid(W xf |X t |+W hf H t-1 +b f )
I t ＝sigmoid(W xi |X t |+W hi H t-1 +b i )
O t ＝sigmoid(W xo |X t |+W ho H t-1 +b o )
N t ＝tanh(W xn |X t |+W hn H t-1 +b n )
C t ＝F t ⊙C t-1 +I t ⊙N t
Sequentially calculating the unit states of all the circulating units so as to obtain the target state output by the last circulating unit; wherein F is t Represents the t-th forget gate, I t Represents the t-th input gate, N t Represents the t-th newly added information, O t Represents the t-th output gate, C t Represents the t-th cell state, X t Represents the t-th score vector, W xf 、W hf 、b f 、W xi 、W hi 、b i 、W xo 、W ho 、b o 、W xn 、W hn 、b n Are all preset parameters.
8. The course recommending means as defined in claim 6, wherein said recommending means comprises:
a first determining submodule, configured to determine a target course associated with the learning target vector, where the target course includes at least one basic lesson, each basic lesson includes at least one of teaching resources, lesson content, job content, and experiment content, and the target course includes a target lesson corresponding to all or part of the basic lessons;
the second determining submodule is used for determining the node number of each target course according to the target course and the learning target vector;
a third determining sub-module for determining a curriculum schedule for the user based on the number of nodes of each target curriculum;
and the recommending sub-module is used for recommending relevant courses for the user to learn according to the course schedule.
9. The course recommendation device of claim 6, further comprising:
the vector acquisition module is used for acquiring the scientific achievement after the user learns the related course setting time and vectorizing the scientific achievement to obtain an actual achievement vector;
the unit adding module is used for adding a circulation unit for the long-period and short-period memory network, inputting the target state vector and the actual achievement vector, and obtaining an actual state vector;
the vector mapping module is used for performing sigmoid nonlinear mapping on the actual state vector to a (0, 1) interval, and solving a linear distance between the actual state vector and the 1 to obtain an actual indication parameter;
a parameter comparison module for comparing the actual indication parameter with the set threshold;
the course learning module is used for learning according to recommended preset courses if the indication parameters are normal; the preset course is a course of a learning plan of the user in a set time period.
10. The curriculum recommendation device of claim 8, wherein said recommendation module further comprises:
the dimension value acquisition sub-module is used for acquiring the dimension value of each subject according to the latest achievement vector;
the dimension level computing sub-module is used for obtaining the dimension level of each subject according to the dimension value of each subject;
and the related course obtaining sub-module is used for obtaining one or more related courses corresponding to the dimension levels of the subjects respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310099795.5A CN116342335B (en) | 2023-02-03 | 2023-02-03 | Course recommendation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310099795.5A CN116342335B (en) | 2023-02-03 | 2023-02-03 | Course recommendation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116342335A true CN116342335A (en) | 2023-06-27 |
| CN116342335B CN116342335B (en) | 2023-11-14 |
Family
ID=86882942
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310099795.5A Active CN116342335B (en) | 2023-02-03 | 2023-02-03 | Course recommendation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116342335B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106875309A (en) * | 2017-04-01 | 2017-06-20 | 福建云课堂教育科技有限公司 | A kind of course recommends method and system |
| CN109146174A (en) * | 2018-08-21 | 2019-01-04 | 广东恒电信息科技股份有限公司 | A kind of elective course accurate recommendation method based on result prediction |
| CN111274411A (en) * | 2020-01-22 | 2020-06-12 | 文思海辉智科科技有限公司 | Course recommendation method and device, electronic equipment and readable storage medium |
| CN111460249A (en) * | 2020-02-24 | 2020-07-28 | 桂林电子科技大学 | Personalized learning resource recommendation method based on learner preference modeling |
| CN111581529A (en) * | 2020-05-07 | 2020-08-25 | 之江实验室 | Course recommendation method and device combining student fitness and course collocation degree |
| KR20210020718A (en) * | 2019-08-14 | 2021-02-24 | (주)강안교육 | System for calculating study achievement and determining course class |
| CN112632385A (en) * | 2020-12-29 | 2021-04-09 | 中国平安人寿保险股份有限公司 | Course recommendation method and device, computer equipment and medium |
| CN113065060A (en) * | 2021-02-18 | 2021-07-02 | 山东师范大学 | Deep learning-based education platform course recommendation method and system |
| CN114077704A (en) * | 2020-08-19 | 2022-02-22 | Tcl科技集团股份有限公司 | Learning content recommendation method and device, intelligent terminal and storage medium |
| CN114418807A (en) * | 2022-03-30 | 2022-04-29 | 北京英华在线科技有限公司 | Course recommendation method and system of online education platform based on historical scores |
| KR102398317B1 (en) * | 2021-07-01 | 2022-05-16 | (주)뤼이드 | Method for, device for, and system for recommending education contents for fairness of the education |
| CN115146162A (en) * | 2022-06-30 | 2022-10-04 | 武汉美和易思数字科技有限公司 | Online course recommendation method and system |
-
2023
- 2023-02-03 CN CN202310099795.5A patent/CN116342335B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106875309A (en) * | 2017-04-01 | 2017-06-20 | 福建云课堂教育科技有限公司 | A kind of course recommends method and system |
| CN109146174A (en) * | 2018-08-21 | 2019-01-04 | 广东恒电信息科技股份有限公司 | A kind of elective course accurate recommendation method based on result prediction |
| KR20210020718A (en) * | 2019-08-14 | 2021-02-24 | (주)강안교육 | System for calculating study achievement and determining course class |
| CN111274411A (en) * | 2020-01-22 | 2020-06-12 | 文思海辉智科科技有限公司 | Course recommendation method and device, electronic equipment and readable storage medium |
| CN111460249A (en) * | 2020-02-24 | 2020-07-28 | 桂林电子科技大学 | Personalized learning resource recommendation method based on learner preference modeling |
| CN111581529A (en) * | 2020-05-07 | 2020-08-25 | 之江实验室 | Course recommendation method and device combining student fitness and course collocation degree |
| CN114077704A (en) * | 2020-08-19 | 2022-02-22 | Tcl科技集团股份有限公司 | Learning content recommendation method and device, intelligent terminal and storage medium |
| CN112632385A (en) * | 2020-12-29 | 2021-04-09 | 中国平安人寿保险股份有限公司 | Course recommendation method and device, computer equipment and medium |
| CN113065060A (en) * | 2021-02-18 | 2021-07-02 | 山东师范大学 | Deep learning-based education platform course recommendation method and system |
| KR102398317B1 (en) * | 2021-07-01 | 2022-05-16 | (주)뤼이드 | Method for, device for, and system for recommending education contents for fairness of the education |
| CN114418807A (en) * | 2022-03-30 | 2022-04-29 | 北京英华在线科技有限公司 | Course recommendation method and system of online education platform based on historical scores |
| CN115146162A (en) * | 2022-06-30 | 2022-10-04 | 武汉美和易思数字科技有限公司 | Online course recommendation method and system |
Non-Patent Citations (5)
| Title |
|---|
| 上海市软件行业协会组: "深度学习与小标注数据训练", 《2019年软件工程论文专集》, pages 77 * |
| 于冰;袁贝贝;庄可馨;潜静;倪晓燕;章美仁;: "应用卷积神经网络的学习推荐系统", 福建电脑, no. 04 * |
| 查鲁C阿加沃尔: "长短期记忆网络", 《神经网络与深度学习》, pages 226 - 229 * |
| 焦李成等: "长短期记忆网络", 《现代机器学习 第1版》, pages 157 - 161 * |
| 王涛;翟雪松;: "基于神经网络的智能课程推荐模型", 合肥学院学报, no. 01 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116342335B (en) | 2023-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111582694B (en) | Learning evaluation method and device | |
| CN112085168A (en) | Knowledge tracking method and system based on dynamic key value gating circulation network | |
| CN113361685B (en) | Knowledge tracking method and system based on learner knowledge state evolution expression | |
| WO2015009286A1 (en) | System and method for learning recommendation simulation | |
| CN114429212A (en) | Intelligent learning knowledge ability tracking method, electronic device and storage medium | |
| CN116342335B (en) | Course recommendation method and device | |
| CN113591988B (en) | Knowledge cognitive structure analysis method, system, computer equipment, medium and terminal | |
| Caro et al. | Student academic performance stochastic simulator based on the Monte Carlo method | |
| Diao et al. | Personalized exercise recommendation driven by learning objective within e-learning systems | |
| CN114997461B (en) | Time-sensitive answer correctness prediction method combining learning and forgetting | |
| Berzina et al. | Promoting of lifelong learning in engineering | |
| CN114971066A (en) | Knowledge tracking method and system integrating forgetting factor and learning ability | |
| CN114973815A (en) | System and method for testing students in online teaching | |
| CN113590961A (en) | Personalized exercise recommendation method and device based on cognition and state evaluation and intelligent terminal | |
| Teodorescu et al. | Continuous Student Knowledge Tracing Using SVD and Concept Maps | |
| Shah et al. | Spaced Repetition for Slow Learners | |
| Salimovna et al. | Improving the structure and modelinteraction of information processes on distance learning | |
| Abbasov et al. | Informational modeling of the behavior of a teacher in the learning process based on fuzzy logic | |
| Wahyono et al. | A novel intelligent learning for teaching using artificial neural network | |
| JP7417061B2 (en) | memorization system | |
| Almond et al. | Vertical and stationary scales for progress maps | |
| Marković et al. | Applications of Artificial Neural Networks in Multi-Criteria Decision-Making in Civil Engineering | |
| Bogomolova | Essays in Heterogeneous Learning | |
| Chen et al. | A study of marking utility models based on principal component analysis and integer programming | |
| Islam et al. | Reinforcement learning based interactive agent for personalized mathematical skill enhancement |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |